ซ>EPA
          United States
          Environmental Protection
          Agency
            Research And Development  EPA-600-4-91-002
            (MD-591)         July 1994
Short-Term Methods For
Estimating The Chronic Toxicity
Of Effluents And Receiving Water
To Freshwater Organisms

Third Edition

-------
                                  DISCLAIMER                           :

   This document has been reviewed by the Environmental Monitoring Systems
Laboratory-Cincinnati (EMSL-Cincinnati), U.S. Environmental-Protection Agency
(USEPA) and approved for publication.  The mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
The results of data analyses by computer programs described in the section on
data analysis were verified using data.commonly obtained from effluent
toxicity tests.  However, these computer programs may not be applicable to all
data, and the USEPA assumes no responsibility for their use.

-------
                                   FOREWORD


   Environmental measurements! are required to determine the quality of ambient
waters and the character of waste effluents.  The Environmental Monitoring
Systems Laboratory-Cincinnati (EMSL-Cincinnati) conducts research to:

   •  Develop and evaluate analytical methods to identify :and measure the
      concentration of chemical  pollutants in drinking waters, surface waters,
      groundwaters, wastewaters, sediments, sludges, and solid wastes.

   •  Investigate methods for the identification and measurement of viruses,
      bacteria and other microbiological organisms in aqueous samples and to
      determine the response of aquatic organisms to water quality.

   •  Develop and operate a quality assurance program to support the
      achievement of data quality objectives in measurements of pollutants in
      drinking water, surface water, groundwater, wastewater, sediment and
      solid waste.                                        .

   •  Develop methods and models to detect and quantify responses in aquatic
      and terrestrial organisms exposed to environmental stressors and to
      correlate the exposure with effects on chemical and.biological
      indicators.
   The Federal Water Pollution Control Act Amendments of 1972 (PL 92-500), the
Clean Water Act (CWA) of 1977 (PL 95-217), and the Water Quality Act of 1987
(PL 100-4) explicitly state that it is the national policy that the discharge
of toxic substances in toxic amounts be prohibited.  The detection of
chronically toxic effects, therefore, plays an important role in identifying
and controlling toxic discharges to surface waters.  This manual is a third
edition of the freshwater chronic toxicity test manual for effluents first
published (EPA/600/4-85/014) by EMSL-Cincinnati in December 1985 and revised
(EPA/600/4-89/001) in March, 1989.  It provides updated methods for estimating
the chronic toxicity of effluents and receiving waters to freshwater organisms
for use by the U.S. Environmental Protection Agency (USEPA).regional and state
programs, and National Pollutant Discharge Elimination System (NPDES)
permittees.                                              '
                                          Thomas A. Clark,, Director
                                          Environmental Monitoring Systems
                                          Laboratory-Cincinnati

-------
                                    PREFACE


   This manual represents the third edition of the Agency's methods manual for
estimating the chronic toxicity of effluents and receiving waters to  :
freshwater organisms initially published by USEPA's Office of Research^ and
Development, Environmental Monitoring  and Support Laboratory (EMSL-
Cincinnati) in December 1985.  This edition reflects changes recommended by
the Toxicity Assessment Subcommittee of the Biological Advisory Committee,
USEPA headquarters, program offices, and regional staff, other Federal
agencies, state and interstate water pollution control programs, environmental
protection groups, trade associations, major industries, consulting firms,
academic institutions engaged in aquatic toxicology research, and other
interested parties in the private sector.

   The membership of the Toxicity Assessment Subcommittee, USEPA's Biological
Advisory Committee is as follows:

   William Peltier, Subcommittee Chairman,                            ;
       Environmental Services Division, Region 4             .         ;
   Peter Nolan, Environmental Services Division, Region 1
   Jim Ferretti, Environmental Services Division, Region 2            j
   Ronald Preston, Environmental Services Division, Region 3          ;
   Charles Steiner, Environmental Services Division, Region 5         '
   Evan Hornig, Environmental Services Division, Region 6
   Terry Hollister, Environmental Services Division, Region 6
   Michael Tucker, Environmental Services Division, Region 7
   Loys Parrish, Environmental Services Division, Region 8            ;
   Peter Husby, Environmental Services Division, Region 9             ;
   Joseph Cummins, Environmental Services Division, Region 10         .
   Gretchen Hayslip, Water Monitoring and Analysis Section, Region 10 !
   Bruce Binkley, National Enforcement Investigations Center, Denver  ;
   Wesley Kinney, Environmental Monitoring Systems Laboratory-Las Vega|s
   James Lazorchak, Environmental Monitoring Systems Laboratory-Cincinnati
   George Morrison, Environmental Research Laboratory-Narragansett
   Douglas Middaugh, Environmental Research Laboratory-Gulf Breeze    ;
   Teresa Norberg-King, Environmental Research Laboratory-Duluth      i
   Donald Klemm, Environmental Monitoring Systems Laboratory-Cincinnati
   Philip Lewis, Environmental Monitoring Systems Laboratory-Cincinnati
   Cornelius I. Weber, Environmental Monitoring Systems Laboratory-
      Cincinnati       ,                                               !
   Richard Swartz, Environmental Research Laboratory-Newport          \
   Margarete Heber, Human Health and Ecological Criteria Division,
      Office of Science and Technology (OST), Office of Water (OW)
   Bruce Newton, Assessment and Watershed Protection Division, Office bf
      Wetlands, Oceans, and Watersheds, OW
   Christopher Zarba, Human Health and Ecological Criteria Division, Office
      of Science and Technology, OW                                   ;
   Daniel Rieder, Hazard Evaluation Division, Office of Pesticides Programs
   Jerry Smrchek, Health and Environmental Review Division, Office of ;
      Toxic Substances                                                I

-------
Gail Hansen, Office of Solid Waste
Royal Nadeau, Emergency Response Team, Edison
                           Teresa J. Norberg-King
                           Chairman, Biological Advisory Committee
                           Regulatory Ecotoxicology Branch, ERL-Duluth

-------
                                   ABSTRACT                            j
                                                                       i

   This manual describes four short-term (four- to seven-day) methods for
estimating the chronic toxicity of effluents and receiving waters to three
freshwater species:  the fathead minnow, Pimephales promelas, a daphnid,
Ceriodaphnia dubia, and a green alga, Selenastrum capricornutum.  The methods
include single and multiple concentration static renewal and non-renewal
toxicity tests for effluents and receiving waters.  Also included are
guidelines on laboratory safety, quality assurance, facilities, equipment and
supplies; dilution water; effluent and receiving water sample collection,
preservation, shipping, and holding; test conditions; toxicity test data
analysis; report preparation; and organism culturing, holding, and handling.
Examples of computer input and output for Dunnett's Procedure, Probit
Analysis, Trimmed Spearman-Karber Method and the Linear Interpolation Method
are provided in the Appendices.

-------
                                   CONTENTS


                             :                            !                 Page
Foreword	!	iii
Preface	.,	  .;	  .   iv
Abstract	;	   vi
Figures	,  .	xii
Tables  . . .	;	   xv
Acknowledgments .		:  ........   xx

Section Number               '<                                             Page

     1.  Introduction	     1
     2.  Short-Term Methods for Estimating Chronic Toxicity	     3
            Introduction	:	     3
            Types of Tests	:	•.  .     5
            Static Tests  	     6
            Advantages and Disadvantages of Toxicity Test Types  	     7
     3.  Health and Safety	     8
            General Precautions 	  	     8
            Safety Equipment  	  	     8
            General Laboratory and Field Operations 	     8
            Disease Prevention	  .     9
            Safety Manuals	;	     9
            Waste Disposal  .....	     9
     4.  Quality Assurance	;	   10
            Introduction  .	   10
            Facilities, Equipment, and Test Chambers	  .   10
            Test Organisms  .'.	;	   11
            Laboratory Water Used for Culturing and
              Test Dilution Water	;	   11
            Effluent and. Receiving Water Sampling and
              Handling  \	.......;	   11
            Test Conditions	   11
            Quality of Test Organisms	  .   12
            Food Quality  .	   12
            Acceptability of Short-Term Chronic Toxicity Tests   	   13
            Analytical Methods  	  	   13
            Calibration and Standardization .	:	   14
            Replication and Test Sensitivity	........   14
            Variability in Toxicity Test Results	   14
            Test Precision  . . . \. .	   14
            Demonstrating Acceptable Laboratory Performance  	   16
            Documenting Ongoing Laboratory Performance   .!  	   16
            Reference.Toxicants 	  	   17
            Record Keeping	   19
            Video Tapes of USEPA Culture and Toxicity    ,
              Test Methods	   19
            Supplemental Reports for Training Video Tapes:	   20
                                      vn

-------
                             CONTENTS (CONTINUED)

Section Number                                                            Page


     5.  Facilities, Equipment, and Supplies  .... 	.  . .  I .    21
            General Requirements  	 ............    21
            Test Chambers	'<. .    22
            Cleaning Test Chambers and Laboratory Apparatus .........    22
            Apparatus and Equipment for Culturing and Toxicity Tests  '. .    23
            Reagents and Consumable Materials	:. .    24
            Test Organisms	 .    24
            Supplies		    24
     6.  Test Organisms	    26
            Test Species	• . .  I .    26
            Sources of Test Organisms	'. .    27
            Life Stage	    28
            Laboratory Culturing	;. .    28
            Holding and Handling Test Organisms	I .    28
            Transportation to the Test Site	I .    29
            Test Organism Disposal	;. .    30
     7.  Dilution Water	|. .    31
            Types of Dilution Water	 .    31
            Standard, Synthetic Dilution Water  	 .......    31
            Use of Receiving Water as Dilution Water  ..........    32
            Use of Tap Water as Dilution Water	 .    34
            Dilution Water Holding	........  i .    34
     8.  Effluent and Receiving Water Sampling,  Sample Handling,      ;
              and Sample Preparation for Toxicity Tests	i. .    35
            Effluent Sampling 	  .........    35
            Effluent Sample Types	 .    35
            Effluent Sampling Recommendations 	  . .    36
            Receiving Water Sampling  	  . .    37
            Effluent and Receiving Water Sample Handling,
              Preservation,  and Shipping	 .    38
            Sample Receiving	 .    39
            Persistence of Effluent Toxicity During  Sample            i
              Shipment and Holding	 .  i .    39
            Preparation of Effluent and Receiving Water Samples
              for Toxicity Tests  	 .....;.    39
            Preliminary Toxicity Range-Finding Tests  	  I .    42
            Multi-concentration (Definitive) Effluent
              Toxicity Tests	i .    42
            Receiving Water Tests	; .    43
     9.  Chronic Toxicity Test Endpoints and Data Analysis  	  i .    44
            Endpoints	 .    44
            Relationship Between Endpoints Determined by
              Hypothesis Testing and Point Estimation Techniques  . .  ; .    45
            Precision	j .    47
            Data Analysis	I .    47
            Choice of Analysis	i .    49
            Hypothesis Tests  .  .	; .    52
            Point Estimation Techniques	i .    53

                                     viii

-------
                             CONTENTS (CONTINUED)        I

Section Number                                                            Page


     10. Report Preparation .;	'	   55
            Introduction  . ; . .	  .   55
            Plant Operations  . .	   55
            Source of Effluent, Receiving Water, and Dilution Water  ...   55
            Test Methods	„ .	   56
            Test Organisms	   56
            Quality Assurance	   56
            Results	   56
            Conclusions and Recommendations  	  	   57
     11. Test Method:  Fathead Minnow, Pimephales promelas,  Larval
            Survival and Growth Test Method  1000.0   .	   58
              Scope and Application	•...,.	   58
              Summary of Method	1	   58
              Interferences	   58
              Safety	I-	   59
              Apparatus and Equipment	„	   59
              Reagents and Consumable Materials  ..... 	   60
              Effluent and Receiving Water Collection, Preservation,
                and Storage	   68
              Calibration and Standardization	,	   68
              Quality Control	: .	   69
              Test Procedures	   69
              Summary of Test Conditions and Test Acceptability
                Criteria	   76
              Acceptability of Test Results	   81
              Data Analysis	   81
              Precision and Accuracy	;	108
     12. Test Method:  Fathead Minnow, Pimephales promelas,  Embryo-larval
            Survival and Teratogenicity Test Method  1001.0   	   114
              Scope and Application	   114
              Summary of Method	:	114
              Interferences	i	114
              Safety  . . . :	, .  . '	115
              Apparatus and Equipment	,	115
              Reagents and Consumable Materials	   116
              Effluent and Receiving Water Collection, Preservation,
                and Storage	:	118
              Calibration and Standardization  	   118
              Quality Control 	  .........   118
              Test Procedures	118
              Summary of Test Conditions and Test Acceptability
                Criteria	: .	126
              Acceptability of Test Result	126
              Data Analysis	130
              Precision and Accuracy	;	139
                                       IX

-------
                             CONTENTS (CONTINUED)
Section Number
Page
     13. Test Method:  Daphnid, Ceriodaphnia dubia, Survival and      :
            Reproduction Test Method 1002.0  .....  	 .  .  144
              Scope and Application	  .  144
              Summary of Method	j.  .  144
              Interferences	...........  144
              Safety	:.  .  145
              Apparatus and  Equipment		  . i.  .  145
              Reagents and Consumable Materials	;.  .  147
              Effluent and Receiving Water Collection, Preservation,  •
                and Storage	•  •  158
              Calibration and Standardization  .	158
              Quality Control	  .  158
              Test Procedures	  158
              Summary of Test Conditions and Test Acceptability       ;
                Criteria	.........;..  165
              Acceptability  of Test Results	i.  .  165
              Data Analysis	170
              Precision and  Accuracy   	  189
     14. Test Method:  Green Alga, Selenastrum capn'cornutum, Growth
            Test Method 1003.0	196
              Scope and Application	196
              Summary of Method 	  196
              Interferences	i  .  196
              Safety	197
              Apparatus and  Equipment	197
            ,  Reagents and Consumable Materials  	  198
              Effluent and Receiving Water Collection, Preservation,
                and Storage	'.  .  203
              Calibration and Standardization	:.  .  203
              Quality Control		203
              Test Procedures	  .  203
              Summary of Test Conditions and Test Acceptability
                Criteria	  .  209
              Acceptability  of Test Results	;.  .  209
              Data Analysis	212
              Precision and  Accuracy	i.  .  225

     Cited References	!.  .  229
     Bibliography	,  .  239
     Appendices	;.  .  246
         A. Independence, Randomization, and Outliers	;.  .  248
         B. Validating Normality and Homogeneity of Variance
              Assumptions	254
         C. Dunnett's Procedure	271
         D. T test with Bonferroni's Adjustment	  283
         E. Steel's Many-one Rank Test	i.  .  289
         F. Wilcoxon Rank Sum Test	  293

-------
                             CONTENTS (CONTINUED)

Section Number               ;                                             Page


         G. Fisher's Exact Test .  .	,'	299
         H. Single Concentration Toxicity Test - Comparison
              of Control with 100% Effluent or Receiving Water  	  308
         I. Probit Analysis	312
         J. Spearman-Karber Method	315
         K. Trimmed Spearman-Karber Method  	 . 	  320
         L. Graphical Method  	  325
         M. Linear Interpolation Method 	  329
         Cited References	340

-------
                                    FIGURES                           '•


SECTIONS 1-10                                                         ;

     Number                                                           !    page

     1.  Control (cusum) charts	!. .   is

     2.  Flowchart for statistical analysis of test data  ...... I .   50

SECTION 11

     Number                                                               Page

     1.  Data form for the fathead minnow, Pimephales promelas,
         larval survival and growth test.  Routine chemical  and
         physical determinations	, .    73

     2.  Mortality data for the fathead minnow, Pimephales promelas,  -
         larval survival and growth test	    75

     3.  Weight data for the fathead minnow,  Pimephales promelas,
         larval survival and growth test	    77

     4.  Summary data for the fathead minnow,  Pimephales promelas,
         larval survival and growth test	    78

     5.  Flowchart for statistical analysis of fathead minnow,
         Pimephales promelas, larval  survival  data by hypothesis      ^
         testing	i  .    83

     6.  Flowchart for statistical analysis of fathead minnow,
         Pimephales promelas, larval  survival  data by point            :
         estimation	      84

     7.  Plot of survival  proportion  data in Table 3	i  .    86

     8.  Output for USEPA Probit  Analysis program,  Version 1.5   .  .  . ;  .    95

     9.  Flowchart  for statistical analysis of fathead minnow,
         Pimephales promelas, larval  growth data   	 .  .    96

     10.  Plot of weight data from fathead minnow,  Pimephales
         promelas,  larval  survival and  growth  test for point
         estimate testing	'.  .  .  . J  .    98

     11.  Plot of raw data,  observed means,  and smoothed means for      ' .
         the  fathead minnow,  Pimephales promelas,  growth data          ;
         in Tables  2 and 18	;    106
                                     xn

-------
                              FIGURES  (CONTINUED)

SECTION 11 (Continued)       j

     Number                                              ;                 Page

     12. ICPIN program output for the IC25	109

     13. ICPIN program output for the IC50	110

SECTION 12

     Number                                                               Page

     1.  Data form for the fathead minnow, Pimephales promelas,
         embryo-larval survival and teratogenicity test.  Routine
         chemical and physical determinations	,.	122

     2.  Data form for the fathead minnow, Pimephales promelas,
         embryo-larval survival and teratogenicity test.: Survival
         and terata data  '.  .	124

     3.  Summary data for the  fathead minnow,  Pimephales promelas,
         embryo-larval survival and teratogenicity test  . 	  127

     4.  Flowchart for statistical analysis of fathead minnow,
         Pimephales promelas,  embryo-larval data   	  131

     5.  Plot of fathead minnow, Pimephales promelas, total  mortality
         data from the embryo-larval  test		134

     6.  Output for USEPA Probit Analysis  program, Version  1.5   	  140

 SECTION  13

     Number                                              '                  Page

     1.   Examples of  a test  board  and randomizing template	156

     2.  Data form  for the daphnid,  Ceriodaphnia  dubia,  survival
          and reproduction test.   Routine chemical and physical
         determinations	   163

     3.   Data form  for the daphnid,  Ceriodaphnia  dubia,  survival
          and reproduction test.   Daily summary of data   .  	  : .  .   166

     4.   Flowchart  for  statistical  analysis  of the daphnid,
          Ceriodaphnia dubia, survival data  	  	   172

     5.   Output  for USEPA Trimmed Spearman-Karber program 	   175
                                      xm

-------
                              FIGURES (CONTINUED)                      :
SECTION 13  (Continued)
     Number                                                               Page
     6.  Flowchart for statistical analysis of the daphnid,
         Cen'odaphnia dubia, reproduction data	  176
     7.  Plot of number of young per adult female from a daphnid,
         •Cen'odaphnia dubia, survival and reproduction test	j.  .  17.8
     8.  Plot of raw data, observed means, and smoothed means for
         the daphnid, Cen'odaphnia dubia, reproductive data  	  187
     9.  Example of ICPIN program output for the IC25 . . •	  190
                                                                      !
     10. Example of ICPIN program output for the IC50	j.  .  191
SECTION 14
     Number                                                           !    Page
     1.  Data form for the green alga,  Selenastrum capn'cornutum,     I
         growth test.  Routine chemical  and physical                   l
         determinations	  .  206
     2.  Data form for the green alga,  Selenastrum capn'cornutum,
         growth test, cell density determinations  .  .  .  .	 L  .  209
     3.  Flowchart for statistical  analysis of the green alga,
         Selenastrum capn'cornutum,  growth response data	:.  .  213
     4.  Plot of Iog10 transformed cell count data from the green
         alga,  Selenastrum capn'cornutum, growth response test in     :
         Table 4	......!..  215
     5.  Plot of raw data and observed  means  for the  green alga,
         Selenastrum capn'cornutum,  growth data	 .  224
     6.  ICPIN program output for the IC25	  . .  . ;. .  226
     7.  ICPIN program output for the IC50  	  ........ I .  227
                                     xiv

-------
                                    TABLES

                                                         I
SECTIONS 1-10

     Number                                   '           <                 Page

     1.  National interlaboratory study of chronic toxicity
         test precision, 1991:  Summary of responses using
         a reference toxicant	.;	   15

     2.  Commercial suppliers of brine shrimp (Artemia) cysts 	   25
                             i                            ;,
     3.  Preparation of synthetic freshwater using reagent
         grade chemicals	;	   33

     4.  Preparation of synthetic freshwater using
         mineral water	   34

     5.  Percent unionized NH3 in aqueous ammonia solutions:
         Temperatures 15-26ฐC and pH 6.0-8.9  	   41

SECTION 11

     Number                                              :                 Page

     1.  Summary of test conditions and test acceptability
         criteria for fathead minnow, Pimephales promelas,
         larval  survival and growth toxicity tests with
         effluents and receiving waters  	   79
                             ,                            1
     2.  Summary of survival and growth data for fathead minnow,
         Pimephales promelas, larvae exposed to a reference
         toxicant for seven days	   81

     3.  Fathead minnow, Pimephales promelas, survival data  .......   85

     4.  Centered observations for Shapiro-Wilk's example  	   87

     5.  Ordered centered observations for the
         Shapiro-Wilk's example	;	   87

     6.  Coefficients and differences for Shapiro-Wilk's example   ....   88

     7.  ANOVA  table	'	   90

     8.  ANOVA  table for Dunnett's Procedure  example   	   91

     9.  Calculated t values	!.  .	   92

     10. Data for  Probit Analysis	   94
                                       xv

-------
                              TABLES  (CONTINUED)

SECTION 11  (Continued)                                                i

     Number                                                           [    Page

     11. Fathead minnow, Pimephales promelas, growth data	(.  .   97

     12. Centered observations for Shapiro-Milk's example	!.  .   97

     13. Ordered centered observations for Shapiro-Milk's  example  ....   99

     14. Coefficients and differences for Shapiro-Milk's example   ....  100

     15. ANOVA table	  102

     16. ANOVA table for Dunnett's Procedure example	;.  .  103

     17. Calculated t values		104

     18. Fathead minnow, Pimephales promelas, mean growth response  .
         after smoothing	'.  .  107

     19. Precision of the fathead minnow, Pimephales promelas,
         larval survival and growth test, using NaPCP as a            ;
         reference toxicant	;  '.  108

     20. Combined frequency distribution for survival NOECs
         for all laboratories	;  .  112

     21. Combined frequency distribution for weight NOECs             :
         for all laboratories	,  .  113

SECTION 12                                                            '

     Number                                                               Page

     1.  Summary of test conditions and test acceptability criteria   '
         for fathead minnow, Pimephales promelas,  embryo-larval       I
         survival and teratogenicity toxicity tests with effluents
         and receiving waters	]   .  128

     2.  Data from fathead minnow, Pimephales promelas, embryo-larval \
         toxicity test with ground water effluent	i   .  132

     3.  Fathead minnow, Pimephales promelas, embryo-larval total
         mortality data	:   .  133

     4.  Centered observations for Shapiro-Milk's  example 	 j   .  135

     5.  Ordered centered observations for the Shapiro-Milk's example .'   .  135
                                """'                                     I
                                                                      I

                                     xvi                              i

-------
                              TABLES (CONTINUED)

SECTION 12 (Continued)                                   '

     Number                  :                                             Page

     6.  Coefficient and differences for the Shapiro-WiIks example  ...  136

     7.  Assigning Ranks to the control and 3.125% effluent
         concentration for the Steel's Many-One Rank Test: . . 	  137

     8.  Table of Ranks for Steel's Many-One Rank Test	138

     9.  Rank Sums	138

     10. Data for Probit Analysis	  139

     11. Precision of the fathead minnow, Pimephales promelas,
         embryo-larval survival and teratogenicity test, using
         cadmium as a reference toxicant	141

     12. Precision of the fathead minnow, Pimephales promelas,
         embryo-larval, survival and teratogenicity toxicity test,
         using diquat as a reference toxicant	142

     13. Precision of the fathead minnow, Pimephales promelas,
         embryo-larval survival and teratogenicity static-renewal
         test conducted with trickling filter  effluent   .  .	143

SECTION  13

     Number                                              :                  Page
                                                         i
     1.  Nutrient stock solutions  for maintaining algal  stock
         cultures  	  	  ........  150

     2.  Final concentration of macronutrients and micronutrients
         in  the culture medium	151

     3.  Summary of test conditions and  test  acceptability
         criteria for daphnid, Ceriodaphnia dubia, survival
         and reproduction toxicity tests with  effluents
         and receiving waters	  168

     4.  Summary of survival and reproduction  data for the
         daphnid, Ceriodaphnia dubia,  exposed  to  an  effluent
         for seven days	,	170
                                                         i
     5.   Format of the  2x2  contingency table   	  .........  173

     6.   2x2 contingency table for control  and 25% effluent	173


                                      xv ii

-------
                              TABLES  (CONTINUED)
SECTION 13  (Continued)                                                ,
     Number                                                                Page
     7.  Data for Trimmed Spearman-Karber Analysis	'.  .   174
     8.  The daphnid, Ceriodaphm'a dubia, reproduction data  	 .  .   177
     9.  Centered observations for Shapiro-Wilk's example  ...  .  . .. L  .   179
     10. Ordered centered observations for Shapiro-Wilk's  example  .  . •.  .   180
     11. Coefficients and differences for Shapiro-Wilk's example   .  . i.  .   181
     12. ANOVA table	 -. .;  .  .... ,  .  ...... |.  .   183
     13. ANOVA table for Dunnett's Procedure example  ...  	 I  .   185
     14. Calculated t values	i.  .   185
     15. Daphnid, Ceriodaphm'a dubia, reproduction mean
         response after smoothing	[..  .   188
     16. Single laboratory precision of the daphnid, Ceriodaphnia     \
         dubia, survival and reproduction test, using NaPCP  as a      ;
         reference toxicant	  .   192
     17. The daphnid, Ceriodaphnia dubia, seven-day survival and
         reproduction test precision for a single laboratory using
         NaPCP as the reference toxicant	'.	193
     18. Inter!aboratory precision of the daphnid, Ceriodaphnia
         dubia, survival and reproduction test with copper            \
         spiked effluent	!.  .   194
     19. Inter!aboratory precision data for the daphnid,
         Ceriodaphnia dubia, summarized for eight reference
         toxicants and effluents	195
SECTION 14                                                            I
     Number                                                           ;     Page
     1.  Nutrient stock solutions for maintaining algal  stock         ;
         cultures and test control cultures		L  .   200
     2.  Final  concentration of macronutrients and micronutrients
         in the culture medium	I   .   202
                                     xvm

-------
                              TABLES (CONTINUED)
SECTION 14 (Continued)
     Number
Page
     3.  Summary of test conditions and test acceptability criteria
         for green alga, Selenastrum capricornutum, growth toxicity
         tests with effluents and receiving waters  ..... 	  210
     4.  Green alga, Selenastrum capricornutum, growth
         response data	212
     5.  Centered observations for Shapiro-Wilk's example 	  214
     6.  Ordered centered observations for Shapiro-Wilk's example ....  216
     7.  Coefficients and differences for Shapiro-Wilk's example  ....  217
     8.  ANOVA table	:	219
     9.  ANOVA table for Dunnett's Procedure example  .	  220
     10. Calculated t values	'	  221
     11. Algal mean growth response after smoothing ... 	  223
     12. Single laboratory precision of the green alga,
         Selenastrum capriccurnutum, 96-h toxicity tests,
         using the reference toxicant cadmium chloride   	  228
                                     xix

-------
                                ACKNOWLEDGMENTS


   The principal authors of this document are  Philip Lewis, Donald Klemm,
Cornelius Weber, James Lazorchak, and Florence Fulk, Environmental Systems
Laboratory-Cincinnati, OH; William Peltier, Environmental Services Division,
Region 4, Athens, GA; Teresa Norberg-King, Environmental Research Laboratory,
Duluth, MM; and Cathy Poore, Computer Sciences Corporation, Cincinnati;, OH.
Contributors to specific sections of this manual are listed below.

1.  Sections 1-10; General Guidelines                                 \

    Margarete Heber, OST, Office of Water                             ;
    Donald Klemm, EMSL-Cincinnati                                     '
    Philip Lewis, EMSL-Cincinnati                                     :
    George Morrison, ERL-Narragansett                                 !
    Teresa Norberg-King, ERL-Duluth                                   !
    William Peltier, ESD, Region 4                                    •
    Cornelius Weber, EMSL-Cincinnati

2.  Sections 11-13; Toxicity Test Methods                             j

    Donald Klemm, EMSL-Cincinnati                                     ;
    James Lazorchak, EMSL-Cincinnati                                  ;
    Philip Lewis, EMSL-Cincinnati
    Donald Mount, ERL-Duluth
    Teresa Norberg-King, ERL-Duluth
    Quentin Pickering, EMSL-Cincinnati                                !
                                                                      1
3.  Data Analysis (Sections 9, 11-13, and Appendices)                 j

    Florence Fulk, EMSL-Cincinnati                                    !
    Laura Gast, Technology Application, Inc (TAI)                     |
    Cathy Poore, Computer Sciences Corporation (CSC)                  !

      Review comments from the following persons are gratefully acknowledged:

Celeste Philbrick Barr, Environmental Services Division, Biology Section,
  U.S. Environmental Protection Agency, Region 1, Lexington, MA.
Michael Tucker, Bjoassay Lab. Environmental Services Division, U.S.   :
  Environmental Protection Agency, Region 7, Kansas City, MO.         \
Michael Morton, Environmental Services Division, U.S. Environmental
 Protection Agency, Region 6, Dallas, TX.
Jerry Smrchek, Environmental Effects Branch, Health and Environmental [Review
  Division, U.S. Environmental Protection Agency, Washington, DC.
Robert Donaghy, Environmental Services Division, U.S. Environmental Protection
  Agency, Region 3, Wheeling, WV.
Philip Crocker, Water Quality Management Branch, U.S. Environmental   \.
 Protection Agency, Region 6, Dallas, TX.                             !
Chick Steiner, Central Regional Laboratory, U.S. Environmental Protection
  Agency, Region 5, Chicago, IL.                                      \
                                      xx

-------
                          ACKNOWLEDGMENTS (CONTINUED)


Tom Waller, Institute of Applied Sciences, University of North Texas, Denton,
TX.                          ,

   Many useful public comments on the second edition of the freshwater
toxicity test methods "Short-Term Methods for Estimating the Chronic Toxicity
of Effluents and Receiving Waters to Freshwater Organisms", EPA/600/4-89/001
(USEPA, 1989a) were received in response to the proposed rule, published in
the Federal Register, December 4, 1989,  [FR 54(231):50216-50224], regarding
the Agency's intent to include the short-term chronic toxicity tests in
Table IA, 40 CFR Part 136.  These comments were considered in the preparation
of the third edition of the manual, and  are included in the Public Docket for
rulemaking, located in room 2904, USEPA  Headquarters, Washington, DC.

   Materials in this manual w.ere taken in part from the following sources:
Methods for Acute Toxicity Tests with Fish, Macroinvertebrates, and
Amphibians, Environmental Research Laboratory, U. S. Environmental Protection
Agency, Duluth, MN, EPA-660/3-75-009 (USEPA, 1975); Handbook for Analytical
Quality Control in Water and Wastewater  Laboratories, Environmental Monitoring
and Support Laboratory-Cincinnati, U. S. Environmental Protection Agency,
Cincinnati, OH, EPA-600/4-79/019 (USEPA, 1979a); Methods for chemical analysis
of water and wastes, Environmental Monitoring and Support Laboratory-
Cincinnati, U. S. Environmental Protection Agency, Cincinnati, OH, EPA-600/4-
79-020 (USEPA, 1979b); Interim NPDES Compliance Biomonitoring Inspection
Manual, Enforcement Division, Office of  Water Enforcement, U. S. Environmental
Protection Agency, Washington, DC, (USEPA, 1979c); NPDES compliance inspection
manual, Office of Water Enforcement and  Permits (EN-338), U. S. Environmental
Protection Agency, Washington, D. C. (USEPA, 1988a); Methods for Measuring the
Acute Toxicity of Effluents to Freshwater and Marine Organisms, Environmental
Monitoring and Support Laboratory-Cincinnati, U. S. Environmental Protection
Agency, Cincinnati, OH, EPA-600/4-85/013 (USEPA, 1985a); Short-term Methods
for Estimating the Chronic Toxicity of Effluents and Receiving Waters to
Freshwater Organisms, Environmental Monitoring Systems Laboratory-Cincinnati,
U.S. Environmental Protection. Agency, Cincinnati, OH, EPA/600/4-89/001 (USEPA
1989a); A Seven-day Life-cycle Cladoceran Test,  Environ. Toxicol. Chem.
3:425-434 (Mount, D. I. and 1. J. Norberg, 1984); A New Fathead Minnow
(Pimephales promelas) Subchronic Toxicity Test, Environ. Toxicol. Chem.
4:711-718 (Norberg, T. J., arid D. I. Mount, 1985); The Selenastrum
capn'cornutum Printz Algal Assay Bottle  Test, Environmental Research
Laboratory, U. S. Environmental Protection Agency, Corvallis, OR,
EPA-600/9-78-018 (USEPA, 1978a); Short-term Methods for Estimating the Chronic
Toxicity of Effluents and Receiving Waters to Marine and Estuarine Organisms,
Environmental Monitoring and Support Laboratory-Cincinnati, U. S.
Environmental Protection Agency, Cincinnati, OH, EPA-600/4-87/028 (USEPA,
1988b); Short-term Methods for Estimating the Chronic Toxicity of Effluents
and Receiving Waters to Marine and Estuarine Organisms (Second Edition),
Environmental Monitoring Systems Laboratory-Cincinnati, U.S. Environmental
Protection Agency, Cincinnati, OH, EPA/600/4-91/003 (USEPA, 1993a); Methods
for Measuring the Acute Toxicity of Effluents and Receiving Waters to
Freshwater and Marine Organisms (Fourth  Edition), Environmental Monitoring

                                     xxi                '

-------
                          ACKNOWLEDGMENTS (CONTINUED)
Systems Laboratory-Cincinnati, U.S. Environmental Protection Agency,
Cincinnati, OH, EPA/600/4-90/027F  (USEPA, 1993b); and Technical Support
Document for Water Quality-Based Toxics Control, Office of Water Enforcement
and Permits, Office of Water Regulations and Standards, U.S. Environmental
Protection Agency, Washington, DC, EPA/505/2-90-001 (USEPA, 1991a).   ;

   The fathead minnow, Pimephales promelas, larval survival and growth test
method and the daphnid, Cen'odaphnia dubia, survival and reproduction ;test
method in this manual were adapted from methods developed at the ERL-Duluth by
Donald Mount and Teresa Norberg-King.  The fathead minnow, Pimephales '•
promelas, embryo-larval survival and teratogenicity test method was developed
by Wesley Birge and Jeffrey Black, University of Kentucky, Lexington, junder a
cooperative agreement with the EMSL-Cincinnati.  The algal growth test method
was adapted from the green alga, Selenastrum capn'cornutum, algal assay bottle
test developed at the Environmental Research Laboratory-Corvallis by William
Miller, Joseph Greene, and Tamotsu Shiroyama.                         \
                                                                      i
   Debbie Hall, Bioassessment and  Ecotoxicology Branch, and Mary Sullivan,
Quality Assurance Research Division, provided valuable secretarial assistance,
and Betty Thomas, Technical Information Manager, EMSL-Cincinnati, provided an
editorial review.                                                     i
                                     xxn

-------
                                   SECTION 1

                                 INTRODUCTION


1.1  This manual describes chronic toxicity tests for use in the National
Pollutant Discharge Elimination System (NPDES) Permits Program to identify
effluents and receiving waters containing toxic materials in chronically toxic
concentrations.  The methods included in this manual are referenced in
Table IA, 40 CFR Part 136 regulations and, therefore, constitute approved
methods for chronic toxicity tests.  They are also suitable for determining
the toxicity of specific compounds contained in discharges.  The tests may be
conducted in a central laboratory or on-site, by the regulatory agency or the
permittee.                                               ;

1.2  The data are used for NPDES permits development and to determine
compliance with permit toxicity limits.  Data can also be:used to predict
potential acute and chronic toxicity in the receiving water, based on the
LC50, NOEC, IC50 or IC25 (see Section 9, Chronic Toxicity Endpoints and Data
Analysis) and appropriate dilution, application, and persistence factors.  The
tests are performed as a part of self-monitoring permit requirements,
compliance biomonitoring inspections, toxics sampling inspections, and special
investigations.  Data from chronic toxicity tests performed as part of permit
requirements are evaluated during compliance evaluation inspections and
performance audit inspections.

1.3  Modifications of these tests are also used in toxicity reduction
evaluations and toxicity identification evaluations to identify the toxic
components of an effluent, to aid in the development and implementation of
toxicity reduction plans, and to compare and control the effectiveness of
various treatment technologies for a given type of industry, irrespective of
the receiving water (USEPA, 1988c; USEPA, 1989b; USEPA 1989c; USEPA, 1989d;
USEPA, 1989e; USEPA, 1991a; USEPA, 1991b; and USEPA, 1992).

1.4  This methods manual serves as a companion to the acute toxicity test
methods for freshwater and marine organisms (USEPA, 1993b), the short-term
chronic toxicity test methods for marine and estuarine organisms (USEPA,
1993a), and the manual for evaluation of laboratories performing aquatic
toxicity tests (USEPA, 1991c).

1.5  Guidance for the implementation of toxicity tests in the NPDES program is
provided in the Technical Support Document for Water Quality-based Toxics
Control (USEPA, 1991a).      ;

1.6  These freshwater short-term toxicity tests are similar to those developed
for marine and estuarine organisms to evaluate the toxicity of effluents
discharged to marine and estuarine waters under the NPDES permit program.
Methods are presented in this manual for three species of freshwater organisms
from three phylogenetic groups.  The methods are all static renewal  type
seven-day tests except the green alga,  Selenastrum capricornutum,  test which
lasts four days.

-------
1.7  The three species for which test methods are provided are the fathead
minnow, Pimephales promelas-, the daphnid, Ceriodaphnia dubia; and the .green
alga, Selenastrum capri'cornutum.

1.7.1  The tests included in this document are based on the following ^ethods:

   1. "A new fathead minnow (Pimephales promelas) subchronic toxicity ftest,"
      by Teresa J. Norberg and Donald I. Mount, 1985, Environmental Toxicology
      and Chemistry (Norberg and Mount, 1985).                        '

   2. "In-situ acute/chronic toxicological monitoring of industrial effluents
      for the NPDES biomonitoring program using fish and amphibian    ;
      embryo/larval stages as a test organism," by Wesley J. Birge and Jeffrey
      A. Black, 1981, OWEP-82-001, Office of Water Enforcement and Permits,
      U.S. Environmental Protection Agency, Washington, DC (USEPA, 1981).

   3. "A seven-day life-cycle cladoceran test,", by Donald I. Mount arid
      Teresa Norberg, 1984, Environmental Toxicology and Chemistry (Mount and
      Norberg, 1984).                                                 j

   4. "The Selenastrum capricornutum Printz algal assay bottle test,";by
      William E. Miller, Joseph C. Greene and Tamotsu Shiroyama,  1978,
      Environmental research Laboratory, U.S. Environmental  Protection
      Agency, Con/all is, OR.  EPA/600/9-78/018  (USEPA, 1978a).        j

1.7.2  Two of the methods incorporate the chronic endpoint of growth  in
addition to lethality and one incorporates reproduction.  The fathead;minnow,
Pimephales promelas, embryo-larval survival and teratogenicity test
incorporates teratogenic effects  in addition to lethality.   The green,alga,
Selenastrum capricornutum, growth test  has the  advantage of  a relatively  short
exposure period (96 h).

1.8  The validity of the freshwater chronic methods  in predicting adverse
ecological impacts of toxic discharges  was demonstrated in field  studies
(USEPA, 1984; USEPA, 1985b; USEPA, 1985c; USEPA, 1985d; USEPA, 1986a;:USEPA,
1986b; USEPA, 1986c; USEPA, J,986d; Birge et al., 1989; and Eagleson et  al.,
1990).                                                                ;

1.9  The use of any test species  or test conditions  other than those  described
in the methods summary tables in  this manual  shall be  subject to  application
and  approval of alternate test  procedures under 40 CFR 136.4 and  40 CFR 136.5.

1.10 These methods are restricted to  use by,  or under the supervision  of,
analysts experienced in the use or conduct of aquatic  toxicity tests  and  the
interpretation of data from aquatic toxicity  testing.  Each  analyst must
demonstrate the ability to generate acceptable  test  results  with  these  methods
using the procedures described  in this  methods  manual.                ;

1.11 This manual was  prepared  in the  established  EMSL-Cincinnati format
(USEPA, 1983).                                                        i

-------
                                  SECTION 2

              SHORT-TERM METHODS FOR  ESTIMATING CHRONIC TOXICITY


2.1   INTRODUCTION
                                                          I
                                                          i
2.1.1  The objective of aquatic toxicity tests with effluents or pure
compounds is to estimate the  "safe" or  "no effect" concentration of these
substances, which  is defined  as the concentration which will permit normal
propagation of fish and other aquatic life in the receiving waters.  The
endpoints that have been considered in  tests to determine the adverse effects
of toxicants include death and survival, decreased reproduction and growth,
locomotor activity, gill ventilation rate, heart rate, blood chemistry,
histopathology, enzyme activity, olfactory function, and terata.  Since it is
not feasible to detect and/or measure all of these (and other possible)
effects of toxic substances on a routine basis, observations in toxicity tests
generally have been limited to only a few effects, such a<; mortality, growth,
and reproduction.

2.1.2  Acute lethality is an  obvious and easily observed effect which accounts
for its wide use in the early period of evaluation of the toxicity of pure
compounds and complex effluents.  The results of these tests were usually
expressed as the concentration lethal to 50% of the test organisms (LC50) over
relatively short exposure periods (one-to-four days).     '

2.1.3  As exposure periods of acute tests were lengthened, the LC50 and lethal
threshold concentration were  observed to decline for many ;compounds.  By
lengthening the tests to include one or more complete life cycles and
observing the more subtle effects of the toxicants, such as a reduction in
growth and reproduction, more accurate, direct, estimates of the threshold or
safe concentration of the toxicant could be obtained.  However,  laboratory
life-cycle tests may not accurately estimate the "safe" concentration of
toxicants because they are conducted with a limited number of species under
highly controlled, steady-state conditions,  and the results do not include the
effects of the stresses to which the organisms would ordinarily be exposed in
the natural  environment.

2.1.4  An early published account of a full  life-cycle, fish toxicity test was
that of Mount and Stephan (1967).   In this study,  fathead minnows,  Pimephales
promelas, were exposed to a graded series of pesticide concentrations
throughout their life cycle,   and the effects of the toxicant on  survival,
growth, and reproduction were measured and evaluated.  This work was soon
followed by full  life-cycle tests using other toxicants and fish species.

2.1.5  McKim (1977) evaluated the data from 56 full  life-cycle tests,  32- of
which used the fathead minnow, Pimephales promeTas,  and concluded that the
embryo-larval  and early juvenile life-stages were  the most sensitive stages.
He proposed the use of partial life-cycle toxicity tests with the early
life-stages (ELS)  of fish to  establish water quality criteria.

-------
2.1.6  Macek and Sleight (1977) found that exposure of critical life-stages of
fish to toxicants provides estimates of chronically safe concentrations
remarkably similar to those derived from full life-cycle toxicity tests.  They
reported that "for a great majority of toxicants, the concentration which will
not be acutely toxic to the most sensitive life stages is the chronically safe
concentration for fish, and that the most sensitive life stages are the
embryos and fry".  Critical life-stage exposure was considered to be exposure
of the embryos during most, preferably all, of the embryogenic (incubation)
period, and exposure of the fry for 30 days post-hatch for warm water jfish
with embryogenic periods ranging from one-to-fourteen days, and for 60 days
post-hatch for fish with longer embryogenic periods.  They concluded that in
the majority of cases, the maximum acceptable toxicant concentration (MATC)
could be estimated from the results of exposure of the embryos during
incubation, and the larvae for 30 days post-hatch.
                                                            *

2.1.7  Because of the high cost of full life-cycle fish toxicity tests and the
emerging consensus that the ELS test data usually would be adequate for
estimating chronically safe concentrations, there was a rapid shift by aquatic
toxicologists to 30 - 90-day ELS toxicity tests for estimating chronically
safe concentrations in the late 1970s.  In 1980, USEPA adopted the policy that
ELS test data could be used in establishing water quality criteria if idata
from full life-cycle tests were not available (USEPA, 1980a).         i
2.1.8  Published reports of the results of ELS tests indicate that
relative sensitivity of growth and survival as endpoints may be species
dependent, toxicant dependent, or both.  Ward and Parrish (1.980) examined the
literature on ELS tests that used embryos and juveniles of the sheepshead
minnow, Cypn'nodon variegatus, and found that growth was not a statistically
sensitive indicator of toxicity in 16 of 18 tests.  They suggested that the
ELS tests be shortened to 14 days posthatch and that growth be eliminated as
an indicator of toxic effects.

2.1.9  In a review of the literature on 173 fish full life-cycle and ELS tests
performed to determine the chronically safe concentrations of a wide variety
of toxicants, such as metals, pesticides, organics, inorganics, detergents,
and complex effluents, Weltering (1984) found that at the lowest effect
concentration, significant reductions were observed in fry survival in 57%,
fry growth in 36%, and egg hatchability in 19% of the tests.  He also found
that fry survival and growth were very often equally sensitive, and concluded
that the growth response could be deleted from routine application of the ELS
tests.  The net result would be a significant reduction in the duration and
cost of screening tests with no appreciable impact on estimating MATCs for
chemical hazard assessments.  Benoit et al . (1982), however, found larval
growth to be the most significant measure of effect, and survival to be
equally or less sensitive than growth in early life-stage tests with four
organic chemicals.                                                    i
                                                                      i
2.1.10  Efforts to further reduce the length of partial life-cycle toxicity
tests for fish without compromising their predictive value have resulted in
the development of an eight-day, embryo-larval survival and teratogenicity
test for fish and other aquatic vertebrates (USEPA, 1981; Birge et all, 1985),
and a seven-day larval survival and growth test (Norberg and Mount, 1985).

-------
 2.1.11  The  similarity  of  estimates  of  chronically  safe  concentrations  of
 toxicants derived  from  short-term, embryo-larval  survival  and  teratogenicity
 tests to those derived  from  full  life-cycle  tests has  been demonstrated by
 Birge et al.  (1981),  Birge and Cassidy  (1983),  and  Birge et  al.  (1985).

 2.1.12  Use  of a seven-day,  fathead  minnow,  Pimephales promelas,  larval
 survival and growth test was first proposed  by  Norberg and; Mount  at  the 1983
 annual meeting of  the Society for Environmental Toxicology arid Chemistry
 (Norberg and Mount, 1983).   This  test was  subsequently used  by Mount and
 associates  in field demonstrations at Lima,  OH  (USEPA, 1984),  and  at many
 other locations.   Growth was frequently found to  be more sensitive than
 survival in  determining the  effects  of  complex  effluents.

 2.1.13  Norberg and Mount  (1985)  performed three  single  toxicant  fathead
 minnow larval growth tests with zinc, copper, and DURSBANฎ,  using  dilution
 water from Lake Superior.  The results were  comparable to, and had confidence
 intervals that overlapped  with, chronic values  reported  in the literature for
 both ELS and full  life-cycle tests.

 2.1.14  Mount and  Norberg  (1984)  developed a seven-day cladoceran  partial
 life-cycle test and experimented  with a number  of diets  for  use in culturing
 and testing  the daphnid, Ceriodaphm'a reticulata  (Norberg  and Mount,  1985).
 As different laboratories  began to use this  cladoceran test, it was  discovered
 that apparently more than  one species was  involved  in  the  tests conducted by
 the same laboratory.  Berner (1986)  studied  the problem  and  determined  that
 perhaps as many as three variant  forms were  involved and it  was decided  to
 recommend the use  of the more common Ceriodaphm'a dubia  rather than  the
 originally reported Ceriodaphm'a  reticulata.  The method was adopted  for use
 in the first edition of the freshwater short-term chronic methods  (USEPA,
 1985e),

 2.1.15  The  green  alga, Selenastrum capricornutum, bottle test was developed,
 after extensive design, evaluation, and application, for the National
 Eutrophication Research Program (USEPA,  1971).  The test was later modified
 for use in the assessment of receiving waters and the  effects of wastes
 originating  from industrial,  municipal,  and agricultural  point and non-point
 sources (USEPA,  1978a).

 2.1.16  The  use of short-term toxicity tests including subchronic  and chronic
 tests in the NPDES Program is especially attractive because they provide a
more direct  estimate of the safe concentrations of effluents in receiving
waters than was provided by acute toxicity tests,  at an only slightly
 increased level  of effort,  compared to the fish full life-cycle chronic  and
 28-day ELS tests and the 21-day daphnid, Daphm'a magna, life-cycle test.

 2.2  TYPES OF TESTS

2.2.1  The selection of the test type will  depend on the  NPDES permit
requirements, the objectives  of the test,  the available resources, the
requirements of the test organisms,  and  effluent characteristics  such as
fluctuations in  effluent toxicity.

-------
2.2.2  Effluent chronic toxicity i's generally measured using a multi- I
concentration, or definitive test, consisting of a control and a minimum of
five effluent concentrations.  The tests are designed to provide dose-response
information, expressed as the percent effluent concentration tha't affects the
hatchability, gross morphological abnormalities, survival, growth, and/or
reproduction within the prescribed period of time (four to seven days)..  The
results of the tests are expressed in terms of the highest concentration that
has no statistically significant observed effect on those responses when
compared to the controls or the estimated concentration that causes a
specified percent reduction in responses versus the controls.         \

2.2.3  Use of pass/fail tests consisting of a single effluent concentration
(e.g., the receiving water concentration or RWC) and a control is not
recommended.  If the NPDES permit has a whole effluent toxicity limit for
acute toxicity at the RWC, it is prudent to use that permit limit as the
midpoint of a series of five effluent concentrations.  This will ensur;e that
there is sufficient information on the dose-response relationship.  For
example, the effluent concentrations utilized in a test may be:
(1) 100% effluent, (2) (RWC + 100)/2, (3) RWC,  (4) RWC/2, and (5) RWC/14.  More
specifically, if the RWC = 50%, the effluent concentrations used in the
toxicity test would be 100%, 75%, 50%, 25%, and 12.5%.                ;

2.2.4  Receiving (ambient) water toxicity tests commonly employ two   >
treatments, a control and the undiluted receiving water, but may also consist
of a series of receiving water dilutions.

2.2.5  A negative result from a chronic toxicity test does not preclude the
presence of toxicity.  Also, because of the potential temporal variability in
the toxicity of effluents, a negative test result with a particular sample
does not preclude the possibility that samples  collected at some otheir time
might exhibit chronic toxicity.

2.2.6  The frequency with which chronic toxicity tests are conducted under a
given NPDES permit is determined by the regulatory agency on the basis of
factors such as the variability and degree of toxicity of the waste,
production schedules, and process changes.

2.2.7  Tests recommended for use in this methods manual may be staticinon-
renewal or static renewal.   Individual methods  specify which static type of
test is to be conducted.                                              :

2.3  STATIC TESTS

2.3.1  Static non-renewal tests  - The test organisms are exposed to the same
test solution for the duration of the test.                           t

2.3.2  Static-renewal tests  - The test organisms are exposed to  a fresh
solution of the same concentration of sample every 24 h or other prescribed
interval, either by transferring the test organisms from one test chamber to
another, or by replacing all or  a portion of solution in the test chambers.

-------
2.4  ADVANTAGES AND DISADVANTAGES OF TOXICITY TEST TYPES

2.4.1  STATIC NON-RENEWAL,  SHORT-TERM TOXICITY TESTS:     |

   Advantages:

   1. Simple and inexpensive.
   2. Very cost effective in determining compliance with permit conditions.
   3. Limited resources (space, manpower, equipment) required; would permit
      staff to perform many more tests in the same amount of time.
   4. Smaller volume of effluent required than for static renewal or flow-
      through tests.                                      i

   Disadvantages:

   1. Dissolved oxygen (DO) depletion may result from high chemical  oxygen
      demand (COD), biological oxygen demand (BOD), or metabolic wastes.
   2. Possible loss of toxicants through volatilization and/or adsorption to
      the exposure vessels.
   3. Generally less sensitive than static renewal, because the toxic
      substances may degrade or be adsorbed, thereby reducing the apparent
      toxicity.  Also, there is less chance of detecting slugs of toxic
      wastes, or other temporal variations in waste properties.

2.4.2  STATIC RENEWAL, SHORT-TERM TOXICITY TESTS:         |

   Advantages:

   1. Reduced possibility of DO depletion from high COD and/or BOD,  or ill
      effects from metabolic wastes from organisms in the test solutions.
   2. Reduced possibility of loss of toxicants through volatilization and/or
      adsorption to the exposure vessels.
   3. Test organisms that rapidly deplete energy reserves are fed when the
      test solutions are renewed, and are maintained in a healthier state.

   Disadvantages:                                         ,

   1. Require greater volume of effluent than non-renewal tests.
   2. Generally less chance of temporal variations in waste properties.

-------
                                   SECTION 3

                               HEALTH AND SAFETY
3.1  GENERAL PRECAUTIONS

3.1.1  Each laboratory should develop and maintain an effective health and
safety program, requiring an ongoing commitment by the laboratory management.
This program should include (1) a safety officer with the responsibility and
authority to develop and maintain a safety program, (2) the preparation of a
formal, written, health and safety plan, which is provided to each of the
laboratory staff, (3) an ongoing training program on laboratory safety, and
(4) regularly scheduled, documented, safety inspections.

3.1.2  Collection and use of effluents in toxicity tests may involve  ;
significant risks to personal safety and health.  Personnel collecting
effluent samples and conducting toxicity tests should take all safety
precautions necessary for the prevention of bodily injury and illness which
might result from ingestion or invasion of infectious agents, inhalation or
absorption of corrosive or toxic substances through skin contact, and
asphyxiation due to lack of oxygen or presence of noxious gases.      :

3.1.3  Prior to sample collection and laboratory work, personnel will
determine that all necessary safety equipment and materials have been ^obtained
and are in good condition.

3.1.4  Guidelines for the handling and disposal of hazardous materials must be
strictly followed.

3.2  SAFETY EQUIPMENT

3.2.1  PERSONAL SAFETY GEAR                                           •

3.2.1.1  Personnel should use safety equipment, as required, such as rubber
aprons, laboratory coats, respirators, gloves, safety glasses, hard hats, and
safety shoes.  Plastic netting on glass beakers, flasks, and other glassware
minimizes breakage and subsequent shattering of the glass.

3.2.2  LABORATORY SAFETY EQUIPMENT

3.2.2.1  Each laboratory (including mobile laboratories) should be provided
with safety equipment such as first aid kits, fire extinguishers, fire
blankets, emergency showers, chemical spill clean up kits, and eye fountains.

3.2.2.2  Mobile laboratories should be equipped with a telephone or other
means to enable personnel to summon help in case of emergency.

3.3  GENERAL LABORATORY AND FIELD OPERATIONS

3.3.1  Work with effluents should be performed in compliance with accepted
rules pertaining to the handling of hazardous materials (see safety manuals

                                       8

-------
 listed  in  Section 3, Health  and  Safety,  Subsection  3.5).   It  is  recommended
 that  personnel  collecting  samples  and  performing  toxicity  tests  not work
 alone.                                                     ;

 3.3.2   Because  the chemical  composition  of  effluents  is usually  only  poorly
 known,  they should be considered as potential  health  hazards,  and  exposure to
 them  should be  minimized.  Fume  and canopy  hoods  over the  toxicity test areas
 must  be used whenever possible.                            ;

 3.3.3   It  is advisable to  cleanse  exposed parts of  the body immediately after
 collecting effluent samples.                    '           ,  '

 3.3.4   All containers are  to be  adequately  labeled  to indicate their  contents.

 3.3.5   Staff should be familiar  with safety guidelines on  Material Safety Data
 Sheets  for reagents and other chemicals  purchased from suppliers.
 Incompatible materials should not  be stored together.  Good housekeeping
 contributes to  safety and  reliable results.

 3.3.6   Strong acids and volatile organic solvents employed in glassware
 cleaning must be used in a fume  hood or  under  an exhaust canopy  over  the work
 area.

 3.3.7   Electrical equipment or extension cords not  bearing the approval of
 Underwriter Laboratories must not  be used.  Ground-fault interrupters must be
 installed  in all "wet" laboratories,where electrical  equipment is  used.

 3.3.8   Mobile laboratories should be properly  grounded to  protect  against
 electrical shock.

 3.4  DISEASE PREVENTION

 3.4.1   Personnel handling samples which are known or  suspected to  contain
 human wastes should be immunized against tetanus, typhoid  fever, polio, and
 hepatitis B.

 3.5  SAFETY MANUALS                                        |

 3.5.1   For further guidance on safe practices when collecting effluent samples
 and conducting toxicity tests,  check with the permittee and consult general
 safety manuals,  including USEPA  (1986e) and Walters and Jameson  (1984).

 3.6  WASTE DISPOSAL

 3.6.1  Wastes generated during toxicity testing must be properly handled and
 disposed of in an appropriate manner.   Each testing facility will have its own
waste disposal  requirements based on local,  state, and Federal rules and
 regulations.  It is extremely important that these rules and regulations be
 known, understood,  and complied with by all  persons responsible for,  or
 otherwise involved in performing the toxicity testing activities.  Local  fire
officials should be notified of any potentially hazardous  conditions.

-------
                                   SECTION  4

                               QUALITY  ASSURANCE
4.1  INTRODUCTION
4.1.1  Development and maintenance of a toxicity test laboratory quality
assurance (QA) program (USEPA, 1991a) requires an ongoing commitment by
laboratory management.  Each toxicity test laboratory should (1) appoint a
quality assurance officer with the responsibility and authority to deyelop and
maintain a QA program; (2) prepare a quality assurance plan with stated data
quality objectives (DQOs); (3) prepare a written description of laboratory
standard operating procedures (SOPs) for culturing, toxicity-testing,
instrument calibration, sample chain-of-custody procedures, laboratory sample
tracking system, glassware cleaning, etc.; and (4) provide an adequate,
qualified technical staff for culturing and testing the organisms, and
suitable space and equipment to assure reliable data.                :

4.1.2  QA practices for toxicity testing laboratories must address all
activities that affect the quality of the final effluent toxicity test data,
such as:  (1) effluent sampling and handling;  (2) the source and condition of
the test organisms; (3) condition of equipment; (4) test conditions; (5)
instrument calibration; (6) replication; (7) use of reference toxicants;
(8) record keeping; and (9) data evaluation.                         !

4.1.3  Quality control practices, on the other hand, consist of the more
focused, routine, day-to-day activities carried out within the scope of the
overall QA program.   For more detailed discussion of quality assurance and
general guidance on good laboratory practices  and laboratory evaluation
related to toxicity testing, see FDA, (1978);  USEPA, (1979d), USEPA  (1980b),
USEPA  (1980c), and USEPA  (1991c); DeWoskin  (1984); and Taylor (1987).;

4.1.4  Guidance for the evaluation of laboratories performing toxicity tests
and laboratory evaluation criteria may be found in USEPA (1991c).

4.2  FACILITIES, EQUIPMENT, AND TEST CHAMBERS                        •

4.2.1  Separate test  organism culturing and toxicity testing areas shbuld be
provided to  avoid possible loss of cultures due to cross-contamination.
Ventilation  systems should be designed and  operated to prevent recirculation
or leakage of air from chemical analysis laboratories or sample storage and
preparation  areas into organism culturing or testing areas,  and from testing
and sample preparation areas  into culture rooms.

4.2.2  Laboratory and toxicity test  temperature control equipment must  be
adequate to  maintain  recommended test water temperatures.   Recommended
materials must  be used in the fabrication of the  test equipment which  comes  in
contact with the effluent  (see Section 5, Facilities, Equipment and  Supplies;
and specific toxicity test method).                                  ;
                                       10

-------
4.3  TEST ORGANISMS

4.3.1  The test organisms used in the procedures described in this manual are
the fathead minnow, PimephaTes promelas, the daphnid, Cen'odaphnia dubia, and
the green alga, Selenastrum capn'cornutum.  The fish and invertebrates should
appear healthy, behave normally, feed well, and have low mortality in the
cultures, during holding, and in test controls.  Test organisms should be
positively identified to species (see Section 6, Test Organisms).

4.4  LABORATORY WATER USED FOR CULTURING AND TEST DILUTION MATER

4.4.1  The quality of water used for test organism culturing and for dilution
water used in toxicity tests is extremely important.  Water for these two uses
should come from the same source.  The dilution water used in effluent
toxicity tests will depend in part on the objectives of the study and
logistical constraints, as discussed in detail in Section 7, Dilution Water.
For tests performed to meet MPDES objectives, synthetic, moderately hard water
should be used.  The dilution water used for internal quality assurance tests
with organisms, food, and reference toxicants should be the water routinely
used with success in the laboratory.  Types of water are discussed in
Section 5, Facilities, Equipment and Supplies.  Water used for culturing and
test dilution should be analyzed for toxic metals and organics at least
annually or whenever difficulty is encountered in meeting minimum
acceptability criteria for control  survival and reproduction or growth.  The
concentration of the metals Al, As, Cr, Co, Cu, Fe, Pb, Ni, and Zn, expressed
as total metal, should not exceed 1 /^g/L each, and Cd,  Hg,  and Ag, expressed
as total metal, should not exceed 100 ng/L each.  Total organochlorine
pesticides plus PCBs should be less than 50 ng/L (APHA, 1992).  Pesticide
concentrations should not exceed USEPA's Ambient Water Quality chronic
criteria values where available.

4.5  EFFLUENT AND RECEIVING MATER SAMPLING AND HANDLING

4.5.1  Sample holding times and temperatures of effluent samples collected for
on-site and off-site testing must conform to conditions described in
Section 8, Effluent and Receiving Water Sampling,  Sample Handling, and Sample
Preparation for Toxicity Tests.

4.6  TEST CONDITIONS                                     ',
                                                         i
4.6.1  Water temperature must be maintained within the  limits specified for
each test.  The temperature of test solutions must be measured by placing the
thermometer or probe directly into  the test solutions,  or by placing the
thermometer in equivalent volumes of water in surrogate vessels positioned at
appropriate locations among the test vessels.  Temperature  should be recorded
continuously in at least one test vessel for the duration of each test.'  Test
solution temperatures must be maintained within the limits  specified for each
test.  DO concentration and pH  should be checked at the beginning of each test
and daily throughout the test period.
                                      11

-------
4.7  QUALITY OF TEST ORGANISMS
                                                                      I
4.7.1  If the laboratory performs short-term chronic toxicity tests routinely
but does not have an ongoing test organism culturing program and must obtain
the test organisms from an outside source, the sensitivity of a batch of test
organisms must be determined with a reference toxicant in a short-term chronic
toxicity test performed monthly (see Subsections 4.14, 4.15, 4.16, and, 4.17).
Where acute or short-term chronic toxicity tests are performed with effluents
or receiving waters using test organisms obtained from outside the test
laboratory, concurrent toxicity tests of the same type must be performed with
a reference toxicant, unless the test organism supplier provides control chart
data from at least the last five monthly short-term chronic toxicity tests
using the same reference toxicity and control conditions (see Section 6, Test
Organism).

4.7.2  The supplier should certify the species identification of the test
organisms, and provide the taxonomic reference (citation and page) or !name(s)
of the taxonomic expert(s) consulted.                                 j

4.7.3  If the laboratory maintains breeding cultures, the sensitivity !of the
offspring should be determined in a short-term chronic toxicity test performed
with a reference toxicant at least once each month (see Subsections 4.|l4,
4.15, 4.16, and 4.17).  If preferred, this reference toxicant test may be
performed concurrently with an effluent toxicity test.  However, if a given
species of test organism produced by inhouse cultures is used only monthly, or
less frequently in toxicity tests, a reference toxicant test must be performed
concurrently with each short-term chronic effluent and/or receiving water
toxicity test.

4.7.4  If routine reference toxicant tests fail to meet acceptability
criteria, the test must be immediately repeated.  If the failed reference
toxicant test was being performed concurrently with an effluent or receiving
water toxicity test, both tests must be repeated (For exception, see  i
Section 4, Quality Assurance, Subsection 4.16.5).                     |

4.8  FOOD QUALITY

4.8.1  The nutritional quality of the food used in culturing and testing fish
and invertebrates is an important factor in the quality of the toxicity test
data.  This is especially true for the unsaturated fatty acid content of brine
shrimp nauplii, Artemia.  Problems with the nutritional suitability of the
food will be reflected in the survival, growth, and reproduction of the test
organisms in cultures and toxicity tests.  Artemia cysts, and other foods must
be obtained as described in Section 5, Facilities, Equipment, and Supplies.

4.8.2  Problems with the nutritional suitability of food will be reflected  in
the survival, growth, and reproduction of the test organisms in cultures and
toxicity tests.  If a batch of food is suspected to be defective, the
performance of organisms fed with the new food can be compared with the
performance of organisms fed with a food of known quality in side-by-side
tests.  If the food is used for culturing, its suitability should be
determined using a short-term chronic test which will determine the affect  of

                                      12

-------
food quality on growth or reproduction of each of the  relevant test  species  in
culture, using four replicates with each food source.  Where  applicable,  foods
used only in chronic toxicity tests can be compared with  a  food  of known
quality in side-by-side, multi-concentration chronic tests, using the
reference toxicant regularly employed in the laboratory QA  program.  For list
of commercial sources of Artemia cysts, see Table 2 of Section 5, Facilities,
Equipment, and Supplies.
                                                          s
4.8.3  New batches of food used in culturing and testing  s;hould  be analyzed
for toxic organics and metals or whenever difficulty is encountered  in meeting
minimum acceptability criteria for control survival and reproduction or
growth.  If the concentration of total organochlorine  pesticides exceeds  0.15
ng/g wet weight, or the concentration of total organochlorine pesticides  plus
PCBs exceeds 0.30 /j,g/g wet weight, or toxic metals (Al, As, Cr,  Cd,  Cu, Pb,
Ni, Zn, expressed as total metal) exceed 20 ng/g wet weight, the food should
not be used (for analytical methods see AOAC, 1990 and USDA,  1989).  For  foods
(e.g., such as YCT) which are used to culture and test organisms, the quality
of the food should meet the requirements for the laboratory water used for
culturing and test dilution water as described in Section 4.4 above.

4.9  ACCEPTABILITY OF SHORT-TERM CHRONIC TOXICITY TESTS

4.9.1  For the tests to be acceptable, control survival in fathead minnow,
Pimephales promeTas, and the daphnid, Cen'odaphm'a dubia, tests must be 80% or
greater.  At the end of the test, the average dry weight of surviving
seven-day-old fathead minnows in control chambers must equal or exceed
0.25 mg.  In Cen'odaphm'a dubia controls, at least 60% of the surviving adults
should have produced their third brood in 7 ฑ 1 days,  and the number of young
per surviving adult must be 15 or greater.  In algal  toxicity tests, the mean
cell density in the controls after 96 h must equal or exceed 2 x 10  cells/mL
and not vary more than 20% among replicates.  If these criteria are not met,
the test must be repeated.

4.9.2  An individual test may be conditionally acceptable if temperature,  DO,
and other specified conditions fall outside specifications, depending on the
degree of the departure and the objectives of the tests (see test condition
summaries).   The acceptability of the test would depend on the experience and
professional  judgment of the laboratory investigator and the reviewing staff
of the regulatory authority.   Any deviation from test specifications must be
noted when reporting data from the test.

4.10  ANALYTICAL METHODS

4.10.1  Routine chemical  and physical  analyses for culture and dilution water,
food,  and test solutions must include established quality assurance practices
outlined in  USEPA methods manuals (USEPA,  1979a and USEPA, 1979b).

4.10.2  Reagent containers should be dated and catalogued when received from
the supplier,  and the shelf life should not be exceeded.   Also,  working
solutions should be dated when prepared,  and the recommended shelf life should
be observed.
                                      13

-------
4.11  CALIBRATION AND STANDARDIZATION                                 ;

4.11.1  Instruments used for routine measurements of chemical and physical
parameters such as pH, DO, temperature, and conductivity, must be calibrated
and standardized according to instrument manufacturer's procedures asj
indicated in the general section on quality assurance (see USEPA Methods
150.1, 360.1, 170.1, and 120.1 in USEPA, 1979b).  Calibration data are
recorded in a permanent log book.                                     ;

4.11.2  Wet chemical methods used to measure hardness, alkalinity and total
residual chlorine must be standardized prior to use each day according to the
procedures for those specific USEPA methods (see USEPA Methods 130.2 and 310.1
in USEPA, 1979b).

4.12  REPLICATION AND TEST SENSITIVITY

4.12.1  The sensitivity of the tests will depend in part on the number of
replicates per concentration, the significance level selected, and the type of
statistical analysis.   If the variability remains constant, the sensitivity of
the test will increase  as the number of replicates is increased.  Thelminimum
recommended number of replicates varies with the objectives of the test and
the statistical method  used for analysis of the data.

4.13  VARIABILITY IN TOXICITY TEST RESULTS

4.13.1  Factors which can affect test  success and precision  include  (1) the
experience and skill of the laboratory  analyst;  (2) test organism age*
condition, and sensitivity; (3) dilution water quality;  (4) temperature
control; and  (5) the quality and quantity of food provided.  The results will
depend  upon the species used and the strain or source of the test organisms,
and test conditions, such as temperature, DO, food, and water quality.  The
repeatability or precision of toxicity  tests is  also a function of the number
of test organisms used  at each toxicant concentration.  Jensen  (1972)
discussed the relationship between sample size  (number of  fish) and  the
standard error of the test, and considered 20 fish per concentration  as
optimum for Probit Analysis.

4.14  TEST PRECISION

4.14.1  The ability of  the laboratory  personnel  to obtain  consistent, precise
results must  be demonstrated with reference toxicants before they attempt  to
measure effluent toxicity.  The  single-laboratory precision  of  each  type  of
test  to be used in  a  laboratory  should be determined by  performing at least
five  tests with a reference toxicant.

4.14.2  Test  precision  can be estimated by using the same  strain of  organisms
under the same test conditions  and employing a  known toxicant,  such  as  a
reference toxicant.

4.14.3   Inter!aboratory precision data from chronic toxicity tests with two
species using the reference toxicants  potassium  chloride  and copper  sulfate
are shown in  Table  1.   Additional precision data for each  of the tests

                                       14                              '

-------
described in this manual are presented in the sections describing the
individual test methods.
  TABLE 1. NATIONAL INTERLABORATORY STUDY OF CHRONIC TOXICITY TEST PRECISION,
           1991:  SUMMARY OF RESPONSES USING A REFERENCE TOXICANT1
  Organism
Endpoint
No. Labs   % Effluent'
SD
CV(%)
Pimephales
promelas


Cen'odaphnia
dubia


Survival, NOEC
Growth, IC25
Growth, IC50
Growth, NOEC
Survival, NOEC
Reproduction, IC25
Reproduction, IC50
Reproduction, NOEC
146
124
117
142
162
155
150
156
NA
4.67
6.36
NA
NA
2.69
3.99
NA
NA
1.87
2.04
NA
i NA
1.96
i 2.35
NA
NA
40.0
32.1
NA
NA
72.9
58.9
NA
     From a national study of interlaboratory precision of toxicity  test data
     performed  in  1991 by the  Environmental  Monitoring Systems  Laboratory-
     Cincinnati ,, U.S.  Environmental  Protection Agency,  Cincinnati,  OH  45268.
     Participants included Federal, state, and private laboratories engaged in
     NPDES permit compliance monitoring.                  ;
     Expressed  as % effluent; in  reality it was  a reference  toxicant  (KC1)
     but was not known by the persons  conducting the tests.
 4.14.4  Additional  information  on  toxicity  test  precision!is  provided  in  the
 Technical  Support Document  for  Water  Quality-based  Control  (see  pp.  2-4,  and
 11-15  in USEPA,  1991a).                                   ;

 4.14.5  In cases where the  test data  are  used  in Probit  Analysis or  other
 point  estimation techniques (see Section  9,  Chronic Toxicity  Test Endpoints
 and  Data Analysis),  precision can  be  described by the  mean, standard
 deviation,  and  relative  standard deviation  (percent coefficient  of variation,
 or CV)  of  the calculated endpoints from the replicated tests.   In cases where
 the  test data are used in the Linear  Interpolation  Method,  precision can  be
 estimated  by empirical confidence  intervals derived by using  the ICPIN Method
 (see Section 9,  Chronic  Toxicity Test Endpoints  and Data Analysis).   However,
 in cases where  the  results  are  reported in  terms of the  No-Observed-Effect
 Concentration  (NOEC)  and Lowest-Observed-Effect  Concentration (LOEC) (see
 Section 9, Chronic  Toxicity Test Endpoints  and Data Analysis)  precision can
 only be described by listing the NOEC-LOEC  interval for  each  test.  It is not
 possible to express precision  in terms of a commonly used statistic.  However,
 when all tests  of the same  toxicant yield the  same  NOEC-LOEC  interval, maximum
                                       15

-------
 precision  has  been attained.   The "true"  no  effect  concentration  could  fall
 anywhere within  the interval,  NOEC ฑ (NOEC minus  LOEC).

 4.14.6   It should be noted here that the  dilution factor  selected for;a test
 determines the width of the NOEC-LOEC interval  and  the  inherent maximum
 precision  of the test.   As the absolute value  of  the  dilution  factor
 decreases,  the width of the NOEC-LOEC interval  increases,  and  the inherent
 maximum precision of the test  decreases.  When  a  dilution  factor  of 0.3 is
 used, the  NOEC could be considered to have a relative variability as  high as
 ฑ 300%.  With  a  dilution factor of 0.5, the  NOEC  could  be  considered  to have  a
 relative variability of ฑ 100%.   As a result of the variability of different
 dilution factors,   USEPA recommends the use  of  the  dilution  factor of 0.5 or
 greater.   Other  factors which  can affect  test  precision include:   test
 organism age,  condition,  and sensitivity; temperature control; and feeding.

 4.15  DEMONSTRATING ACCEPTABLE LABORATORY PERFORMANCE

 4.15.1  It is  a  laboratory's responsibility  to  demonstrate its ability  to
 obtain  consistent,  precise results with reference toxicants  before it [performs
 toxicity tests with effluents  for permit  compliance purposes.  To meet  this
 requirement, the intralaboratory precision,  expressed as percent  coefficient
 of variation (CV%),  of  each type of test  to  be  used in the laboratory should
 be determined  by performing five or more  tests  with different  batches !of test
 organisms,  using the same reference toxicant, at  the  same  concentrations, with
 the same test  conditions  (i.e.,  the same  test duration, type of dilution
 water,  age  of  test  organisms,  feeding, etc.), and the same data analysis
 methods.  A reference toxicant concentration series (0.5 or  higher) should be
 selected that  will  consistently  provide partial mortalities  at two or more
 concentrations.

 4.16  DOCUMENTING ONGOING LABORATORY PERFORMANCE

 4.16.1  Satisfactory laboratory  performance  is  demonstrated  by performing at
 least one  acceptable test per  month with  a reference  toxicant for each|
 toxicity test  method commonly  used in  the laboratory.  For a given  test
 method, successive  tests  must  be performed with the same reference  toxicant,
 at the  same concentrations, in the same dilution  water, using the same  data
 analysis methods.   Precision may vary  with the  test species, reference
 toxicant, and  type  of test.

 4.16.2  A control chart  should be  prepared for  each combination of  reference
 toxicant, test species, test conditions,  and endpoints.  Toxicity endpoints
 from five or six tests  are  adequate  for establishing  the control   charts.
 Successive toxicity  endpoints  (NOECs,  IC25s,  LC50s,  etc.)   should  be plotted
 and examined to determine  if the  results  (X.) are  within prescribed limits
 (Figure 1).  The types  of control  charts  illustrated  (see  USEPA,   1979a) are
 used to evaluate the cum'ulative  trend  of results  from a series of samples.
 For endpoints  that  are  point estimates (LC50s and IC25s),   the cumulative
mean (X) and upper  and  lower control limits  (ฑ  2S)  are re-calculated with each
 successive test result.   Endpoints  from hypothesis  tests (NOEC, NOAEC);  from
each test are  plotted directly on  the  control chart.  The  control   limits would
consist of one concentration interval  above and below the  concentration

                                       16

-------
representing the central tendency.  After two years of data collection, or a
minimum of 20 data points, the control (cusum) chart should be maintained
using only the 20 most recent data points.

4.16.3  The outliers, which are values falling outside the upper and lower
control limits, and trends of increasing or decreasing sensitivity, are
readily identified.  In the case of endpoints that are point estimates (LCBOs
and IC25s), at the Pp 05 probability level, one in 20 tests would be expected
to fall outside of the control limits by chance alone.  If more than one out
of 20 reference toxicant tests fall outside the control limits, the effluent
toxicity tests conducted during the month in which the second reference
toxicant test failed are suspect, and should be considered as provisional and
subject to careful review.  Control limits for the NOECs will also be exceeded
occasionally, regardless of how well a laboratory performs.

4.16.4  If the toxicity value from a given test with a reference toxicant
falls well outside the expected range for the other test organisms when using
the standard dilution water and other test conditions, the sensitivity of the
organisms and the overall credibility of the test system are suspect.  In this
case, the test procedure should be examined for defects and should be repeated
with a different batch of test organisms.

4.16.5  Performance-should improve with experience, and the control limits for
endpoints that are point estimates should gradually narrow.  However, control
limits of ฑ 2S will be exceeded 5% of the time by chance alone, regardless of
how well a laboratory performs.  Highly proficient laboratories which develop
very narrow control limits may be unfairly penalized if a test result which
falls just outside the control limits is rejected de facto.  For this reason,
the width of the control limits should be considered by the permitting
authority in determining whether the outliers should be rejected.

4.17  REFERENCE TOXICANTS

4.17.1  Reference toxicants such as sodium chloride (NaC'l), potassium chloride
(KC1), cadmium chloride  (CdCU),  copper sulfate (CuSOJ,  sodium dodecyl
sulfate (SDS), and potassium dichromate (K2Cr207), are  suitable for use in the
NPDES Program and other Agency programs requiring aquatic toxicity tests.
EMSL-Cincinnati hopes to -release USEPA-certified solutions of cadmium and
copper for use as reference toxicants through cooperative research and
development agreements with commercial suppliers, and will, continue to develop
additional reference toxicants for future release.  Interested parties can
determine the availability of "EPA Certified" reference toxicants by checking
the EMSL-Cincinnati electronic bulletin board, using a modem to access the
following telephone number:  513-569-7610.  Standard reference materials also
can be obtained from commercial supply houses, or can be prepared inhouse
using reagent grade chemicals.  The regulatory agency should be consulted
before reference toxicant(s) are selected and used.
                                       17

-------
Where:
 a
 o
           UPPER CONTROL LIMIT
                       CENTRAL TENDENCY
           LOWER CONTROL LIMIT
  0
                 10
                                    15
                                 20
O
O
O"
        UPPER CONTROL LIMIT (X + 2 S)
                      CENTRAL TENDENCY
       LOWER^ONTROL LIMIT (X - 2 S)
  0      5       10      15      20 ""
 TOXICITYTEST WITH REFERENCE TOXICANTS
             X =
                    n
     S =
         \
               Ey- 2  _   ji=l
               ^    	
                        n
                   n-l
         X
         S
  = Successive toxicity values from toxicity tests.
  = Number of tests.                             •
  = Mean toxicity value.
  - Standard deviation.
                18

-------
4.18  RECORD KEEPING                                      :

4.18.1  Proper record keeping is important.  A complete file should be
maintained for each individual toxicity test or group of tests on closely
related samples.  This file should contain a record of the sample chain-of-
custody; a copy of the sample log sheet; the original bench sheets for the
test organism responses during the toxicity test(s); chemical analysis data on
the sample(s); detailed records of the test organisms used in the test(s),
such as species, source, age, date of receipt, and other pertinent information
relating to their history and health; information on the calibration of
equipment and instruments; test conditions employed; and results of reference
toxicant tests.  Laboratory data should be recorded on a real-time basis to
prevent the loss of information or inadvertent introduction of errors into the
record.  Original data sheets should be signed and dated by the laboratory
personnel performing the tests.                           :

4.18.2  The regulatory authority should retain records pertaining to discharge
permits.  Permittees are required to retain records pertaining to permit
applications and compliance for a minimum of 3 years [40 CFR 122.41(j)(2)].

4.19  VIDEO TAPES OF USEPA CULTURE AND TOXICITY TEST METHODS

4.19.1  Three video-based training packages are available :from the National
Technical Information Service i(NTIS), Department of Commerce, 5285 Port Royal
Road, Springfield, VA  22161.  Credit card orders can be placed by calling
toll-free (800) 788-6282, or by FAX at 703-321-8547, or by mail at the above
address.  .For other information call 703-487-4650.

    1. Order # ELA18254: "U.S. EPA Freshwater Culturing Methods for
      Ceriodaphnia dubia and the Fathead Minnow (Pimephales promelas),"
      consisting of a 24-minute video and 33-page supplemental report on
      culturing Ceriodaphnia, and an 18 minute video and 22-page report on
      culturing fathead minnows, and a copy of Short-term Methods for
      Estimating the chronic Toxicity of Effluents and Receiving Waters to
      Freshwater Organisms  (EPA-600/4-89/001).  Price $60..00..

    2. Order # ELA18036: "U.S. EPA Test Methods for Freshwater Toxicity
      Tests," consisting of a 23-minute video and 26-page ,supplemental report
      on Ceriodaphnia survival and reproduction toxicity tests, and a 15-
      minute video and 18-page report on fathead minnow survival and growth
      toxicity  tests, and a copy of Short-term Methods for Estimating the
      chronic Toxicity of Effluents and Receiving Waters to Freshwater
      Organisms  (EPA-600/4-89/001).  Price $45.00.        ;

    3. Order # ELA18301: U.S.  EPA Culturing and Test Methods for Freshwater
      Effluent  Toxicity Tests using Ceriodaphnia dubia and Fathead Minnows
       (Pimephales promelas), consisting of all four videos and supplemental
      reports,  and a copy of the short-term toxicity test manual.  Price
      $90.00.
                                       19

-------
4.20  SUPPLEMENTAL REPORTS FOR TRAINING VIDEO TAPES

4.20.1  Supplemental Reports for Training Video Tapes are included in draining
packages above.
                                      20

-------
                                 SECTION 5

                     FACILITIES, EQUIPMENT, AND SUPPLIES


5.1  GENERAL REQUIREMENTS
                                                         1
5.1.1  Effluent toxicity tests may be performed in a fixed or mobile
laboratory.  Facilities must include equipment for rearing and/or holding
organisms.  Culturing facilities for test organisms may be desirable in fixed
laboratories which perform large numbers of tests.  Temperature control can be
achieved using circulating water baths, heat exchangers, 'or environmental
chambers.  Water used for rearing, holding, acclimating, and testing organisms
may be ground water, receiving water, dechlorinated tap water, or
reconstituted synthetic water.  Dechlorination can be accomplished by carbon
filtration, or the use of sodium thiosulfate.  Use of 3.6 mg (anhydrous)
sodium thiosulfate/L will reduce 1.0 mg chlorine/L.  After dechlorination,
total residual chlorine should be non-detectable.  Air used for aeration must
be free of oil and toxic vapors.  Oil-free air pumps should be used where
possible.  Particulates can be removed from the air using BALSTONฎ Grade BX or
equivalent filters (Balston, Inc., Lexington, Massachusetts), and oil and
other organic vapors can be removed using activated carbon filters (BALSTONฎ,
C-l filter, or equivalent).

5.1.2  The facilities must be well ventilated and free from fumes.  Laboratory
ventilation systems should be checked to ensure that return air from chemistry
laboratories and/or sample holding areas is not circulated to test organism
culture rooms or toxicity test rooms, or that air from toxicity test rooms
does not contaminate culture areas.  Sample preparation, culturing, and
toxicity test areas should be separated to avoid cross cointamination of
cultures or toxicity test solutions with toxic fumes.  Air pressure
differentials between such rooms should not result in a net flow of
potentially contaminated air to sensitive areas through open or loosely-
fitting doors.  Organisms should be shielded from external  disturbances.

5.1.3  Materials used for exposure chambers, tubing, etc.,  that come in
contact with the effluent and dilution water should be carefully chpsen.
Tempered glass and perfluorocarbon plastics (TEFLONฎ) should be used whenever
possible to minimize sorption and leaching of toxic substances.  These
materials may be reused following decontamination.  Containers made of
plastics, such as polyethylene, polypropylene, polyvinyl chloride, TYGONฎ,
etc., may be used as test chambers or to ship, store and transfer effluents
and receiving waters, but they should not be reused unless absolutely
necessary, because they could carry over adsorbed toxicants from one test to
another, if reused.  However, these containers may be repeatedly reused for
storing uncontaminated waters, such as deionized or laboratory-prepared
dilution waters and receiving waters.  Glass or disposable polystyrene
containers can be used for test chambers.  The use of large (> 20 L) glass
carboys is discouraged for safety reasons.

5.1.4  New plastic products of a type not previously used should be tested for
toxicity before initial use by exposing the test organisms in the test system

                                      21

-------
where the material  is  used.   Equipment  (pumps, valves, etc.) which  cannot be
discarded after each use  because of cost, must be decontaminated  according to
the cleaning procedures listed below  (see Section 5,  Facilities,  Equipment and
Supplies, Subsection 5.3.2).  Fiberglass and  stainless steel,  in  addition to
the previously mentioned  materials, can be  used for holding, acclimating, and
dilution water storage tanks, and  in  the water delivery system, but once
contaminated with pollutants  the fiberglass should not be reused.   All
material should be  flushed or rinsed  thoroughly with  the test  media before
using in the test.

5.1.5  Copper, galvanized material, rubber, brass, and lead must  not come in
contact with culturing, holding, acclimation, or dilution water,  or wi|th
effluent samples and test solutions.  Some materials, such as  several types of
neoprene rubber (commonly used for stoppers), may be  toxic and should be
tested before use.

5.1.6  Silicone adhesive  used to construct glass test chambers absorbs some
organochlorine and  organophosphorus pesticides, which are difficult to remove.
Therefore, as little of the adhesive  as possible should be in  contact with
water.  Extra beads of adhesive inside the containers should be removed.

5.2  TEST CHAMBERS

5.2.1  Test chamber size  and  shape are varied according to size of  the test
organism.  Requirements are specified in each toxicity test method.   ;

5.3  CLEANING TEST CHAMBERS AND LABORATORY APPARATUS

5.3.1  New plasticware used for sample collection or  organism  exposure vessels
does not require thorough cleaning before use.  It is sufficient  to ri;nse new
sample containers once with dilution water before use.  New glassware ;must be
soaked overnight in 10% acid  (see below) and  rinsed well in deionized water
and dilution water.

5.3.2  All non-disposable sample containers,  test vessels, tanks, and [other
equipment that have come  in contact with effluent must be washed  after use to
remove contaminants as described below.                               ;

   1. Soak 15 min in tap  water and scrub with detergent, or clean in an
      automatic dishwasher.
   2. Rinse twice with tap water.                                     ;
   3. Carefully rinse  once with fresh, dilute (10%, V:V) hydrochloric jor
      nitric acid to remove scale, metals and bases.  To prepare  a  10%
      solution of acid, add 10 ml of concentrated acid to 90 mL of deiionized
      water.                                                          ;
   4. Rinse twice with deionized water.
   5. Rinse once with  full-strength, pesticide-grade  acetone to remove
      organic compounds (use  a fume hood or canopy).
   6. Rinse three times with  deionized water.

5.3.3  Special  requirements for cleaning glassware used in the green alga,
Selenastrum capricornutum, toxicity tests (Method 1003.0 Section  14).  Prepare

                                      22

-------
2,
3,
4,
5.
 all  graduated cylinders,  test flasks,  bottles,  volumetric flasks,  centrifuge
 tubes and vials used in algal assays as follows:

    1. Wash with non-phosphate detergent solution,  preferably heated to > 50ฐC
       Brush the inside of flasks with  a stiff-bristle brush to loosen any
       attached material.   The use of a commercial  laboratory glassware washer
       or heavy-duty kitchen  dishwasher (under-counter type) is highly
       recommended.
       Rinse with tap water.
       Test flasks should  be  thoroughly rinsed with acetone and a 10% solution
       (by volume) of reagent grade hydrochloric acid  (HC1).  It may be
       advantageous  to soak the flasks  in 10% HC1  for  several  days.   Fill  vials
       and centrifuge tubes with the 10% HC1  solution  and  allow to  stand a few
       minutes;  fill  all larger containers to about one-tenth  capacity with HC1
       solution  and  swirl  so  that the entire  surface is  bathed
       Rinse twice with MILLIPOREฎ MILLI-Qฎ OR QPAK™2, or  equivalent,  water.
       New test  flasks,  and all  flasks  which  through use may become
       contaminated  with toxic organic  substances,  must  be Tinsed with
       pesticide-grade acetone or heat-treated before  use.   To thermally
       degrade organics, place glassware in a high  temperature oven  at 400ฐC
       for 30  min.   After  cooling,  go to 7.   If  acetone  is used,  go  to 6.
   6.  Rinse thoroughly with  MILLIPOREฎ MILLI-Qฎ or QPAK™,,  or equivalent
       water,  and  dry in an 105ฐC oven.   All  glassware should  be  autoclaved
       before  use  and between  uses.                        ;
   7.  Cover the  mouth  of  each chamber  with aluminum foil  or other closure,  as
       appropriate,  before storing.                        :

 5.3.4 The  use  of sterile, disposable  pipets  will  eliminate the  need  for  pipet
 washing  and minimize the  possibility of contaminating the cultures  with toxic
 substances.
                                                          i

 5.3.5  All  test chambers  and  equipment  must  be  thoroughly rinsed with  the
 dilution  water  immediately prior  to  use  in each test.

 5.4  APPARATUS AND EQUIPMENT  FOR  CULTURIN6 AND TOXICITY TESTS

 5.4.1  Apparatus  and equipment  requirements for culturing and testing  are
 specified in  each toxicity test method.  Also,  see  USEPA,  1993b.

 5.4.2  WATER  PURIFICATION  SYSTEM                          \
                                                          i
 5.4.2.1  A good quality deionized water, providing  18  mega-ohm, laboratory
grade water,  should be available  in the laboratory  and in  sufficient capacity
 tor laboratory needs.  Deionized water may be obtained from MILLIPOREฎ Milli-
Qฎ, MILLIPOREฎ QPAK™  or equivalent system.  If large  quantities of high
quality deionized water are needed,  it may be advisable to supply the
 laboratory grade deionizer with preconditioned water from  a Culliqanฎ
Continentalฎ, or equivalent.
                                  23

-------
                                     MO
5.5  REAGENTS AND CONSUMABLE MATERIALS

5.5.1  SOURCES OF FOOD FOR CULTURE AND TOXICITY TESTS

   1. Brine shrimp, Artemia sp., cysts -- A list of commercial sources is
      listed in Table 2.
   2. Frozen adult brine shrimp, Artemia -- Available from most pet supply
      shops or from San Francisco Bay Brand, 8239 Enterprise Dr., Newark, CA
      94560  (415-792-7200).                                          ,
   3. Flake fish food -- TETRAMINฎ and BIORILฎ are available from most pet

   4  Trout'chow -- Available from Zeigler Bros., P.O. Box 95, Gardners, PA
      17324 (717-677-6181 or 800-841-6800); Glencoe Mills, 1011 Elliott St.,
      Glencoe, MN 55336 (612-864-3181); or Murray Elevators, 118 West,4800
      South, Murray, UT 84107 (800-521-9092).                         :
   5. CEROPHYLLฎ -- Available from Ward's Natural Science Establishment,
      Inc., P.O. Box 92912, Rochester, NY 14692-9012  (716-359-2502) or as
      cereal leaves from Sigma  Chemical Company, P.O. Box 14508, St. Louis,
      63178 (800-325-3010).
   6. Yeast -- Packaged dry yeast, such as Fleischmann's, or equivalent, can
      be purchased at the local grocery store or is available  from  Lake States
      Yeast, Rhine!and, WI.                                       .
   7. Alfalfa Rabbit Pellets -- Available from feed stores as  Purina rabbit
      chow                                                            '
   8. Algae - Available from (1) the American Type Culture Collection; 12301
      Parklawn Drive, Rockville, MD 10852; or  (2) the Culture  Collection of
      Algae, Botany Department, University of Texas,  Austin, TX  78712.

 5.5.1.1  All food  should be tested for nutritional suitability and  chemically
 analyzed for organochlorine pesticides,  PCBs,  and toxic  metals (see Section 4,
 Quality Assurance).

 5.5.2  Reagents  and consumable  materials  are  specified  in each toxicity  test
 method section.  Also,  see  Section 4,  Quality  Assurance.

 5.6  TEST  ORGANISMS                                                   '

 5.6.1  Test  organisms  should be obtained  from inhouse cultures or from
 commercial  suppliers  (see  specific test  method;  Section  4,  Quality Assurance;
 and  Section  6, Test Organisms).                                      '
                                                                      I
 5.7   SUPPLIES                                                         ;

 5.7.1  See test  methods (see  Sections  11-14)  for specific supplies.
24

-------
    TABLE 2.  COMMERCIAL SUPPLIERS OF BRINE SHRIMP  (ARTEMIA) CYSTS1'2
Aquafauna Biomarine
P.O. Box 5
Hawthorne, CA 90250
Tel. (310) 973-5275
Fax. (310) 676-9387
(Great Salt Lake North Arm,
San Francisco Bay)             i

Argent Chemical
8702 152nd Ave. NE
Redmond, WA  98052
Tel. (800) 426-6258
Tel. (206) 855-3777
Fax. (206) 885-2112
(Platinum Label - San Francisco
Bay; Gold Label - San Francisco
Bay, Brazil;  Silver Label -
Great Salt Lake, Australia;
Bronze Label - China, Canada,
other)

Bonneville Artemia Intl., Inc.
P.O. Box 511113
Salt Lake City, UT  84151-1113
Tel. (801) 972-4704
Fax. (801) 972-7495

Ocean Star International
P.O. Box 643
Snowville, UT 84336
Tel. (801) 872-8217
Fax.  (801) 872-8272
(Great Salt Lake)
Sanders Brine Shrimp Co.
3850 South 540 West
Ogden, UT 84405
Tel. (801) 393-5027
(Great Salt Lake)
Sea Critters Inc.
P.O. Box 1508
Tavernier, FL 33070
Tel. (305) 367-2672
Aquarium Products   !
180L Penrod Court
Glen Burnie, MD 21061
Tel. (800) 368-2507
Tel. (301) 761-2100
Fax.   (410) 761-6458
(Columbia)

INVE Artemia Systems
Oeverstraat 7
B-9200 Baasrode, Belgium
Tel. 011-32-52-331320
Fax. 011-32-52-341205
(For marine species - AF  grade)
[small nauplii], UL grade [large
nauplii], for freshwater species-
Hi grade [small nauplii], EG grade
[large nauplii]
Golden West Artemia
411 East 100 South
Salt Lake City, UT 84111
Tel. (801) 975-1222
Fax. (801) 975-1444:

San Francisco Bay Brand
8239 Enterprise Drive
Newark, CA 94560    ;
Tel. (510) 792-7200
Fax. (510) 792-5360
(Great Salt Lake,
San Francisco Bay)

Western Brine Shrimp
957 West South Temple
Salt Lake City, UT 84104
Tel. (801) 364-3642
Fax. (801) 534-0211:
(Great Salt Lake)
2 List from David A.  Bengston,  University of Rhode Island,  Narragansett,  RI
  The  geographic sources from which the vendors obtain the  brine shrimp
  cysts are shown in parentheses.                          i


                                      25

-------
                                  SECTION 6

                                TEST ORGANISMS
6.1  TEST SPECIES                                                    ;

6.1.1  The species used in characterizing the chronic toxicity of effluents
and/or receiving waters will depend on the requirements of the regulatory
authority and the objectives of the test.  It is essential that good quality
test organisms be readily available throughout the year from inhouse or
commercial sources to meet NPDES monitoring requirements.  The organisms used
in the toxicity tests must be identified to species.  If there is any doubt as
to the identity of the test organism, representative specimens should be sent
to a taxonomic expert to confirm the identification.

6.1.2  Toxicity test conditions and culture methods for the species listed in
Subsection 6.1.3 are provided in this manual also, see USEPA, 1993b. ;

6.1.3  The organisms used in the short-term chronic toxicity tests described
in this manual are the fathead minnow, Pimephales promeTas, the daphnid,
Ceriodaphnia dubia (Berner, 1986), and the green alga, Selenastrum
capricornutum.

6.1.4  Some states have developed culturing and testing methods for indigenous
species that may be as sensitive, or more sensitive, than the species
recommended in Subsection 6.1.3.  However, USEPA allows the use of indigenous
species only where state regulations require their use or prohibit importation
of the recommended species  in Subsection 6.2.6.  Where state regulations
prohibit  importation of non-native fishes or the use of recommended test
species,  permission must be requested from the appropriate state agency prior
to their  use.

6.1.5  Where states have developed culturing and testing methods, for
indigenous species other than those recommended in this manual, data comparing
the  sensitivity of the substitute species and the one or more recommended
species must be obtained in side-by-side toxicity tests with reference
toxicants and/or effluents, to ensure that the species selected are at least
as sensitive as the recommended species.  These data must be submitted to the
permitting authority (State or Region) if required.  USEPA acknowledges that
reference toxicants prepared from pure chemicals may not  always be   j
representative of effluents.  However, because of the observed and/or
potential variability  in the quality and toxicity of effluents, it is not
possible  to specify a  representative effluent.

6.1.6  Guidance for the selection of test organisms where the salinity of the
effluent  and/or receiving water requires special consideration is provided  in
the  Technical Support  Document for Water Quality-based Toxics Control  (USEPA,
1991a).

   1. Where the salinity of the receiving water is <  l%o, freshwater lorganisms
      are used regardless of the  salinity of the effluent.

                                      26

-------
   2. Where the salinity of the receiving water is > l%o, :the choice of
      organisms depends on state water quality standards and/or permit
      requirements.                                       ,

6.2  SOURCES OF TEST ORGANISMS

6.2.1  The test organisms recommended in this manual can be cultured in the
laboratory using culturing and handling methods for each organism described in
the respective test method sections.  The fathead minnow, Pimephales promelas,
culture method is given in Section 11 and not repeated in Section 12.  Also,
see USEPA (1993b).                                        ;'

6.2.2  Inhouse cultures should be established wherever it is cost effective.
If inhouse cultures cannot be maintained or it is not. cost effective, test
organisms or starter cultures should be purchased from experienced commercial
suppliers (see USEPA, 1993b).

6.2.3  Starter cultures of the green algae, Selenastrum cupricornutum,
5. minutum, and Chlamydomonas reinhardti are available from the following
sources:

   1. American Type Culture Collection (Culture No. ATCC 22662), 12301
      Parklawn Drive, Rockville, MD 10852.
   2. Culture Collection of Algae, Botany Department, University of Texas,
      Austin, TX 78712.

6.2.4  Because the daphnid, Cen'odaphnia dubia, must be cultured individually
in the laboratory for at least seven days before the test begins, it will be
necessary to obtain a starter culture from a commercial source at least three
weeks before the test is to begin if they are not being cultured inhouse.

6.2.5  If, because of their source, there is any uncertainty concerning the
identity of the organisms, it is advisable to have them examined by a
taxonomic specialist to confirm their identification.  For detailed guidance
on identification, see the individual test methods.

6:2.6  FERAL (NATURAL OCCURRING, WILD CAUGHT) ORGANISMS

6.2.6.1  The use of test organisms taken from the receiving water has strong
appeal, and would seem to be a logical approach.  However, it is generally
impractical and not recommended for the following reasons:

   1. Sensitive organisms may not be present in the receiving water because of
      previous exposure to the effluent or other pollutants.
   2. It is often difficult to collect organisms of the required age and
      quality from the receiving water.
   3. Most states require collecting permits, which may be difficult to
      obtain.  Therefore, it is usually more cost effective to culture the
      organisms in the laboratory or obtain them from private, state, or
      Federal sources.  The fathead minnow, Pimephales promelas, the daphnid,
      Cen'odaphnia dubia, and the green alga, Selenastrum capricornutum, are
      easily cultured in the laboratory or readily available commercially.

                                      27

-------
   4. The  required  QA/QC  records,  such  as  the  single laboratory precision
      data, would not  be  available.                                   ;
   5. Since it  is mandatory  that the  identity  of the test organism be!
      known to  species  level,  it would  be  necessary to examine each organism
      caught  in the wild  to  confirm its  identity.  This would usually be
      impractical or,  at  the least, very stressful to the organisms.
   6. Test organisms obtained  from the wild must be observed in the laboratory
      for  a minimum of  one week prior to use,  to assure that they arelfree of
      signs of  parasitic  or  bacterial infections and other adverse effects.
      Fish captured by  electroshocking must not be used in toxicity testing.

6.2.6.2  Guidelines for collecting natural occurring organisms are provided in
USEPA (1973), USEPA (1990) and USEPA  (1993c).                         '

6.2.7  Regardless of their source, test  organisms should be carefully'observed
to ensure  that  they are free of signs of stress and disease, and in good
physical condition.  Some species  of test  organisms can be obtained from
commercial stock certified as  "disease-free".                         •

6.3  LIFE STAGE

6.3.1  Young  organisms  are often more sensitive to toxicants than are|adults.
For this reason, the use  of  early  life stages, such as larval fish, is
required for  all tests.   There may be special  cases, however, where the
limited availability of organisms  will require some deviation from the
recommended life stage.   In  a given test,  all ^organisms should be     ;
approximately the same  age and should be taken from the same source.  (Since
age may affect the  results of the  tests, it would enhance the value and
comparability of the data if the same species  in the same life stages iwere
used throughout a monitoring program at  a  given facility.             •

6.4  LABORATORY CULTURIN6
                                                                      I
6.4.1  Instructions for culturing  and/or holding the recommended test |
organisms are included  in the respective test methods (also, see USEPA,
1993b).

6.5  HOLDING AND HANDLING TEST ORGANISMS

6.5.1  Test organisms should not be subjected to changes of more than |3ฐC in
water temperature in any  12  h period or  2  units of pH in any 24-h period.

6.5.2  Organisms should be handled as little as possible.  When handling is
necessary, it should be done as gently,   carefully, and quickly as possible to
minimize stress.  Organisms  that are dropped or touch a dry surface or are
injured during handling must be discarded.  Dipnets are best for handling
larger organisms.  These nets are  commercially available or can be made from
small-mesh nylon netting, silk batting cloth, plankton netting, or similar
material.   Wide-bore,  smooth glass tubes (4 to 8 mm ID) with rubber bulbs or
pipettors (such as  PROPIPETTEฎ) should be  used for transferring smaller
organisms such as larval fish.                                         I


                                      28

-------
6.5.3  Holding tanks for fish are supplied with good quality water (see
Section 5, Facilities, Equipment, and Supplies) with flow-through rate of at
least two tank volumes per day.  Otherwise use a recirculation system where
water flows through an activated carbon or undergravel filter to remove
dissolved metabolites.  Culture water can also be piped through high intensity
ultraviolet light sources for disinfection, and to photodegrade dissolved
organics.

6.5.4  Crowding must be avoided because it will stress the organisms and lower
the DO concentrations to unacceptable levels.  The solution of oxygen depends
on temperature and altitude.  The DO must be maintained at a minimum of 4.0
mg/L.  Aerate gently if necessary.

6.5.5  The organisms should be observed carefully each day for signs of
disease, stress, physical damage, or mortality.  Dead and abnormal organisms
should be removed as soon as observed.  It is not uncommon for some fish
mortality (5-10%) to occur during the first 48 h in a holding tank because of
individuals that refuse to feed on artificial food and die of starvation.
Organisms in the holding tanks should generally be fed as in the cultures (see
culturing methods in the respective methods).

6.5.6  Fish should be fed as much as they will eat at least once a day with
live brine shrimp nauplii, Artemia, or frozen adult brine shrimp, or dry food
(frozen food should be completely thawed before use).  Adult brine shrimp can
be supplemented with commercially prepared food such as TETRAMINฎ.or BIORILฎ
flake food, or equivalent.  Excess food and fecal material should be removed
from the bottom of the tanks at least twice a week by siphoning.

6.5.7  Fish should be observed carefully each day for signs of disease,
stress, physical damage, and mortality.  Dead and abnormal specimens should be
removed as soon as observed.  It is not uncommon to have some fish (5-10%)
mortality during the first 48 h in a holding tank because of individuals that
refuse to feed on artificial food and die of starvation. :Fish in the holding
tanks should generally be fed as in the cultures (see culturing methods in the
respective methods).

6.5.8  A daily record of feeding, behavioral observations^ and mortality
should be maintained.

6.6  TRANSPORTATION TO THE TEST SITE                     ;

6.6.1  Organisms are transported from the base or supply laboratory to a
remote test site in culture water or standard dilution water in plastic bags
or large-mouth screw-cap (500 ml) plastic bottles in styrofoam coolers.
Adequate DO is maintained by replacing the air above the water in the bags
with oxygen from a compressed gas cylinder, and sealing tine bags or by use of
an airstone supplied by a portable pump,  the DO concentration must not fall
below 4.0 mg/L.

6.6.2  Upon arrival at the test site, the organisms are transferred to
receiving water if receiving water is to be used as the test dilution water.
All but a small volume of the holding water (approximately 5%) is removed by

                                      29

-------
siphoning and replaced slowly over a 10 to 15 minute period with dilution
water.  If receiving water is to be used as the dilution water, caution must
be exercised in exposing the test organisms to it, because of the possibility
that it might be toxic.  For this reason, it is recommended that only
approximately 10% of the test organisms be exposed initially to the dilution
water.  If this group does not show excessive mortality or obvious signs of
stress in a few hours, the remainder of the test organisms may be transferred
to the dilution water.
                                                            *
6.6.3  A group of organisms must not be used for a test if they appear:to be
unhealthy, discolored, or otherwise stressed, or if mortality appears to
exceed 10% preceding the test.  If the organisms fail to meet these criteria,
the entire group must be discarded and a new group obtained.  The mortality
may be due to the presence of toxicity, if the receiving water is used; as
dilution water, rather than a diseased condition of the test organisms.  If
the acclimation process is repeated with a new group of test organisms and
excessive mortality occurs, it is recommended that an alternative source of
dilution water be used.                                               ;

6.7  TEST ORGANISM DISPOSAL

6.7.1  When the toxicity test(s) is concluded, all test organisms (including
controls) should be humanely destroyed and disposed of in an appropriate
manner.
                                      30

-------
                                  SECTION 7

                                DILUTION WATER           !


7.1  TYPES OF DILUTION WATER

7.1.1  The type of dilution water used in effluent toxicity tests will  depend
largely on the objectives of the study.

7.1.1.1  If the objective of the test is to estimate the chronic toxicity of
the effluent, which is the primary objective of NPDES permit-related toxicity
testing, a synthetic (standard) dilution water (moderately hard water)  is
used.  If the test organisms have been cultured in water which is different
from the test dilution water, a second set of controls, using culture water,
should be included in the test.                 '         ;

7.1.1.2  If the objective of the test is to estimate the chronic toxicity of
the effluent in uncontaminated receiving water, the test may be conducted
using dilution water consisting of a single grab sample of receiving water (if
non-toxic), collected either upstream and outside the influence of the
outfall, or with other uncontaminated natural water (ground or surface water)
or standard dilution water having approximately the same characteristics
(hardness, alkalinity, and conductivity) as the receiving^water.  Seasonal
variations in the quality of receiving waters may affect effluent toxicity.
Therefore, the pH, alkalinity, hardness, and conductivity1of receiving water
samples should be determined before each use.  If the test organisms have been
cultured in water which is different from the test dilution water, a second
set of controls, using culture water, should be included in the test.

7.1.1.3. If the objective of the test is to determine the1additive or
mitigating effects of the discharge on already contaminated receiving water,
the test is performed using dilution water consisting of receiving water
collected immediately upstream or outside the influence of the outfall.
A second set of controls, using culture water, should be included in the test.

7.2  STANDARD, SYNTHETIC DILUTION WATER

7.2.1  Standard, synthetic dilution water is prepared with deionized water and
reagent grade chemicals or mineral water (Tables 3 and 4).  The source water
for the deionizer can be ground water or tap water.

7.2.2  DEIONIZED WATER USED TO PREPARE STANDARD, SYNTHETIC, DILUTION WATER

7.2.2.1  Deionized water is obtained from a MILLIPOREฎ MlLLI-Qฎ, MILLIPOREฎ
QPAK™2 or equivalent system.   It is advisable to provide a preconditioned
(deionized)  feed water by using a Culliganฎ, Continentalฎ;, or equivalent
system  in front of the MILLIPOREฎ System to extend the life of the MILLIPOREฎ
cartridges  (see Section 5, Facilities, Equipment, and Supplies).

7.2.2.2  The recommended order of the cartridges in a four-cartridge deionizer
(i.e., MILLI-Qฎ System or equivalent)  is (1) ion exchange, (2) ion exchange,

                                      31

-------
(3) carbon, and (4) organic cleanup (such as ORGANEX-Qฎ, or equivalent)
followed by a final bacteria filter.  The QPAK™2 water system is a sealed
system which does not allow for the rearranging of the cartridges.  However,
the final cartridge is an ORGANEX-Qฎ filter, followed by a final bacteria
filter.  Commercial laboratories using this system have not experience^1 any
difficulty in using the water for culturing or testing.  Reference to jthe
MILLI-Qฎ systems throughout the remainder of the manual includes all
HILLIPOREฎ or equivalent systems.

7.2.3  STANDARD, SYNTHETIC FRESHWATER                                 j

7.2.3.1  To prepare 20 L of synthetic, moderately hard, reconstituted water,
use the reagent grade chemicals in Table 3 as follows:

   1. Place 19 L of MILLI-Qฎ, or equivalent, water in a properly cleaned
      plastic carboy.
   2. Add 1.20 g of MgS04,  1.92 g NaHC03,  and  O.OSOg  KC1 to the  carboy.
   3. Aerate overnight.
   4. Add 1.20 g of CaS04ซ2H20  to  1  L  of  MILLI-Qฎ  or  equivalent  deionized
      water in a separate flask.  Stir on magnetic stirrer until calcium
      sulfate is dissolved, add to the 19 L above, and mix well.      '
   5. For Cen'odaphnia dubia culturing and testing, add sufficient sodium
      selenate (Na2Se04)  to provide  2  /ig  selenium  per liter of final dilution
      water.                                                          '•
   6. Aerate the combined solution vigorously for an additional  24 h tb
      dissolve the added chemicals and stabilize the medium.
   7. The measured pH, hardness, etc., should be as listed in Table 3.

7.2.3.2  If large volumes of synthetic reconstituted water will  be needed, it
may be advisable to mix 1 L portions of concentrated stock solutions of
NaHC03,  MgS04,  and  KC1  for  use  in  preparation  of the  reconstituted  waters.

7.2.3.3  To prepare 20 L of standard,  synthetic,  moderately hard,
reconstituted water, using mineral water such as PERRIERฎ Water, or equivalent
(Table 4), follow the instructions below.                             !

   1. Place 16 L of MILLI-Qฎ or equivalent water in a properly cleaned plastic
      carboy.
   2. Add 4 L of PERRIERฎ Water, or equivalent.                       ;
   3. Aerate vigorously for 24 h to stabilize the medium.              i
   4. The measured pH, hardness and alkalinity of the aerated water will be as
      indicated in Table 4.                                          .[
   5. This synthetic water is referred to as diluted mineral  water (DMW) in
      the toxicity test methods.                                      !

7.3  USE OF RECEIVING WATER AS DILUTION WATER                         ;

7.3.1  If the objectives of the test require the use of uncontaminated
receiving water as dilution water, and the receiving water is uncontaminated,
it may be possible to collect a sample of the receiving water upstream of, or
close to, but outside of the zone influenced by the effluent.  However, if the
receiving water is contaminated, it may be necessary to collect the sample in

                                      32                              ;

-------
      TABLE 3.   PREPARATION OF SYNTHETIC FRESHWATER USING REAGENT GRADE
                CHEMICALS1
Water
Type
Reagent
Added (mg/L)2
NaHC03 CaS04.2H20
Very soft
Soft
Moderately
Hard
Hard
Very hard
12.
48.

96.
192.
384.
0
0

0
0
0
7.
30.

60.
120.
240.
5
0

0
0
0
MgS04
7
30

60
120
240
.5
.0

.0
.0
.0
KC1
0.5
2.0

4.0
8.0
16.0
Final Water
•z /,
pH3
6.4-6
7.2-7

7.4-7
7.6-8
8.0-8
.8
.6

.8
.0
.4
Hardness
: 10-13
40-48

: 80-100
1 160-180
280-320
Quality
Alka-
linity4
10-13
30-35

60-70
110-120
225-245
       Taken in part from Marking and Dawson (1973).
       Add reagent grade chemicals to deionized water.   i
       Approximate equilibrium pH after 24 h of aeration.
       Expressed as mg CaC03/L.                           ;
an area "remote" from the discharge site, matching as closely as possible the
physical and chemical characteristics of the receiving water near the outfall.

7.3.2  The sample should be collected immediately prior to the test, but never
more than 96 h before the test begins.  Except where it is used within 24 h,
or in the case where large volumes are required for flow through tests, the
sample should be chilled to 4ฐC during or immediately following collection,
and maintained at that temperature prior to use in the test.

7.3.3  Receiving water containing debris or indigenous organisms that may be
confused with or attack the test organisms should be filtered through a sieve
having 60 /m mesh openings prior to use.

7.3.4  Where toxicity-free dilution water is required in ;a test, the water is
considered acceptable if test organisms show the required survival, growth,
and reproduction in the controls during the test.        ;

7.3.5  The regulatory authority may require that the hardness of the dilution
water be comparable to the receiving water at the discharge site.  This
requirement can be satisfied by collecting an uncontaminetted receiving water
with a suitable hardness, or adjusting the hardness of an otherwise suitable
receiving water by addition of reagents as indicated in Table 3.
                                      33

-------
      TABLE  4.   PREPARATION  OF  SYNTHETIC  FRESHWATER  USING MINERAL WATER1
                                                          Final Water Quality
Volume of
Water
Type
Mineral Watef
Added
(mi/ir
Proportion
of Mineral
Water
(%)

PH3

Hardness4
Al ka-
linity4
  Very soft           50
  Soft               100
  Moderately Hard    200
  Hard               400
  Very hard
 2.5
10.0
20.0
40.0
7.2-8.1
7.9-8.3
7.9-8.3
7.9-8.3
  10-13
  40-48
 80-100
160-180
  10-13
  30-35
  60-70
110-120
     From Mount et al.  (1987), and data provided by Philip Lewis,
     EMSL-Cincinnati, OH.                                              ;
     Add mineral water  to Milli-Qฎ water, or equivalent, to prepare Diluted
     Mineral Water (DMW).
     Approximate equilibrium pH after 24 h of aeration.
     Expressed as mg CaCO,/L.
     Dilutions of PERRIERฎ Water form a precipitate when concentrations
     equivalent to "very hard water" are aerated.                      :
7.4  USE OF TAP WATER AS DILUTION WATER

7.4.1  The use of tap water as dilution water is discouraged unless it Jis
dechlorinated and passed through a deionizer and carbon filter.  Tap water can
be dechlorinated by deionization, carbon filtration, or the use of sodium
thiosulfate.  Use of 3.6 mg/L (anhydrous) sodium thiosulfate will reduce
1.0 mg chlorine/L (APHA, 1992).  Following dechlbrination, total residual
chlorine should not exceed 0.01 mg/L.  Because of the possible toxicity of
thiosulfate to test organisms, a control lacking thiosulfate should be
included in toxicity tests utilizing thiosulfate-dechlorinated water.

7.4.2  To be adequate for general laboratory use following dechlorinatiion, the
tap water is passed through a deionizer and carbon filter to remove toxic
metals and organics, and to control hardness and alkalinity.

7.5  DILUTION WATER HOLDING

7.5.1  A given batch of dilution water should not be used for more than
14 days following preparation because of the possible build-up of bacterial,
fungal, or algal  slime growth and the problems associated with it.  The
container should be kept covered and the contents should be protected from
light.                                 .                                !
                                      34

-------
                                   SECTION 8

           EFFLUENT AND RECEIVING WATER SAMPLING, SAMPLE HANDLING,
                AND SAMPLE PREPARATION FOR TOXICITY TESTS


8.1  EFFLUENT SAMPLING

8.1.1  The effluent sampling point should be the same as that specified in the
NPDES discharge permit (USEPA, 1988a).  Conditions for exception would be:
(1) better access to a sampling point between the final treatment and the
discharge outfall; (2) if tne processed waste is chlorinated prior to
discharge, it may also be desirable to take samples prior to contact with the
chlorine to determine toxicity of the unchlorinated effluent; or (3) in the
event there is a desire to evaluate the toxicity of the influent to municipal
waste treatment plants or separate wastewater streams in industrial facilities
prior to their being combined with other wastewater streams or non-contact
cooling water, additional sampling points may be chosen.

8.1.2  The decision on whether to collect grab or composite samples is based
on the objectives of the test and an understanding of the short and long-term
operations and schedules of the discharger.  If the effluent quality varies
considerably with time, which can occur where holding times are short, grab
samples may seem preferable because of the  ease of collection and the
potential of observing peaks  (spikes) in toxicity.  However, the sampling
duration of a grab sample is  so short that  full characterization of an
effluent over a 24-h period would require a prohibitively large number of
separate samples and tests.   Collection of  a 24-h composite .sample, however,
may  dilute toxicity spikes, and average the quality of the effluent over the
sampling period.  Sampling recommendations  are provided below (also see USEPA,
1993b).

8.1.3  Aeration during collection and transfer of effluents  should be
minimized to reduce the  loss  of volatile  chemicals.

8.1.4  Details  of date,  time,  location, duration, and  procedures used  for
effluent  sample  and dilution  water  collection should be recorded.

8.2   EFFLUENT  SAMPLE TYPES

8.2.1  The  advantages  and disadvantages  of  effluent grab  and composite samples
are  listed  below:

8.2.1.1   GRAB  SAMPLES                                   \

     Advantages:

    1.  Easy  to  collect;  require a minimum of equipment  and on-site  time.
    2.  Provide  a measure of  instantaneous  toxicity.   Toxicity spikes  are  not
       masked by dilution.
                                       35

-------
    Disadvantages:

   1. Samples are collected over a very short period of time and on a
      relatively infrequent basis.  The chances of detecting a spike ip
      toxicity would depend on the frequency of sampling and the probability
      of missing a spike is high.

8.2.1.2  COMPOSITE SAMPLES

   Advantages:
   1.
   2,
       A single effluent sample is collected over a 24-h  period.
       The sample is collected over a much longer period  of time  than  a single
       grab sample and contains all  toxicity spikes.

    Disadvantages:

    1.  Sampling equipment is  more  sophisticated  and expensive,  and must:be
       placed  on-site for at  least 24 h.
    2.  Toxicity spikes may not be  detected because  they are masked by  dilution
       with less toxic wastes.

8.3  EFFLUENT SAMPLING RECOMMENDATIONS

8.3.1   When tests  are conducted on-site,  test solutions  can be renewed daily
with freshly  collected samples, except for  the  green alga, Selenastrum
capricornutum,  test  which  is  not  renewed.

8.3.2   When tests  are conducted off-site, a minimum of three samples  are
collected.  If these  samples  are  collected  on Test Days  1, 3, and 5,  the first
sample  would  be  used  for test  initiation, and for  test solution renewal on Day
2.  The second  sample would be used  for test solution renewal on Days 3 and 4.
The third  sample would  be  used for test solution renewal  on Days 5,  6, and 7.
                                                                       -,
8.3.3   Sufficient  sample volume must be collected to perform the required
toxicity and  chemical  tests.  A 4-L  (1-gal) CUBITAINERฎ will provide
sufficient sample  volume for most tests.                                '

8.3.4  THE FOLLOWING  EFFLUENT SAMPLING METHODS ARE RECOMMENDED:        !

8.3.4.1  Continuous Discharges

   1.  If the facility discharge is continuous,  but the calculated retention
      time of a continuously discharged effluent is less  than 14 days  and the
      variability of the waste is unknown, at a minimum,  four grab samples or
      four composite samples are collected over a 24-h period.   For  example,  a
      grab sample is taken every.6 h (total  of four samples) and each  sample
      is used  for a separate toxicity test,  or four successive  6-h composite
      samples  are taken and each is used  in a separate test.

   2.  If the calculated retention  time of a continuously  discharged  effluent
      is greater than 14 days, or  if it can be  demonstrated that  the   :

                                      36

-------
      wastewater does not vary more than 10% in toxicity over a 24-h period,
      regardless of retention time, a single grab sample is collected for a
      single toxicity test.

   3.  The retention time of the effluent in the wastewater treatment facility
      may be estimated from calculations based on the volume of the retention
      basin and rate of wastewater inflow.   However,  the calculated retention
      time may be much greater than the actual time because of
      short-circuiting in the holding basin.  Where short-circuiting is
      suspected, or sedimentation may have reduced holding basin capacity, a
      more accurate estimate of the retention time can be obtained by carrying
      out a dye study.                                   !

8.3.4.2.  Intermittent discharges                        ;

8.3.4.2.1  If the facility discharge is intermittent, a single grab sample is
collected midway during each discharge period.  Examples of intermittent
discharges are:                                          ;

   1.  When the effluent is continuously discharged during a single 8-h work
      shift (one sample is collected) or two successive 8-h work shifts (two
      samples are collected).
   2.  When the facility retains the wastewater during an B-h work shift, and
      then treats and releases the wastewater as a batch discharge (one sample
      is collected).
   3.  When, at the end the shift,  clean up activities result, in the discharge
      of a slug of toxic wastes (one sample is collected).

8.4  RECEIVING WATER SAMPLING

8.4.1   Logistical problems and difficulty in securing sampling equipment
generally preclude the collection  of composite receiving water samples for
toxicity tests.  Therefore,  based  on the requirements of the test, a single
grab sample or daily grab sample of receiving water  is collected for use  in
the test.

8.4.2   The sampling point is determined by the objectives, of the test.  In
rivers,  samples should be collected from mid-stream  and at mid-depth, if
accessible.   In lakes the samples  are collected at mid-depth.

8.4.3   To determine the extent of  the zone of toxicity in the receiving water
downstream from the outfall, receiving water  samples are collected at several
distances downstream  from the discharge.  The time required for the effluent-
receiving-water mixture to travel  to sampling points downstream from the
outfall, and  the rate and degree of mixing, may be difficult to ascertain.
Therefore,  it may  not be possible  to correlate downstream toxicity with
effluent toxicity  at  the discharge point unless a dye study is performed.  The
toxicity of receiving water  samples from five stations downstream from the
discharge point can  be evaluated using  the  same number of test vessels and
test organisms  as  used  in one effluent  toxicity test with five effluent
dilutions.


                                      37                 ;

-------
 8.5   EFFLUENT  AND  RECEIVING WATER SAMPLE  HANDLING,  PRESERVATION, AND SHIPPING

 8.5.1   Unless  the  samples  are  used in  an  on-site  toxicity  test  the day of
 collection,  they should  be chilled and maintained at  4ฐC until  used to inhibit
 microbial  degradation, chemical  transformations,  and  loss  of  highly volatile
 toxic  substances.                                                     :
8.5.2   Composite  samples  should  be  chilled  as  they  are collected.
samples  should  be chilled immediately  following collection.
Grab
8.5.3   If  the  effluent  has  been  chlorinated, total residual chlorine must be
measured immediately  following sample  collection.

8.5.4   Sample  holding time  begins when the last grab sample in a series  is
taken  (i.e., when  a series  of four grab samples are taken over a 24-h period),
or when a  24-h  composite  sampling period  is completed.   If the data from the
samples are to  be  acceptable for use in the NPDES Program, the lapsed ^time
(holding time)  from sample  collection  to  first use of the sample in test
initiation must not exceed  36 h.  EPA  believes that 36 h is adequate time to
deliver the samples to  the  laboratories performing the test in most cases.  In
the  isolated cases, where the permittee can document that this delivery time
cannot  be met,  the permitting authority can allow an option for on-site
testing or a variance for an extension of shipped sample holding time.  The
request for a variance  in sample holding  time, directed to the USEPA Regional
Administrator under 40  CFR  136.3(e) must  include supportive data which show
that the toxicity of  the effluent sample  is not reduced  (e.g., because of
volatilization  and/or sorption of toxics  on the sample container surfaces) by
extending the holding time  beyond more than 36 h.  However in no case should
more than 72 h  elapse between collection  and first use of the sample.  In
static-renewal  tests, the original sample may also be used to prepare test
solutions for renewal at 24 h and 48 h after test initiation,  if stored at
4ฐC, with minimum head  space, as described in Subsection 8.5.   Guidance for
determining the persistence of the sample is provided in Subsection 8.7.

8.5.5   To minimize the  loss of toxicity due to volatilization of toxic
constituents, all sample containers should be "completely" filled,  leaving no
air space between the contents and the lid.                           ;

8.5.6   SAMPLES  USED IN  ON-SITE TESTS

8.5.6.1  Samples collected for on-site tests should be used within  24 h.

8.5.7   SAMPLES SHIPPED  TO OFF-SITE FACILITIES                         !

8.5.7.1  Samples collected for off-site toxicity testing are to be  chilled to
4ฐC during or immediately after collection, and shipped iced to the performing
laboratory.  Sufficient ice should be placed with the sample in the shipping
container to ensure that ice will still be present when the sample  arrives at
the laboratory and is unpacked.   Insulating material  must not  be placed
between the ice and the sample in the shipping container.             ;
                                      38

-------
8 5.7.2  Samples may be shipped in one or more 4-L  (1-gal) CUBITAINERSฎ or new
plastic "milk" jugs.  All sample containers should  be rinsed with source water
before being filled with sample.  After use with receiving water or effluents,
CUBITAINERSฎ and plastic jugs are punctured to prevent reuse.

8573  Several sample shipping options are  available,  including Express
Mail, air express, bus, and  courier  service.  Express Mail is  delivered seven
days  a week.   Saturday and Sunday shipping and receiving schedules of  private
.carriers vary  with the carrier.

8.6   SAMPLE RECEIVING

8.6.1 Upon arrival  at the laboratory,  samples are  logged in  and the
temperature is measured and  recorded.   If  the samples are not  immediately
prepared for testing,  they are  stored  at 4ฐC  until  used.

8.6.2 Every effort  must  be  made  to  initiate  the  test with an  effluent sample
on  the day  of  arrival  in  the laboratory, and  the  sample  holding time  should
not exceed  36  h  unless a  variance  has  been granted  by the NPDES permitting
 authority.

 8.7  PERSISTENCE OF  EFFLUENT TOXICITY DURING  SAMPLE SHIPMENT AND HOLDING

 8.7.1 The  persistence of the toxicity of  an  effluent  prior to its  use in  a
 toxicity test  is of interest in assessing  the validity  of toxicity  test data,
 and in determining the possible effects of allowing an  extension of the
 holding  time.   Where a variance in holding time  (> 36  h, but < 72  h)  is
 requested by a permittee, (see Subsection  8.5.4  above),  information on the
 effects  of the extension in holding time on the  toxicity of samples must  be
 obtained by comparing the results of multi-concentration chronic toxicity
 tests performed on effluent samples held 36 h with toxicity test results  using
 the same samples after they were held for the requested,:longer period.  The
 portion of the sample set aside for the second test should be held under the
 same conditions as during shipment and holding.

 8.8  PREPARATION OF EFFLUENT AND RECEIVING WATER SAMPLES FOR TOXICITY TESTS

 8.8.1  When aliquots are removed from the sample container, the head space
 above the remaining sample  should be held to a minimum.  Air which enters a
 container upon removal of sample should be expelled by  compressing the
 container before reclosing,  if possible (i.e., where a  CUBITAINERฎ is used),
 or by using an appropriate  discharge valve (spigot).

 8.8.2   With  the daphnid, Cen'odaphm'a dubia, and  fathead minnow, Pimephales
 promelas, tests, effluents  and receiving waters must be filtered through  a
 60-Aim plankton net to remove indigenous organisms  that  may attack or  be
 confused with the test organisms (see the daphnid, Cen'odaphm'a dubia, test
 method for details).  Receiving waters used  in green alga, Selenastrum
 capn'cornutum, toxicity  tests must  be filtered through  a;0.45-Mm pore diameter
 filter before use.   It may  be necessary to first coarse-filter the dilution
 and/or waste  water  through  a nylon  sieve  having  2- to 4-mm holes to remove
 debris  and/or break  up large floating or  suspended solids.   Because filtration

                                        39

-------
 may increase the DO in the effluent, the DO should be checked both before and
 after filtering.  Caution:  filtration may remove some toxicity.

 8.8.3  If the samples must be warmed to bring them to the prescribed test
 temperature, supersaturation of the dissolved oxygen and nitrogen may'become a
 problem.   To avoid this problem, the effluent-an.d dilution water are checked
 with a DO probe after reaching test temperature and, if the DO is greater than
 100% saturation or lower than 4.0 mg/L, the solutions are aerated moderately
 (approximately 500 mL/min) for a few minutes, using an airstone, until the DO
 is within the prescribed range (> 4.0 mg/L).   Caution:  avoid excessive
 aeration.

 8.8.4  The DO concentration in the samples should be near saturation prior to
 use.  Aeration will  bring the DO and other gases into equilibrium with air,
 minimize  oxygen demand,  and stabilize the pH.  However,  aeration" during
 collection,  transfer,  and preparation of samples should  be minimized to reduce
 the loss  of volatile chemicals.

 8.8.4.1  Aeration during the test may alter the results  and should be iused
 only as a last resort  to maintain the required DO.   Aeration can reduce the
 apparent  toxicity of the test solutions by stripping them of highly volatile
 toxic substances,  or increase their toxicity  by altering pH.  However,  the DO
 in the test  solutions  must not be allowed to  fall  below  4.0 mg/L.

 8.8.4.2  In  static tests (renewal  or non-renewal),  low DOs may commonly occur
 in the higher concentrations of  wastewater.   Aeration is accomplished by
 bubbling  air through a pipet at  a rate  of 100 bubbles/min.   If aeration is
 necessary, all  test  solutions must be aerated.   It  is advisable  to monjitor the
 DO closely during  the  first few  hours of the  test.   Samples with  a potential
 DO problem generally show a downward trend in DO within  4 to 8 h  after the
 test is started.   Unless aeration  is initiated  during the first  8  h of the
 test,  the DO may  be  exhausted during an. unattended  period,  thereby .
 invalidating the  test.

 8.8.5  At a  minimum, pH,  conductivity,  and total  residual  chlorine are
 measured  in  the undiluted  effluent or receiving  water, and  pH  and  conductivity
 are  measured in the  dilution  water.                                    :

 8.8.5.1   It  is  recommended  that  total alkalinity and  total  hardness  also be
 measured  in  the undiluted  effluent test water,  receiving  water, and  the
 dilution  water.

 8.8.6  Total  ammonia is  measured  in  effluent  and receiving  water samples where
 toxicity  may be contributed  by un-ionized  ammonia (i.e.,  where total  ammonia
ฃ  5 mg/L).   The concentration  (mg/L)  of un-ionized  (free)  ammonia  in  a-sample
 is a function of temperature  and pH,  and  is calculated using the percentage
 value obtained from Table 5,  under the appropriate pH and temperature,  and
multiplying  it by the concentration  (mg/L) of total ammonia  in the  sample.

8.8.7  Effluents and receiving waters can be dechlorinated  using 6.7 mg/L
anhydrous  sodium thiosulfate to reduce 1 mg/L chlorine (APHA,  1992).  Note
that the amount of thiosulfate required to dechlorinate effluents  is greater

                                      40                               !

-------
 TABLE 5. PERCENT UNIONIZED  NH3  IN AQUEOUS AMMONIA  SOLUTIONS:
          15-26ฐC AND  pH  6.0-8.91
TEMPERATURES
 pH
                                 TEMPERATURE (ฐC)

6.0
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
7.0
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
8.0
8.1
8.2
8.3
8.4
8.5
8.6
8.7
8.8
8.9
15
0.0274
0.0345
0.0434
0.0546
0.0687
0.0865
0.109
0.137
0.172
0.217
0.273
0.343
0.432
0.543
0.683
0.858
1.08
1.35
1.70
2.13
2.66
3.33
4.16
5.18
6.43
7.97
9.83
12.07
14.7
17.9
16
0.0295
0.0372
0.0468
0.0589
0.0741
0.0933
0.117
0.148
0.186
0.234
0.294
0.370
0.466
0.586
0.736
0.925
1.16
1.46
1.83
2.29
2.87
3.58
4.47
5.56
6.90
8.54
10.5
12.9
15.7
19.0
17
0.0318
0.0400
0.0504
0.0634
0.0799
0.1005
0.127
0.159
0.200
0.252
0.317
0.399
0.502
0.631
0.793
0.996
1.25
1.57
1.97
2.46
3.08
3.85
4.80
5.97
7.40
9.14
11.2
13.8
16.7
20.2
18
0.0343
0.0431
0.0543
0.0683
0.0860
0.1083
0.136
0.171
0.216
0.271
0.342
0.430
0.540
0.679
0.854
1.07
1.35
1.69
2.12
2.65
3.31
4.14
5.15
6.40
7.93
9.78
12.0
14.7
17.8
21.4
19
0.0369
0.0464
0.0584
0.0736
0.0926
0.1166
0.147
0.185
0.232
0.292
0.368
0.462
0.581
0.731
0.918
1.15
1.45
1.82
2.28
2.85
3.56
4.44
5.52
6.86
8.48
10.45
12.8
15.6
18.9
22.7
20
0.0397
0.0500
0.0629
0.0792
0.0996
0.1254
0.158
0.199
0.250
0.314
0.396
0.497
0.625
0.786
0.988
1.24
1.56
1.95
2.44
3.06
3.82
4.76
5.92
7.34
9.07
11.16
13.6
16.6
20.0
24.0
21
0.0427
0.0537
0.0676
0.0851
0.107
0.135
0.170
0.214
0.269
0.338
0.425
0.535
0.672
0.845
1.061
1.33
1.67
2.10
2.62
3.28
4.10
5.10
6.34
7.85
9.69
11.90
14.5
17.6
21.2
25.3
22
0.0459
0.0578
0.0727
0.0915
0.115
0.145
0.182
0.230
0.289
0.363
0.457
0.575
0.722
0.908
1.140
1.43
1.80
2.25
2.82
3.52
4.39
5.46
6.78
8.39
10.3
12.7
15.5
18.7
22.5
26.7
23
0.0493
0.0621
0.0781
0.0983
0.124 ,
0.156
0.196
0.247
0.310
0.390 ••
0.491 :
0.617
0.776
0.975
1.224
1.54 •
1.93
2.41
3.02
3.77
4.70
5.85
7.25
8.96 ;
11.0
13.5
16.4
19.8
23.7
28.2 .
24
0.0530
0.0667
0.0901
0.1134
0.133
0.167
• 0.210
0.265
0.333
0.419
0.527
0.663
0.833
1.05
1.31
1.65
2.07
2.59
3.24
4.04
5.03
6.25
7.75
9.56
11.7
14.4
17.4
21.0
25.1
29.6
25
0.0568
0.0716
0.0901
0.1134
0.143
0.180
0.226
0.284
0.358
0.450
0.566
0.711
0.893
1.12
1.41
1.77
2.21
2.77
3.46
4.32
5.38
6.68
8.27
10.2
12,5
15.2
18.5
22.2
26.4
31.1
26
0.0610
0.0768
0.0966
0.1216
0.153
0.193
0.242
0.305
0.384
0.482
0.607
0.762
0.958
1.20
1.51
1.89
2.37
2.97
3.71
4.62
5.75
7.14
8.82
10.9
13.3
16.2
19.5
23.4
27.8
32.6
     Table provided by Teresa Norberg-King, ERL, Duluth, ;Minnesota.  Also
     see Emerson et al. (1975), Thurston et al. (1974), and USEPA  (1985a).
than the amount needed to dechlorinate tap water  (see Section 7, Dilution
Water, Subsection 7.4.1).  Since thiosulfate may  contribute to sample
toxicity, a thiosulfate control should be used in the test in addition to the
normal dilution water control.                           !

8.8.8  Mortality or impairment of growth or reproduction due to pH  alone may
occur if the pH of the sample falls outside the range of 6.0 - 9.0.  Thus,  the
presence of other forms of toxicity (metals and organics) in the sample may be
masked by the toxic effects of low or high pH.  The question about  the
presence of other toxicants can be answered only  by performing two  parallel
tests, one with an adjusted pH, and one without an adjusted pH.  Freshwater
samples are adjusted to pH 7.0 by adding IN NaOH  or IN  HC1 dropwise, as
required, being careful to avoid overadjustment.
                                       41

-------
 8.9  PRELIMINARY TOXICITY RANGE-FINDING TESTS

 8.9.1  USEPA Regional and State personnel generally have observed that it is
 not necessary to conduct a toxicity range-finding test prior to initiating a
 static,  chronic, definitive toxicity test.  However,  when preparing to perform
 a static test with a sample of completely unknown quality,  or before
 initiating a flow-through test, it is advisable to conduct  a preliminary
 toxicity range-finding test.

 8.9.2  A toxicity range-finding test ordinarily consists of a down-scaled,
 abbreviated static acute test in which groups of five organisms are exposed to
 several  widely-spaced sample  dilutions in a logarithmic series, such as 100%,
 10.0%,  1.00%,  and 0.100%,  and a control,  for 8-24 h.   Caution:   if the sample
 must also be used for the full-scale definitive test,  the 36-h  limit on
 holding  time (see Subsection  8.5.4)  must  not be exceeded before the definitive
 test is  initiated.                                                    :

 8.9.3  It should be noted that the toxicity (LC50)  of a sample  observed in  a
 range-finding  test  may be significantly different from the  toxicity observed
 .in  the follow-up chronic definitive  test  because:   (1)  the  definitive test  is
 longer;  and (2)  the test may  be performed with  a sample collected  at a
 different time,  and possibly  differing significantly  in the level  of toxicity.

 8.10 HULTI-CONCENTRATION (DEFINITIVE)  EFFLUENT TOXICITY TESTS
                                                                      i
 8.10.1   The tests recommended for use  in  determining  discharge  permit >
 compliance  in  the NPDES  program are  multi-concentration,  or definitive,  tests
 which provide  (1) a point  estimate of  effluent  toxicity in  terms of an  IC25,
 IC50, or LC50, or  (2)  a  no-observed-effect-concentration  (NOEC) defined  in
 terms of mortality,  growth, reproduction,  and/or teratogenicity and  obtained
 by  hypothesis  testing.   The tests  may  be  static  renewal  or  static  non-renewal.

 8.10.2   The tests consist  of  a  control  and  a minimum of five  effluent
 concentrations commonly  selected  to  approximate  a geometric  series,  such  as
 100%, 50%,  25%,  12.5%, and 6.25%,  using a > 0.5  dilution  series.

 8.10.3   These  tests  are  also  to be used in determining  compliance  with permit
 limits on the mortality  of the  receiving water concentration  (RWC) of
 effluents by bracketing  the RWC with effluent concentrations  in the  following
manner:   (1) 100% effluent, (2) [RWC +  100]/2,  (3) RWC,  (4)  RWC/2, andi  (5)
 RWC/4.   For example, where the  RWC = 50%, the effluent  concentrations used in
the test would be 100%,  75%,  50%,  25%, and 12.5%.

8.10.4   If  acute/chronic ratios are to be determined by simultaneous acute and
short-term chronic tests with a single species,  using the same  sample,'both
types of tests must use the same test conditions, i.e., pH,  temperature, water
hardness, salinity, etc.
                                      42

-------
8.11  RECEIVING WATER TESTS

8.11.1  Receiving water toxicity tests generally consist of 100% receiving
water and a control.  The total hardness of the control should be comparable
to the receiving water.

8.11.2  The data from the two treatments are analyzed by hypothesis testing to
determine if test organism survival in the receiving water differs
significantly from the control.  Four replicates and 10 organisms per
replicate are required for each treatment (see Summary of Test Conditions and
Test Acceptability Criteria in the specific test method).

8.11.3  In cases where the objective of the test is to estimate the degree of
toxicity of the receiving water, a multi-concentration test is performed by
preparing dilutions of the receiving water, using a > 0.5 dilution series,
with a suitable control water.
                                       43

-------
                                   SECTION 9

               CHRONIC  TOXICITY TEST  ENDPOINTS  AND  DATA ANALYSIS


 9.1   ENDPOINTS

 9.1.1  The  objective of  chronic aquatic  toxicity tests with  effluents
and pure
compounds  is  to  estimate  the  highest  "safe"  or  "no-effect  concentration" of
these  substances.   For  practical  reasons,  the responses  observed  in these
tests  are  usually  limited to  hatchability, gross morphological  abnormalities,
survival,  growth,  and reproduction, and  the  results  of the tests  are  usually
expressed  in  terms of the highest toxicant concentration that has no
statistically significant observed effect  on these responses, when compared to
the controls.  The terms  currently used  to define the endpoints employed in
the rapid,  chronic and  sub-chronic toxicity  tests have been derived from the
terms  previously used for full  life-cycle  tests.  As shorter chronic  tests
were developed,  it became common  practice  to apply the same terminology to the
endpoints.  The  terms used in this manual  are as follows:

9.1.1.1  Safe  Concentration - The highest  concentration  of toxicant that will
permit normal  propagation of  fish and other  aquatic  life in receiving waters.
The concept of a "safe  concentration" is a biological concept, whereas the
"no-observed-effect concentration" (below) is a statistically defined
concentration.

9.1.1.2  No-Observed-Effect-Concentration  (NOEC) - The highest concentration
of toxicant to which organisms  are exposed in a full life-cycle or partial
life-cycle  (short-term) test, that causes  no observable  adverse effects on the
test organisms (i.e., the highest concentration of toxicant in which the
values for the observed responses are not  statistically  significantly
different from the controls).   This value  is used, along with other factors,
to determine toxicity limits  in permits.

9.1.1.3  Lowest-Observed-Effect-Concentration (LOEC) - The lowest
concentration of toxicant  to which organisms are exposed in a life-cycle or
partial life-cycle  (short-term) test,  which  causes adverse effects on the test
organisms (i.e., where the values  for the  observed responses are statistically
significantly different from the  controls).

9.1.1.4  Effective Concentration  (EC)  - A  point estimate of the toxicant
concentration that would  cause  an observable adverse affect on a quantal,  "all
or nothing," response (such as death,  immobilization, or serious
incapacitation)  in a given percent of the organisms, calculated by point
estimation techniques.  If the observable effect is death or immobility, the
term,  Lethal Concentration (LC),  should be used (see Subsection 9.1.1.5).   A
certain EC or LC value might be judged from a biological  standpoint to
represent a threshold concentration,  or lowest concentration that would cause
an adverse effect on the observed response.

9.1.1.5  Lethal Concentration (LC) -  The toxicant concentration that would
cause death in a given percent of the  test population.    Identical  to EC when

                                      44

-------
 the observed  adverse  effect  is  death.   For  example,  the  LC50  is  the
 concentration of toxicant  that  would cause  death  in  50%  of  the test
 population.

 9.1.1.6   Inhibition Concentration  (1C)  -  The  toxicant  concentration  that would
 cause  a given percent reduction in  a non-quanta!  biological measurement for
 the test  population.  For  example,  the  IC25 is the concentration of  toxicant
 that would cause a 25% reduction in mean  young per female or  in  growth for the
 test population, and  the IC50 is the concentration of  toxicant that  would
 cause  a 50% reduction.

 9.2  RELATIONSHIP BETWEEN  ENDPOINTS DETERMINED BY HYPOTHESIS  TESTING AND POINT
     ESTIMATION TECHNIQUES

 9.2.1  If the objective of chronic  aquatic  toxicity  tests with effluents and
 pure compounds is to  estimate the highest "safe or no-effect  concentration" of
 these  substances, it  is imperative  to understand  how the statistical endpoints
 of these  tests are related to the "safe"  or "no-effect" concentration.  NOECs
 and LOECs are determined by  hypothesis  testing (Dunnett's Test,  a t  test with
 the Bonferroni adjustment, Steel's  Many-one Rank  Test, or the Wilcoxon Rank
 Sum Test  with the Bonferroni adjustment), whereas LCs, ICs, and  ECs  are
 determined by point estimation  techniques (Probit Analysis, Spearman-Karber
 Method, Trimmed Spearman-Karber Method, Graphical Method or Linear
 Interpolation Method).  There are inherent  differences between the use of a
 NOEC or LOEC derived  from  hypothesis testing to estimate a  "safe"
 concentration, and the use of a LC, EC, 1C, or other point estimates derived
 from curve fitting, interpolation,  etc.

 9.2.2  Most point estimates, such as the  LC, 1C,   or  EC, are derived  from a
 mathematical model  that assumes a continuous dose-response relationship.   By
 definition, any LC, 1C,  or EC value is  an estimate of some amount of adverse
 effect.  Thus the assessment of a "safe" concentration must be made  from a
 biological standpoint rather than with  a statistical  test,.  In this  instance,
 the^biologist must determine some amount of adverse effect that  is deemed to
 be "safe", in the sense that from a practical  biological  viewpoint it will  not
 affect the normal  propagation of fish and other aquatic life in receiving
waters.

 9.2.3  The use of NOECs  and LOECs,  on the other hand, assumes either (1)  a
continuous dose-response relationship,  or (2)  a non-continuous (threshold)
model  of the dose-response relationship.

9.2.3.1  In the case of a continuous dose-response relationship,  it is also
assumed that adverse effects that are not "statistically  observable"  are  also
not important from a biological  standpoint,  since they are not pronounced
enough to test as  statistically significant against some  measure  of the
natural variability of the responses.

9.2.3.2  In the case of  non-continuous  dose-response  relationships,  it is
assumed that there  exists a true threshold,  or concentration below which  there
is no  adverse effect on  aquatic  life,  and above which there  is an adverse
                                      45

-------
effect.  The purpose of the statistical analysis in this case is to estimate
as closely as possible where that threshold lies.                     i

9.2.3.3  In either case,  it is  important to realize that the amount of adverse
effect that is statistically observable (LOEC) or not observable (NOEC) is
highly dependent on all aspects of the experimental design, such as the number
of concentrations of toxicant,  number of replicates per concentration; number
of organisms per replicate, and use of randomization.  Other factors that
affect the sensitivity of the test include the choice of statistical analysis,
the choice of an alpha level, and the amount  of variability between responses
at a given concentration.                                             i

9.2.3.4  Where the assumption of a continuous dose-response relationship is
made, by definition some  amount of adverse effect might be present at the
NOEC, but is not great enough to be detected  by hypothesis testing.

9.2.3.5  Where the assumption of a non-continuous dose-response relationship
is made, the NOEC would indeed  be an estimate of a  "safe" or "no-effect"
concentration if the amount of  adverse effect that  appears at the threshold is
great enough to test as statistically significantly different from the
'controls in the face of all aspects of the experimental design mentioned
above.   If, however, the  amount of adverse effect at the threshold were not
great enough to test as statistically different, some amount of adverse effect
might be present at the NOEC.   In any case, the estimate of the NOEC with
hypothesis testing is always dependent on the aspects of the experimental
design mentioned above.   For this reason, the reporting and examination of
some measure of the sensitivity of the test  (either the minimum significant
difference or the percent change from the control that this minimum difference
represents) is extremely  important.                                   ;

9.2.4   In summary, the assessment of a "safe" or "no-effect" concentration
cannot be made from the results of statistical  analysis alone, unless;(1) the
assumptions of a strict threshold model are  accepted, and  (2)  it  is assumed
that the amount of adverse  effect present at  the threshold  is  statistically
detectable by hypothesis  testing.   In this case, estimates  obtained from a
statistical analysis are  indeed estimates of  a  "no-effect"  concentration.
If the  assumptions are not  deemed tenable, then estimates  from a  statistical
analysis can only be used in conjunction with an assessment from  a  biological
standpoint of what magnitude of adverse effect  constitutes  a  "safe"
concentration.   In this  instance, a  "safe" concentration  is not necessarily a
truly  "no-effect" concentration,  but rather  a concentration at which  the
effects  are judged to  be  of no  biological significance.               ;

9.2.5  A better  understanding  of  the relationship  between  endpoints derived by
hypothesis testing  (NOECs)  and  point estimation techniques  (LCs,  ICs,j and  ECs)
would be very helpful  in  choosing methods of data  analysis.   Norberg-King
 (1991)  reported  that the  IC25s  were  comparable  to  the  NOECs for 23  effluent
and  reference toxicant data sets  analyzed.   The data  sets  included  short-term
chronic  toxicity tests for  the  fathead minnow,  Pimephales  promelas,  and  the
daphnid, Ceriodaphm'a  dubia.   Birge  et  al.  (1985)  reported  that  LCls  derived
from Probit Analysis of  data  from short-term embryo-larval  tests  with
reference  toxicants were  comparable  to  NOECs  for  several  organisms.

                                       46

-------
Similarly, USEPA (1988d) reported that the IC25s were comparable to the NOECs
for a set of daphnid, Ceriodaphnia dubia, chronic tests with a single
reference toxicant.  However, the scope of these comparisons was very limited,
and sufficient information is not yet available to establish an overall
relationship between these two types of endpoints, especially when derived
from effluent toxicity test data.

9.3  PRECISION

9.3.1  HYPOTHESIS TESTS                                  !

9.3.1.1  When hypothesis tests are used to analyze toxicity test data, it is
not possible to express precision in terms of a commonly used statistic.
The results of the test are given in terms of two endpoints, the No-Observed-
Effect Concentration (NOEC) and the Lowest-Observed-Effect Concentration
(LOEC).  The NOEC and LOEC are limited to the concentrations selected for the
test.  The width of the NOEC-LOEC interval is a function of the dilution
series, and differs greatly depending on whether a dilution factor of 0.3 or
0.5 is used in the test design.  Therefore, USEPA recommends the use of the
> 0.5 dilution factor (see Section 4, Quality Assurance).  It is not possible
to place confidence limits on the NOEC and LOEC derived from a given test, and
it is difficult to quantify the precision of the NOEC-LOEC endpoints between
tests.  If the data from a series of tests performed with the same toxicant,
toxicant concentrations, and test species, were analyzed with hypothesis
tests, precision could only be assessed by a qualitative comparison of the
NOEC-LOEC intervals, with the understanding that maximum precision would be
attained if all tests yielded the same NOEC-LOEC interval;   In practice, the
precision of results of repetitive chronic tests is considered acceptable if
the NOECs vary by no more than one concentration intervaliabove or below a
central tendency.  Using these guidelines, the "normal" range of NOECs from
toxicity tests using a 0.5 dilution factor (two-fold difference between
adjacent concentrations), would be four-fold.

9.3.2  POINT ESTIMATION TECHNIQUES                       :

9.3.2.1  Point estimation techniques have the advantage of providing a point
estimate of the toxicant concentration causing a given amount of adverse
(inhibiting) effect, the precision of which can be quantitatively assessed
(1) within tests by calculation of 95% confidence limits, and (2) across tests
by calculating a standard deviation and coefficient of variation.

9.4  DATA ANALYSIS

9.4.1  ROLE OF THE STATISTICIAN                          \

9.4.1.1  The use of the statistical methods described in this manual for
routine data analysis does not require the assistance of a statistician.
However, the interpretation of the results of the analysis of the data from
any of the toxicity tests described in this manual can become problematic
because of the inherent variability and sometimes unavoidable anomalies in
biological data.   If the data appear unusual in any way, or fail to meet the
necessary assumptions, a statistician should be consulted.   Analysts who are

                                      47

-------
not proficient in statistics are strongly advised to seek the assistance of a
statistician before selecting the method of analysis and using any of the
results.

9.4.1.2  The statistical methods recommended in this manual are not the only
possible methods of statistical analysis.  Many other methods have been
proposed and considered.  Certainly there are other reasonable and defensible
methods of statistical analysis for this kind of toxicity data.  Among
alternative hypothesis tests some, like Williams' Test, require additional
assumptions, while others, like the bootstrap methods, require computer-
intensive computations.  Alternative point estimation approaches most probably
would require the services of a statistician to determine the appropriateness
of the model (goodness of fit), higher order linear or nonlinear models,
confidence intervals for estimates generated by inverse regression, etc.
In addition, point estimation or regression approaches would require the
'specification by biologists or toxicologists of some low level of adverse
effect that would be deemed acceptable or safe.  The statistical methods
contained in this manual have been chosen because they are (1) applicable to
most of the different toxicity test data sets for which they are recommended,
(2) powerful statistical tests, (3) hopefully "easily" understood by
nonstatisticians, and (4) amenable to use without a computer, if necessary.

9.4.2  PLOTTING THE DATA                                              :

9.4.2.1  The data should be plotted, both as a preliminary step to help detect
problems and unsuspected trends or patterns in the responses, and as an aid in
interpretation of the results.  Further discussion and plotted sets of data
are included in the methods and the Appendices.                       ;

9.4.3  DATA TRANSFORMATIONS                                           ;

9.4.3.1  Transformations of the data, (e.g., arc sine square root and logs),
are used where necessary to meet assumptions of the proposed analyses, such as
the requirement for normally distributed data.

9.4.4  INDEPENDENCE, RANDOMIZATION, AND OUTLIERS                      ,

9.4.4.1  Statistical independence among observations is a critical assumption
in all statistical analysis of toxicity data.  One of the best ways to insure
independence is to properly follow rigorous randomization procedures.
Randomization techniques should be employed at the start of the test, ;
including the randomization of the placement of test organisms in the test
chambers and randomization of the test chamber location within the array of
chambers.  Discussions of statistical independence, outliers and
randomization, and a sample randomization scheme, are included in Appendix A.

9.4.5  REPLICATION AND SENSITIVITY                                    !

9.4.5.1  The number of replicates employed for each toxicant concentration is
an important factor in determining the sensitivity of chronic toxicity tests.
Test sensitivity generally increases as the number of replicates is increased,
but the point of diminishing returns in sensitivity may be reached rather

                                      48

-------
 quickly.  The level of sensitivity required by a hypothesis test or the
 confidence interval for a point estimate will determine the number of
 replicates, and should be based on the objectives for obtaining the toxicity
 UclLctป

 9.4.5.2  In a statistical analysis of toxicity data, the choice of a
 particular analysis and the ability to detect departures from the assumptions
 of the analysis,  such as the normal distribution of the data and homogeneity
 of variance,  is also dependent on the number of replicates.  More than the
 minimum number of replicates may be required in situations where it is
 imperative to obtain optimal statistical  results, such as with tests used in
 enforcement cases or when it is not possible to repeat the tests.  For
 example, when the data are analyzed by hypothesis testing,: the nonparametric
 alternatives  cannot be used unless there  are at least four replicates at each
 toxicant concentration.
                                                           i
 9.4.6   RECOMMENDED ALPHA LEVELS                           !

 9.4.6.1  The  data analysis examples included in the manual  specify an alpha
 level  of 0.01 for testing the assumptions of hypothesis tests  and an alpha
 level  of 0.05 for the hypothesis tests themselves.   These levels are common
 and  well  accepted levels  for this type of analysis  and are presented as  a
 recommended minimum significance level  for toxicity test data  analysis.

 9.5  CHOICE OF ANALYSIS

 9.5.1   The  recommended statistical  analysis of  most data from  chronic  toxicity
 tests  with  aquatic organisms follows  a decision process  illustrated  in the
 flowchart  in  Figure 2.  An initial  decision is  made to use  point estimation
 techniques  (the Probit Analysis,  the  Spearman-Karber Method, the Trimmed
 Spearman-Karber Method, the  Graphical  Method, or  Linear  Interpolation Method)
 and/or to use  hypothesis  testing  (Dunnett's Test, the  t  test with the
 Bonferroni  adjustment, Steel's  Many-one Rank Test,  or  the Wilcoxon Rank Sum
 Test with the  Bonferroni  adjustment).  NOTE: For  the NPDES  Permit Program, the
 point  estimation  techniques  are the preferred statistical methods in
 calculating end points for effluent toxicity tests.   If  hypothesis testing  is
 chosen,  subsequent  decisions  are made  on  the appropriate procedure for a given
 set of data, depending on  the results  of  the tests  of  assumptions  as
 illustrated in the  flowchart.   A specific  flow  chart  is  included  in the
 analysis section  for each  test.

 95.2   Since a single chronic toxicity test might yield  information on more
 than one parameter  (such as  survival, growth, and reproduction), the lowest
 estimate of a  "no-observed-effect concentration" for any of the responses
would  be used as the "no-observed-effect concentration" for each test   It
 follows logically that in  the statistical  analysis of the data, concentrations
that had a significant toxic effect on one of the observed responses would not
be subsequently tested for an effect on some other response.'.  This is one
reason  for excluding concentrations that have shown a statistically
significant reduction in survival from a subsequent hypothesis  test for
effects on another parameter such as reproduction.  A second reason is that
the exclusion  of such concentrations usually results in a more  powerful  and

                                      49

-------
                 DATA (SURVIVAL, GROWTH, REPRODUCTION, ETC.)
          POINT
        ESTIMATION
                            HYPOTHESIS TESTING
                             TRANSFORMATION?
    ENDPOINT ESTIMATE
         LC, EC, 1C
                            SHAPIRO-WIUCS TEST
                 NON-NORMAL DISTRIBUTION
                  NORMAL DISTRIBUTION
       HOMOGENEOUS
         VARIANCE
                              BARTLETT'STEST
                                            HETEROGENEOUS
                                               VARIANCE
                           NO STATISTICAL ANALYSIS
                               RECOMMENDED
                 NO
                 •ซ*-
   4 OR MORE
   REPLICATES?
                                                      YES
             EQUAL NUMBER OF
                REPLICATES?
    T-TESTWITH
    BONFERRONI
    ADJUSTMENT
               EQUAL NUMBER OF
                 REPLICATES?
                                        YES
                                                              NO
STEEL'S MANY-ONE
   RANKTEST
  WILCOXON RANK SUM
      TEST WITH
BONFERRONI ADJUSTMENT
                             ENDPOINT ESTIMATES
                                 NOEC, LOEC
Figure 2.    Flowchart for statistical  analysis of  test data.

                                     50

-------
appropriate statistical  analysis.   In  performing  the  point  estimation
techniques recommended  in this manual,  an  all-data  approach is  used.   For
example, data from concentrations  above  the NOEC  for  survival are  included  in
determining ICp estimates using the  Linear Interpolation Method.

9.5.3  ANALYSIS OF GROWTH AND REPRODUCTION DATA           ;

9.5.3.1  Growth data from the fathead  minnow, Pimephales promelas,  larval
survival and growth test are analyzed  using hypothesis testing  or  point
estimation techniques according to the flowchart  in Figure  2.   The above
mentioned growth data may also be  analyzed by generating a  point estimate with
the Linear Interpolation Method.   Data from effluent  conceintrations that have
tested significantly different from  the  control for survival are excluded from
further hypothesis tests concerning  growth effects.   Growth is  defined as the
dry weight per original number of  test organisms  when group weights are
obtained.  When analyzing the data using point estimation techniques, data
from all concentrations are included in  the analysis.

9.5.3.2  Reproduction data from the  daphnid, Ceriodaphnia dubia, survival and
reproduction test are analyzed using hypothesis testing or  point estimation
techniques according to the flowchart  in Figure 2.  In hypothesis  testing,
data from effluent concentrations that have significantly lower survival than
the control, as determined by Fisher's Exact test, are not!included in the
hypothesis tests for reproductive effects.  Data  from all concentrations are
included when using point estimation techniques.

9.5.4  ANALYSIS OF ALGAL GROWTH RESPONSE DATA             •'

9.5.4.1  The growth response data from the green  alga, Selenastrum
capn'cornutum, toxicity test, after  an appropriate transformation, if
necessary, to meet the assumptions of  normality and homogeneity of variance,
may be analyzed by hypothesis testing  according to the flowchart in Figure 2.
Point estimates, such as the IC25 and  IC50, would also be appropriate in
analyzing algal  growth data.

9.5.5  ANALYSIS OF MORTALITY DATA

9.5.5.1  Mortality data are analyzed by Probit Analysis,  if  appropriate, or
other point estimation techniques (i.e., the Spearman-Karber Method, the
Trimmed Spearman-Karber Method,  or the Graphical Method)  (see Appendices I-L
and the discussion below).   The mortality data can also be  analyzed by
hypothesis testing,  after an arc sine  square root transformation (see Appendix
B-F),  according to the flowchart in Figure 2.              i    '
                                                                   •}
9.5.5.2  Mortality data from the daphnid, Ceriodaphnia dubia, survival  and
reproduction test are analyzed by Fisher's Exact Test (Appendix G) prior to
the analysis of the reproduction data.   The mortality data may also be
analyzed by Probit Analysis,  if appropriate or other methods (see
Subsection 9.5.5.1).
                                      51

-------
9.6  HYPOTHESIS TESTS

9.6.1  DUNNETT'S PROCEDURE

9.6.1.1  Dunnett's Procedure is used to determine the NOEC.  The procedure
consists of an analysis of variance (ANOVA) to determine the error term, which
is then used in a multiple comparison procedure for comparing each of;the
treatment means with the control mean, in a series of paired tests (see
Appendix C).  Use of Dunnett's Procedure requires at least three replicates
per treatment to check the assumptions of the test.  In cases where the
numbers of data points (replicates) for each concentration are not equal, a
t test may be performed with Bonferroni's adjustment for multiple comparisons
(see Appendix D), instead of using Dunnett's Procedure.              ;

9.6.1.2  The assumptions upon which the use of Dunnett's Procedure is
contingent are that the observations within treatments are normally
distributed, with homogeneity of variance.  Before analyzing the data, these
assumptions must be tested using the procedures provided in Appendix B.

9.6.1.3  If, after suitable transformations have been carried out, the
normality assumptions have not been met, Steel's Many-one Rank Test should be
used if there are four or more data points (replicates) per toxicant
concentration.  If the numbers of data points for each toxicant concentration
are not equal, the Wilcoxon Rank Sum Test with Bonferroni's adjustment should
be used (see Appendix F).

9.6.1.4  Some indication of the sensitivity of the analysis should be! provided
by calculating (1) the minimum difference between means that can be detected
as statistically significant, and  (2) the percent change from the control mean
that this minimum difference represents for a given test.

9.6.1.5  A step-by-step example of the use of Dunnett's Procedure is provided
in Appendix C.

9.6.2  T TEST WITH THE BONFERRONI ADJUSTMENT                     ,    '

9.6.2.1  At test with Bonferroni's adjustment is used as an alternative to
Dunnett's  Procedure when the number of replicates is not the same for all
concentrations.  This test sets an upper bound of alpha on the overall error
rate,  in contrast to Dunnett's  Procedure, for which the overall error rate is
fixed  at alpha.  Thus Dunnett's Procedure is a more powerful test.

9.6.2.2  The assumptions upon which the use of the t test with Bonferroni's
adjustment  is contingent are that  the observations within treatments are
normally distributed, with homogeneity of variance.  These assumptions must be
tested using the procedures provided  in Appendix B.

9.6.2.3  The estimate of the safe'  concentration derived from this tes't  is
reported in terms of the NOEC.  A  step-by-step example of the use of the
t test with Bonferroni's adjustment is provided in Appendix D.
                                       52

-------
 9.6.3   STEEL'S  MANY-ONE  RANK TEST                        ;

 9.6.3.1  Steel's  Many-one  Rank Test  is  a multiple  comparison  procedure  for
 comparing several  treatments with a  control.   This method  is  similar  to
 Dunnett's Procedure,  except  that  it  is  not  necessary  to  meet  the  assumption  of
 normality.   The data  are ranked,  and the analysis  is  performed  on the ranks
 rather  than  on  the data  themselves.   If the data are  normally or  nearly
 normally  distributed,  Dunnett's Procedure would be more  sensitive (would
 detect  smaller  differences between the  treatments  and control).   For  data that
 are  not normally  distributed,  Steel's Many-one Rank Test can  be much  more    .
 efficient (Hodges  and  Lehmann,  1956).

 9.6.3.2  It  is  necessary to  have  at  least four replicates per toxicant
 concentration to  use  Steel's test.   Unlike  Dunnett's  procedure, the
 sensitivity  of  this test cannot be stated in terms of the minimum difference
 between treatment  means  and  the control  mean that  can be detected as
 statistically significant.              .

 9.6.3.3  The estimate  of the safe concentration is reported as the NOEC.
 A step-by-step  example of  the  use of Steel's Many-one Rank Test is provided  in
 Appendix  E.                                               ;

 9.6.4   WILCOXON RANK SUM TEST  WITH THE  BONFERRONI  ADJUSTMENT

 9.6.4.1   The Wilcoxon  Rank Sum  Test  with  the Bonferroni  Adjustment is a
 nonparametric test for comparing  treatments with a control.   The  data are
 ranked  and the  analysis  proceeds  exactly  as in Steel's Test except that
 Bonferroni's adjustment  for  multiple  comparisons is used instead  of Steel's
 tables.   When Steel's  test can  be used  ('i.e., when  there are  equal numbers of
 data points per toxicant concentration),  it will be more powerful  (able to
 detect  smaller  differences as  statistically significant) than the  Wilcoxon
 Rank Sum  Test with Bonferroni's adjustment.

 9.6.4.2   The estimate  of the safe concentration is  reported as the NOEC.
 A step-by-step  example of  the use of the  Wilcoxon  Rank Sum Test with
 Bonferroni Adjustment  is provided in Appendix F.

 9.6.5   A  CAUTION IN THE USE OF HYPOTHESIS TESTING        '. '

 9.6.5.1   If in  the calculation of an NOEC by hypothesis testing, two tested
 concentrations  cause statistically significant adverse effects,  but an
 intermediate concentration did not cause  statistically significant effects,
 the results should be  used with extreme caution.

 9.7  POINT ESTIMATION TECHNIQUES

 9.7.1   PROBIT ANALYSIS                                    '•

 9.7.1.1   Probit Analysis is used to estimate the LCI,  LC50,  EC1,  or EC50 and
 the associated 95% confidence interval.   The analysis  consists of adjusting
 the data for mortality in the control, and then using  a maximum  likelihood
technique to estimate the parameters  of the  underlying log  tolerance
distribution, which is assumed to  have a particular shape.'

                                      53

-------
9.7.1.2   The assumption upon which the use of Probit Analysis is contingent
is a normal distribution of log tolerances.  If the normality assumption is
not met, and at least two partial mortalities are not obtained, Probit
Analysis should not be used.  It is important to check the results of IProbit?
Analysis to determine if use of the analysis is appropriate.  The chi-square
test for heterogeneity provides one good test of appropriateness of tfje
analysis.  The computer program (see Appendix I) checks the chi-square
statistic calculated for the data set against the tabular value, and provides
an error message if the calculated value exceeds the tabular value.

9.7.1.3  A discussion of Probit Analysis, and examples of computer program
input and output, are found in Appendix I.                            ,

9.7.1.4  In cases where Probit Analysis is not appropriate, the LC50 and
associated confidence interval may be estimated by the Spearman-Karber Method
(Appendix J) or the Trimmed Spearman-Karber Method (Appendix K).  If the test
results in 100% survival and 100% mortality in adjacent treatments (all or
nothing effect), the LC50 may be estimated using the Graphical Method
(Appendix L).

9.7.2  LINEAR INTERPOLATION METHOD                                    ;

9.7.2.1  The Linear Interpolation Method  (see Appendix M) is a procedure to
calculate a point estimate of the effluent or other toxicant concentration
[Inhibition Concentration,  (1C)] that causes a given percent reduction  (e.g.,
25%, 50%, etc.) in the reproduction or growth of the test organisms.  The
procedure was designed for general applicability in the analysis of data from
short-term chronic toxicity tests.

9.7.2.2  Use of the Linear  Interpolation Method is based on the assumptions
that the responses (1) are monotonically non-increasing (the mean response for
each higher concentration is less than or equal to the mean response for the
previous concentration),  (2) follow a piecewise linear response function, and
(3) are from a random, independent, and representative sample  of test data.
The assumption for piecewise linear response cannot be tested  statistically,
and no defined statistical  procedure is provided to test the assumption for
monotonicity.  Where the observed means are not strictly monotonic by
examination, they are adjusted by smoothing.  In cases where the responses at
the low toxicant concentrations  are much  higher than in the controls,jthe
smoothing process may result in  a large upward  adjustment in the control mean.
                                                                      L
9.7.2.3  The inability to test the monotonicity and piecewise  linear response
assumptions for this method makes it difficult  to assess when  the method is,
or is not, producing reliable results.  Therefore, the^ method  should be used
with caution when the results of a toxicity test approach an  "all or nothing"
response from one concentration  to the next in  the concentration series, and
when it appears that there  is a  large deviation from monotonicity.   See
Appendix M for a more detailed discussion  of the use of this method  and a
computer program available  for performing  calculations.
                                       54

-------
                                  SECTION 10              I

                              REPORT PREPARATION


   The toxicity data are reported, together with other appropriate data.  The
following general format and content are recommended for the report:

10.1  INTRODUCTION

    1.   Permit number
    2.   Toxicity testing requirements of permit          I
    3.   Plant location
    4.   Name of receiving water body
   '5.   Contract Laboratory (if the tests are performed under contract)
         a.  Name of firm                                  :
         b.  Phone number                                  !
         c.  Address

10.2  PLANT OPERATIONS

    1.   Product(s)
    2.   Raw materials                                    ',
    3.   Operating schedule
    4.   Description of waste treatment
    5.   Schematic of waste treatment
    6.   Retention time (if applicable)
    7.   Volume of waste flow (MGD,  CFS, GPM)
    8.   Design flow of treatment facility at  time of sampling

10.3  SOURCE OF EFFLUENT, RECEIVING  WATER, AND DILUTION WATER

    1.   Effluent Samples                                 I
         a.  Sampling point                                •
         b.  Collection dates and times                    I
         c.  Sample collection method
         d.  Physical and chemical data
         e.  Mean daily discharge on  sample collection date
         f.  Lapsed time from sample  collection to delivery
         g.  Sample temperature when  received at the laboratory

    2.   Receiving Water Samples                          ;
         a.  Sampling point
         b.  Collection dates and times
         c.  Sample collection method
         d.  Physical and chemical data
         e.  Streamflow (at 7Q10 and  at time of sampling)
         f.  Sample temperature when  received at the laboratory
         g.  Lapsed time from sample  collection to delivery
                                      55

-------
    3.   Dilution Water Samples                                       ;
         a. Source
         b. Collection date(s) and time(s)
         c. Pretreatment                                              ;
         d. Physical and chemical characteristics                     :

10.4  TEST METHODS

    1.   Toxicity test method used (title, number, source)
    2.   Endpoint(s) of test
    3.   Deviation(s) from reference method, if any, and the reason(s)
    4.   Date and time test started
    5.   Date and time test terminated
    6.   Type and volume of test chambers
    7.   Volume of solution used per chamber                          ;
    8.   Number of organisms per test chamber                         ;
    9.   Number of replicate test chambers per treatment
   10.   Acclimation of test organisms (temperature mean and range)   '
   11.   Test temperature (mean and range)
   12.   Specify if aeration was needed
   13.   Feeding frequency, arid amount and type of food               ,

10.5  TEST ORGANISMS                                                  ]

    1.   Scientific name and how determined
    2.   Age
    3.   Life stage
    4.   Mean length and weight  (where applicable)
    5.   Source                   .                                    :
    6.   Diseases and treatment  (where applicable)
    7.   Taxonomic key used for species identification                ;

10.6  QUALITY ASSURANCE                                               ;

    1.   Reference toxicant used routinely; source                    ;
    2.   Date and time of most recent reference toxicant test, test results,
         and current control  (cusum) chart
    3.   Dilution water used  in reference toxicant test
    4.   Results (NOEC or, where applicable, LOEC, LC50, EC50, IC25   !
         and/or  IC50)                                                 !
    5.   Physical and chemical methods used

10.7  RESULTS

    1.   Provide raw toxicity data in tabular form, including daily records of
         affected organisms in each concentration  (including controls), and
         plots of toxicity data
    2.   Provide table of LCSOs, NOECs, IC25, IC50, etc.
    3.   Indicate statistical methods used  to calculate endpoints
    4.   Provide summary table of physical  and chemical data
    5.   Tabulate QA data


                                      56

-------
10.8  CONCLUSIONS AND RECOMMENDATIONS                    i

    1.    Relationship between test endpoints and permit limits
    2.    Actions to be taken                             i
                                     57

-------
                                  SECTION 11

                                 TEST  METHOD

                     FATHEAD  MINNOW,  PIMEPHALES PROMELAS,
                       LARVAL SURVIVAL AND GROWTH TEST
                                 METHOD 1000.0

11.1  SCOPE AND APPLICATION

11.1.1  This method estimates the chronic toxicity of effluents and receiving
water to the fathead minnow,  Pimephales promelas, using newly hatched larvae
in a seven-day, static renewal test.  The effects include the synergistic,
antagonistic, and additive effects of all the chemical, physical, and
biological components which adversely affect the physiological and biochemical
functions of the test organisms.

11.1.2  Daily observations on mortality make it possible to also calculate
acute toxicity for desired exposure periods (i.e., 24-h, 48-h, 96-h LC50s).

11.1.3  Detection limits of the toxicity of an effluent or pure substance are
organism dependent.

11.1.4  Brief excursions in toxicity may not be detected using 24-h composite
samples.  Also, because of the long sample collection period involved in
composite sampling, and because the test chambers are not sealed, highly
degradable or highly volatile toxicants present in the source may not be
detected in the test.

11.1.5  This test method is commonly used in one of two forms: (1) a '
definitive test, consisting of a minimum of five effluent concentrations and a
control, and (2) a receiving water test(s), consisting of one or more
receiving water concentrations and a control.

11.2  SUMMARY OF METHOD

11.2.1  Fathead minnow, Pimephales promelas, larvae are exposed in a static
renewal system for seven days to different concentrations of effluent or to
receiving water.  Test results are based on the survival and weight of the
larvae.

11.3.   INTERFERENCES

11.3.1  Toxic substances may be  introduced by contaminants in dilution water,
glassware, sample hardware, and  testing equipment (see Section 5, Facilities,
Equipment and Supplies).

11.3.2  Adverse effects of low dissolved oxygen  (DO) concentrations, high
concentrations of suspended and/or dissolved solids, and extremes of pH,
alkalinity, or hardness, may mask the  presence  of toxic substances.
                                      58

-------
11.3.3  Improper effluent sampling and sample handling may adversely affect
test results (see Section 8, Effluent and Receiving Water Sampling, Sample
Handling, and Sample Preparation for Toxicity Tests).

11.3.4  Pathogenic and/or predatory organisms in the dilution water and
effluent may affect test organism survival and confound test results.

11.3.5  Food added during the test may sequester metals and other toxic
substances and confound test results.  Daily renewal of solutions, however,
will reduce the probability of reduction of toxicity caused by feeding.

11.4  SAFETY

11.4.1  See Section 3, Health and Safety.                !

11.5  APPARATUS AND EQUIPMENT

11.5.1  Fathead minnow and brine shrimp culture units -- see USEPA, 1985a and
USEPA, 1993b.  This test requires 180-360 larvae.  It is preferable to obtain
larvae from an in-house fathead minnow culture unit.  If it is not feasible to
culture fish in-house, embryos or newly hatched larvae can be shipped in well
oxygenated water in insulated containers.

11.5.2  Samplers -- automatic sampler, preferably with sample cooling
capability, that can collect a 24-h composite sample of 5 L.

11.5.3  Sample containers -- for sample shipment and storage (see Section 8,
Effluent and Receiving Water Sampling, Sample Handling, and Sample Preparation
for Toxicity Tests).

11.5.4  Environmental  chamber or equivalent facility with temperature control
(25 ฑ 16C).

11.5.5  Water purification system -- MILLIPORE MILLI-Qฎ, deionized water or
equivalent (see Section 5, Facilities, Equipment, and Supplies).

11.5.6  Balance -- analytical, capable of accurately weighing to 0.00001 g.

11.5.7  Reference weights, Class S -- for checking performance of balance.
Weights should bracket the expected weights of the weighing pans and the
expected weights of the pans plus fish.                  ;
                                                         i
11.5.8  Test chambers -- four (minimum of three) borosilicate glass or
non-toxic disposable plastic test chambers are required for each concentration
and control.  Test chambers.may be 1 L, 500 ml or 250 ml beakers, 500 ml
plastic cups, or fabricated rectangular (0.3 cm thick) glass chambers,' 15 cm x
7.5 cm x 7.5 cm.  To avoid potential contamination from the air and excessive
evaporation of test solutions during the test, the chambers should be covered
with safety glass plates or sheet plastic (6 mm thick).
                                      59

-------
11.5.9  Volumetric flasks and graduated cylinders -- Class A, borosilicate
glass or non-toxic plastic labware, 10-1000 ml for making test solutions. 5.10
11.5.10  Volumetric pipets -- Class A, 1-100 ml.
11.5.11  Serological pipets -- 1-10 ml, graduated.                    ;
11.5.12  Pipet bulbs and fillers -- PROPIPETฎ, or equivalent.         j
11.5.13  Droppers, and glass tubing with fire polished edges, 4 mm ID -- for
transferring larvae.
11.5.14  Wash bottles -- for rinsing small glassware and instrument electrodes
and probes.
11.5.15  Thermometers, glass or electronic, laboratory grade -- for measuring
water temperatures.                                                   i
11.5.16  Bulb-thermograph or electronic-chart type thermometers -- for
continuously recording temperature.                                   ;
11.5.17  Thermometers, National Bureau of Standards Certified (see USEPA
Method 170.1, USEPA, 1979b) --to calabrate laboratory themometers.
11.5.18  Meters, pH, DO, and specific conductivity -- for routine physical and
chemical measurements.                                                \
11.5.19  Drying oven -- 50-105ฐC range for drying larvae.
11.6  REAGENTS AND CONSUMABLE MATERIALS
11.6.1  Sample containers -- for sample shipment and storage (see Sectjion 8,
Effluent and Receiving Water Sampling, Sample Handling, and Sample Preparation
for Toxicity Tests).
11.6.2  Data sheets (one set per test) -- for recording data.
11.6.3  Vials, marked -- 18-24 per test, containing 4% formalin or 70% ethanol
to preserve larvae (optional).
11.6.4  Weighing boats, aluminum -- 18-24 per test.
11.6.5  Tape, colored -- for labelling test chambers.
11.6.6  Markers, waterproof -- for marking containers, etc.
11.6.7  Reagents for hardness and alkalinity tests -- see USEPA Methods 130.2
and 310.1, USEPA, 1979b.
11.6.8  Buffers, pH 4, pH 7, and pH 10 (or as per instructions of instrument
manufacturer) -- for instrument calibration (see USEPA Method 150.1,  USEPA,
1979b).
                                      60

-------
11.6.9  Specific conductivity standards -- see USEPA Method 120.1, USEPA,
1979b.

11.6.10  Membranes and filling solutions for DO probe (see USEPA Method 360.1,
USEPA, 1979b), or reagents -- for modified Winkler analysis.

11.6.11  Laboratory quality control samples and standards .-- for calibration
of the above methods.

11.6.12  Reference toxicant solutions (see Section 4, Quality Assurance).

11.6.13  Ethanol (70%) or formalin (4%) -- for use as a preservative for the
fish larvae.                                   .   .        '

11.6.14  Reagent water -- defined as distilled or deionized water that does
not contain substances which are toxic to the test organisms (see Section 5,
Facilities, Equipment, and Supplies).

11.6.15  Effluent, receiving water, and dilution water -- ;see Section 7,
Dilution Water; and Section 8,, Effluent and Receiving Water Sampling, Sample
Handling, and Sample Preparation for Toxicity Tests.

11.6.16  Brine Shrimp, Artemia, Nauplii -- for feeding cultures and test
organisms                                                 ;

11.6.16.1  Newly-hatched Artemia nauplii are used as food (see USEPA, 1993b)
for fathead minnow, Pimephales promelas, larvae in toxicity tests and frozen
brine shrimp and flake food are used in the maintenance of continuous stock
cultures.  Although there are many commercial sources of brine shrimp cysts,
the Brazilian or Colombian strains are currently preferred because the
supplies examined have had low concentrations of chemical residues and produce
nauplii of suitably small size.  For commercial sources of brine shrimp,
Artemia, cysts, see Table 2 of Section 5,* Facilities, Equipment, and Supplies
and Section 4, Quality Assurance.

11.6.16.2  Each new batch of brine shrimp, Artemia, cysts must be evaluated
for size (Vanhaecke and Sorgelloos, 1980, and Vanhaecke et al., 1980) and
nutritional suitability (see Leger et al., 1985; Leger et al., 1986) against
known suitable reference cysts by performing a side by side larval growth test
using the "new" and "reference" cysts.  The "reference" cysts used in the
suitability test may be a previously tested and acceptable batch of cysts, or
may be obtained from the Quality Assurance Branch, Environmental Monitoring
Systems Laboratory, Cincinnati, OH 45268; 513-569-7325.  A sample of
newly-hatched Artemia nauplii from each new batch of cysts should be
chemically analyzed.  The Artemia cysts should not be used if the
concentration of total organochlorine exceeds 0.15 ng/g wet weight or the
total concentration of organochlorine pesticides plus PCBs exceeds 0.30 //g/g
wet weight.   (For analytical methods see USEPA, 1982).
                                      61

-------
11.6.16.3 • Artemia nauplii are obtained as follows:

   1. Add 1 L of seawater, or a solution prepared by adding 35.0 g uniodized
      salt (NaCl) or artificial sea salts to 1 L deionized water, to a 2-1
      separatory funnel, or equivalent.
   2. Add 10 ml Artemia cysts to the separatory funnel and aerate for 24-h at
      27ฐC.  (Hatching time varies with incubation temperature and the
      geographic strain of Artemia used) (see USEPA, 1991b; USEPA, 1993b and
      ASTM, 1993).
   3. After 24 h, cut off the air supply in the separatory funnel.  Artemia
      nauplii are phototactic, and will concentrate at the bottom of the
      funnel if it is covered for 5-10 min.  To prevent mortality, do not
      leave the concentrated nauplii at the bottom of the funnel more than 10
      min without aeration.
   4. Drain the nauplii into a beaker or funnel fitted with a < 150 //tn Nitexฎ
      or stainless steel screen, and rinse with seawater, or equivalent,
      before use.

11.6.16.4  Testing Artemia nauplii as food for toxicity test organisms.

11.6.16.4.1  The primary criterion for acceptability of each new supply of
brine shrimp cysts is the ability of the nauplii to support good survival and
growth of the fathead minnow larvae (see Subsection 11.12).  The larvae used
to evaluate the suitability of the brine shrimp nauplii must be of the same
geographical origin, species, and stage of development as those used routinely
in the toxicity tests.  Sufficient data to detect differences in survival and
growth should be obtained by using three replicate test vessels, each
containing a minimum of 15 larvae, for each type of food.

11.6.16.4.2  The feeding rate and frequency, test vessels, volume of control
water, duration of the test, and age of the nauplii at the start of the test,
should be the same as used for the routine toxicity tests.            |

11.6.16.4.3  Results of the brine shrimp nutrition assay, where there are only
two treatments, can be evaluated statistically by use of a t test.  The "new"
food is acceptable if there are no statistically significant differences in
the survival and growth of the larvae fed the two sources of nauplii.

11.6.17  TEST ORGANISMS, FATHEAD MINNOWS, PIMEPHALES PROMELAS         <

11.6.17.1  Newly hatched fish less than 24 h old should be used for the test.
If organisms must be shipped to the testing site, fish up to 48 h old'may be
used, all hatched within a 24-h window.

11.6.17.2  If the fish are kept in a holding tank or container, most of the
water should be siphoned off to concentrate the fish.  The fish are th'en
transferred one at a time randomly to the test chambers until each chamber
contains ten fish.  Alternately, fish may be placed one or two at a time into
small beakers or plastic containers until they each contain five fish.  Three
(minimum of two) of these beakers/plastic containers are then assigned to
randomly-arranged control and exposure chambers.


                                      62

-------
11.6.17.2.1  The fish are transferred directly to the test vessels or
intermediate beakers/plastic containers, using a large-bore, fire-polished
glass tube (6 mm to 9 mm I.D. X 30 cm long) equipped with a rubber bulb, or a
large volumetric pipet with tip removed and fitted with a safety type bulb
filler.  The glass or plastic containers should only contain a small volume of
dilution water.

11.6.17.2.2  It is important to note that larvae should not be handled with a
dip net.  Dipping small fish with a net may result in damage to the fish and
cause mortality.                                         ;

11.6.17.3  The test is conducted with four (minimum of three) test chambers at
each toxicant concentration and control.  Fifteen (minimum of ten) embryos are
placed in each replicate test chamber.  Thus 60 (minimum of 30) fish are
exposed at each test concentration.

11.6.17.4  Sources of organisms

11.6.17.4.1  Fathead minnows, Pimephales promelas, may be obtained from
commercial biological supply houses.  Fish obtained from outside sources for
use as brood stock or in toxicity tests may not always be of suitable age and
quality.  Fish provided by supply houses should be guaranteed to be of (1) the
correct species, (2) disease free, (3) in the requested age range, and (4) in
good condition.  This can be done by providing the record of the date on which
the eggs were laid and hatched, and information on the sensitivity of
contemporary fish to reference toxicants.

11.6.17.5  Inhouse Sources of Fathead Minnows, Pimephales promelas

11.6.17.5.1  Problems in obtaining suitable fish from outside laboratories can
be avoided by developing an inhouse laboratory culture facility.  Fathead
minnows, Pimephales promelas., can be easily cultured in the laboratory from
eggs to adults in static, recirculating, or flow-through 'systems.  The larvae,
juveniles, and adult fish should be kept in 60 L (15 gal) or 76 L (20 gal)
rearing tanks supplied with reconstituted water, dechlorinated tap water, or
natural water.  The water should be analyzed for toxic metals and organics
quarterly (see Section 4, Quality Assurance).            ',

11.6.17.5.1.1  If a static or recirculating system is used, it is necessary to
equip each tank with an outside activated carbon filter system, similar to
those sold for tropical fish hobbyists (or one large activated carbon filter
system for a series of tanks) to prevent the accumulation of toxic metabolic
wastes (principally nitrite and ammonia) in the water.

11.6.17.5.2  Flow-through systems require large volumes of water and may not
be feasible in some laboratories.  The culture tanks should be shielded from
extraneous disturbances using opaque curtains, and should be isolated from
toxicity testing activities to prevent contamination.

11.6.17.5.3  To avoid the possibility of inbreeding of the inhouse brood
stock, fish from an outside source should be introduced yearly into the
culture unit.

                                      63

-------
 11.6.17.5.4  Dissolved oxygen -- The DO concentration in the culture tanks
 should be maintained near saturation, using gentle aeration with 15 cm air
 stones if necessary.  Brungs (1971), in a carefully controlled long-term
 study, found that the growth of fathead minnows was reduced significantly at
 all  dissolved oxygen concentrations below 7.9 mg/L.  Soderberg (1982)
 presented an analytical  approach to the re-aeration of flowing water for
 culture systems.

 11.6.17.5.5  Culture Maintenance

 11.6.17.5.5.1  Adequate procedures for culture maintenance must be followed to
..avoid poor water quality in the culture system.  The spawning and brood stock
 culture tanks should be kept free of debris (excess food,  detritus, waste,
 etc.) by siphoning the accumulated materials (such as dead brine shrimp
 nauplii or cysts) from the bottom of the tanks daily with  a glass siphon tube
 attached to a plastic hose leading to the floor drain.   The tanks are  more
 thoroughly cleaned as required.  Algae, mostly diatoms  and green algae,
 growing on the glass of the spawning tanks are left in  place,  except for the
 front of the tank, which is kept clean for observation.   To avoid excessive
 build-up of algal growth,  the walls of the tanks are periodically scraped.
 The  larval culture tanks are cleaned once or twice a week to reduce the mass
 of fungus growing on the bottom of the tank.                          ;

 11,6.17.5.5.2  Activated.charcoal  and floss in the tank filtration systems
 should be changed weekly,  or more often if needed.  Culture water may  be
 maintained by preparation  of reconstituted water or use of dechlorinated tap
 water.  Distilled or deionized water is added as needed to compensate  for
 evaporation.

 11.6.17.5.5.3  Before new fish are placed in tanks, salt deposits are  removed
 by scraping or with 5% acid solution, the tanks are washed with detergent,
 sterilized with a hypochlorite solution,  and rinsed well  with  hot tap  water
 and  then with laboratory water.                                       :

 11.6.17.5.6  Obtaining Embryos for Toxicity Tests                     ;

 11.6.17.5.6.1  Embryos can be shipped to the laboratory from an outside source
 or obtained from adults  held in the laboratory as described below.     ;

 11.6.17.5.6.2  For breeding tanks,  it is convenient to  use 60  L (15 gal)  or
 76 L (20 gal) aquaria.  The spawning unit is designed to simulate conditions
 in nature conducive to spawning,  such as water temperature and photoperiod.
 Spawning tanks must be held at a temperature of 25 ฑ 2ฐC.   Each aquarium is
 equipped with a heater,  if necessary, a continuous filtering unit,  and
 spawning substrates.   The  photoperiod for the culture system should be1
 maintained at 16 h light and 8 h darkness.   For the spawning tanks,  this
 photoperiod must be rigidly controlled.   A convenient photoperiod is 5:00 AM
 to 9:00 PM.   Fluorescent lights should be suspended about  60 cm above  the
 surface of the water in  the brood  and larval  tanks.  Both  DURATESTฎ and cool-
 white fluorescent lamps  have been  used,  and produce similar results.   A"
 illumination level  of 50 to 100 ft-c is adequate.
                                       64

-------
11.6.17.5.6.3  To simulate the natural spawning environment, it is necessary
to provide substrates (nesting territories) upon which the eggs can be
deposited and fertilized, and which are defended and cared for by the males.
The recommended spawning substrates consist of inverted half-cylinders,
7.6 cm x 7.6 cm (3 in x 3 in) of Schedule 40 PVC pipe.  The substrates should
be placed equi-distant from each other on the bottom of the tanks.

11.6.17.5.6.4  To establish a breeding unit, 15-20 pre-spawning adults six to
eight months old are taken from a "holding" or culture tank and placed in a
76-1 spawning tank.  At this point, it is not possible to distinguish the
sexes.  However, after less than a week in the spawning tank, the breeding
males will develop their distinct coloration and territorial behavior, and
spawning will begin.  As the breeding males are identified, all but two are
removed, providing a final ratio of 5-6 females per male.  The excess spawning
substrates are used as shelter by the females.            i

11.6.17.5.6.5  Sexing of the fish to  ensure a correct female/male ratio in
each tank can be a problem.  However, the task usually becomes easier as
experience is gained (Flickinger, 1966).  Sexually mature females usually have
large bellies and a tapered snout.  The sexually mature males are usually
distinguished by their larger overall size, dark vertical color bands, and the
spongy nuptial tubercles on the snout.  Unless the males  exhibit these
secondary breeding characteristics, no reliable method has been found to
distinguish them from females.  However, using the coloration of the males and
the presence  of enlarged urogenital structures and other  characteristics of
the females,  the correct selection of the  sexes can usually be achieved by
trial and error.

11.6.17.5.6.6 Sexually  immature males are  usually recognized by their
aggressive behavior and  partial banding.   These undeveloped males must be
removed  from  the spawning tanks because they will eat the eggs and  constantly
harass the mature males, tiring them  and reducing the fecundity of  the
breeding  unit.  Therefore, the fish in the  spawning tanks must be carefully
checked  periodically for extra males.

11.6.17.5.6.7 A breeding unit should remain  in their spawning tank about four
months.   Thus, each brood tank or  unit is  stocked with new  spawners about
three times  a year.  However, the  restocking  process  is  rotated so  that at  any
one time the  spawning tanks  contain different  age groups  of  brood fish.

11.6.17.5.6.8  Fathead minnows spawn  mostly in the early  morning  hours.  They
should not be disturbed  except for  a  morning  feeding  (8:00  AM) and  daily
examination  of  substrates for eggs  in late morning or early  afternoon.   In
nature,  the  male  protects, cleans,  and aerates the eggs  until  they  hatch.
 In, the  laboratory,  however,  it  is  necessary to remove the eggs from the tanks
to prevent them  from  being eaten  by the  adults,  for ease  of handling,  for
purposes of  recording embryo count  and hatchability,  and  for  the  use of the
newly hatched young fish for toxicity tests.

 11.6.17.5.6.9  Daily,  beginning  six to eight  hours  after 'the  lights are turned
on (11:00 AM -  1:00 PM), the substrates  in the spawning  tanks are each lifted
carefully and inspected  for  embryos.   Substrates  without embryos  are

                                       65

-------
 immediately returned to the spawning  tank.   Those  with  embryos  are  immersed  in
 clean  water in  a collecting tray,  and replaced  with  a clean  substrate.   A
 daily  record is maintained  of each spawning  site and the  estimated  number  of
 embryos  on  the  substrate.

 11.6.17.5.6.10   Three different  methods  are  described for embryo  incubation.

    1.  Incubation of Embryos on the Substrates:  Several  (2-4) substrates
 are placed  on end in a circular  pattern  (with the  embryos on the  innerside)  in
 10  cm  of water  in a tray.   The tray is then  placed in a constant  temperature
 water  bath,  and the embryos are  aerated  with a  2.5 cm airstone  placed in the
 center of the circle.   The  embryos are examined daily,  and the  dead and
 fungused embryos are counted,  recorded,  and  removed  with  forceps.   At an
 incubation  temperature of 25ฐC,  50% hatch occurs in  five  days.  At  22ฐC
 embryos  incubated on aerated tiles require 7 days  for 50% hatch.

    2.  Incubation of Embryos in a Separatorv  Funnel:  The  embryos  are removed
 from the substrates with a  rolling action of the index  finger ("rolled off")
 (Cast  and Brungs,  1973), their total  volume  is  measured,  and the  number  of
 embryos  is  calculated  using a  conversion factor of approximately  430
 embryos/mL.  The embryos are incubated in about 1.5  L of  water  in a 2 L
 separatory  funnel  maintained in  a  water  bath.   The embryos are  stirred in  the
 separatory  funnel  by bubbling  air  from the tip  of  a  plastic micro-pipette
 placed at the bottom,  inside the separatory  funnel.  During the first two
 days,  the embryos  are  taken from the  funnel  daily, those  that are dead and
 fungused are removed,  and those  that  are alive  are returned to  the  separatory
 funnel in clean  water.  The embryos hatch in four  days  at a temperature  of
 25ฐC.  However,  usually on  day three  the eyed embryos are removed from:the
 separatory  funnel  and  placed in  water in a plastic tray and gently  aerated
 with an  air  stone.   Using this method, the embryos hatch  in five days.;
 Hatching time is  greatly influenced by the amount  of agitation  of the embryos
 and the  incubation  temperature.  If on day three the embryos are transferred
 from the separatory funnel  to  a  static, unaerated  container, a  50%  hatch will
 occur  in six days  (instead  of  five) and a 100%  hatch will  occur in  seven days.
 If the culture  system  is operated  at  22ฐC, embryos incubated on aerated  tiles
 require  seven days  for 50%  hatch.

   3.  Incubation  in  Embryo  Incubation  Cups:  The embryos  are "rolled off" the
 substrates, and the  total number is estimated by determining the volume.  The
 embryos  are then  placed in  incubation  cups attached  to a  rocker arm  assembly
 (Mount,  1968).  Both flow-through  and  static renewal  incubation have been
 used.  On day one,  the embryos are removed from the  cups  and those that  are
dead and fungused  are removed.  After  day one only dead embryos are  removed
from the cups.  During the  incubation  period, the  eggs are examined daily for
viability and fungal growth, until  they hatch.   Unfertilized eggs, and eggs
that have become  infected by fungus,  should be  removed with forceps using a
table top magnifier-illuminator.    Non-viable eggs  become  milky and opaque,  and
are easily recognized.  The  non-viable eggs are very susceptible to  fungal
infection,  which may then spread throughout the egg  mass.  Removal of fungus
should be done quickly, and  the substrates should  be returned to the
incubation tanks as rapidly  as possible so that the  good  eggs are not damaged
by desiccation.   Hatching takes four to five days  at an optimal  temperature of

                                       66

-------
25ฐC.  Hatching can be delayed several (two to four) days by incubating at
lower temperatures.  A large plastic tank receiving recirculating water from a
temperature control unit, can be used as a water bath for incubation of
embryos.                                                 [

11.6.17.5.6.11  Newly-hatched larvae are transferred daily from the egg
incubation apparatus to small rearing tanks, using a large bore pipette, until
the hatch is complete.  New rearing tanks are set up on a daily basis to
separate fish by age group.  Approximately 1500 newly hatched larvae are
placed in a 60-L (15 gal) or 76-L (20 gal) all-glass aquarium for 30 days.
A density of 150 fry per liter is suitable for the first four weeks.  The
water temperature  in the rearing tanks is allowed to follow ambient laboratory
temperatures of 20-25ฐC, but sudden, extreme variations in temperature must be
avoided.

11.6.17.5.7  Food  and Feeding                            \

11.6.17.5.7.1  The amount of food and feeding schedule affects both growth and
egg  production.  The spawning fish  and pre-spawners  in holding tanks usually
are  fed all the adult frozen brine  shrimp and tropical fish flake food or dry
commercial fish food  (No. 1 or No.  2 granules) that  they can eat  (ad libitum)
at the  beginning of the work day and  in  the late afternoon  (8:00 AM and 4:00
PM).  The fish are fed twice a day  (twice a day with dry food and once a day
with  adult shrimp) during the week  and once a day on weekends.

11.6.17.5.7.2  Fathead minnow larvae  are fed  freshly-hatched brine  shrimp
(Artemia) nauplii  twice  daily until they are  four weeks old.  Utilization of
older (larger) brine  shrimp nauplii may  result  in starvation of the young fish
because they  are unable  to  ingest the larger  food organisms  (see
Subsection 11.6.16 or USEPA,  1993b  for  instructions  on the  preparation of
brine shrimp  nauplii).

11.6.17.5.7.3  Fish older  than  four weeks are fed frozen brine shrimp  and
commercial fish  starter  (#1  and  #2),  which  is ground fish meal enriched with
vitamins.  As  the  fish grow,  larger pellet  sizes are used,  as appropriate.
 (Starter,  No.  1  and N. 2 granules,  U.S.  Fish  and Wildlife Service  Formulation
Specification  Diet SD9-30,  can  be obtained  from Zeigler  E!ros.,  Inc.,  P.O. Box
90,  Gardners,  PA 17324).   Newly  hatched  brine shrimp nauplii, and  frozen  adult
brine shrimp  (San  Francisco Bay  Brand)  are  fed  to the  fish  cultures  in  volumes
based on  age,  size,  and  number  of fish  in the tanks.

 11.6.17.5.7.4  Fish  in  the larval tanks  (from hatch to 30 days  old)  are  fed
 commercial  starter fish  food  at  the beginning and  end  of the work day,  and
 newly hatched brine  shrimp nauplii  (from the  brine  shrimp culture unit)  once  a
 day, usually  mid-morning and  mid-afternoon.              ;

 11.6.17.5.7.5' Attempts  should  be made  to avoid introducing Artemia cysts and
 empty shells  when  the brine shrimp  nauplii  are  fed  to  the  fish  larvae.   Some
 of the mortality of the  larval  fish observed  in cultures could  be caused  from
 the ingestion of these materials.                                         •
                                       67

-------
 11.6.17.5.8  Disease Control

 11.6.17.5.8.1  Fish are observed daily for abnormal  appearance or behavior.
 Bacterial  or fungal infections are the most common diseases encountered.
 However,  if normal  precautions are taken,  disease outbreaks will  rarely,  if
 ever,  occur.   Hoffman and Mitchell  (1980)  have put together a list of isome
 chemicals  that have been used commonly for fish diseases  and pests.

 11.6.17.5.8.2  In  aquatic culture systems  where filtration  is utilized, the
 application of certain antibacterial  agents should be  used  with caution.
 A  treatment with a  single dose of antibacterial  drugs  can interrupt  nitrate
 reduction  and stop  nitrification for  various periods of time,  resulting in
 changes  in pH,  and  in ammonia,  nitrite and nitrate concentrations (Collins et
 al.,  1976).   These  changes could cause the death of the culture organisms.

 11.6.17.5.8.3  Do not transfer equipment from one tank to another without
 first  disinfecting  tanks and  nets.  If an  outbreak of  disease occurs, any
 equipment,  such as  nets,  airlines,  tanks,  etc.,  which  has been exposed to
 diseased fish should be disinfected with sodium hypochlorite.   Also  to avoid
 the contamination of cultures or spread of disease, each  time nets are used to
 remove live  or  dead fish from tanks,  they  are first sterilized with  sodium
 hypochlorite  or formalin,  and rinsed  in hot tap  water.  Before a  new lot of
 fish is transferred to culture  tanks,  the  tanks  are cleaned and sterilized as
 described  above.

 11.6.17.5.8.4  It is recommended that  chronic toxicity tests  be performed
 monthly with  a  reference toxicant.  Newly  hatched  fathead minnow  larvae less
 than 24 h  old are used to  monitor the  chronic toxicity of the  reference
 toxicant to the test fish  produced  by  the  culture  unit (see Section  4, Quality
 Assurance).

 11.6.J7.5.9   Record  Keeping

 11.6.17.5.9.1   Records,  kept  in  a bound notebook,  include:  (1)  type  of food
 and time of feeding  for  all fish  tanks; (2)  time  of examination of the tiles
 for embryos,  the estimated number of embryos  on  the tile, and  the  tile:
 position number; (3)  estimated number  of dead  embryos  and embryos  with fungus
 observed during the  embryonic development  stages;  (4)  source of all  fish;  (5)
daily observation of the condition  and  behavior  of the  fish; and  (6J~ dates and
results of reference  toxicant tests performed  (see Section  4,  Quality
Assurance).                                                           ;
                                                                      i
 11.7  EFFLUENT AND RECEIVING WATER COLLECTION, PRESERVATION, AND STORAGE

11.7.1  See Section 8, Effluent  and Receiving Water Sampling, Sample Handling,
and Sample Preparation for Toxicity Tests.

11.8  CALIBRATION AND STANDARDIZATION

11.8.1  See Section 4, Quality Assurance.
                                      68

-------
11.9  QUALITY CONTROL

11.9.1  See Section 4, Quality Assurance.

11.10  TEST PROCEDURES

11.10.1  TEST SOLUTIONS

11.10.1.1  Receiving Waters

11.10.1.1.1  The sampling point is determined by the objectives of the test.
Receiving water toxicity is determined with samples used directly as collected
or after samples are passed through a 60 pm NITEXฎ filter and compared without
dilution, against a control.  Using four replicate chambers per test, each
containing 250 mL, and 400 mL for chemical analyses, would require
approximately 1.5 L or more of sample per test per day.

11.10.1.2  Effluents

11.10.1.2.1  The selection of the effluent test concentrations should be based
on the objectives of the study.  A dilution factor of 0.5 is commonly used.
A dilution factor of 0.5 provides precision of ฑ 100%, and testing of
concentrations between 6.25% and 100% effluent using only five effluent
concentrations (6.25%, 12.5%, 25%, 50%,  and 100%).  Test precision shows
little improvement as the dilution factor is  increased beyond 0.5, and
declines rapidly  if  a smaller dilution factor is used.  Therefore, USEPA
recommends the use of the > 0.5 dilution factor.

11.10.1.2.2   If the  effluent  is known or suspected to be highly toxic, a lower
range of effluent concentrations should  be used  (such as 25%, 12.5%, 6.25%,
3.12%, and 1.56%).   If a high rate of mortality  is observed during the first
1 to 2 h of  the test, additional dilutions should be added at the lower range
of  effluent  concentrations.

11.10.1.2.3   The  volume  of  effluent  required  for daily renewal of four
replicates per concentration, each containing 250 ml of test  solution, is
approximately 2.5 L.  Sufficient test  solution  (approximately  1500 ml) is
prepared  at  each  effluent  concentration  to provide  400 ml  additional volume
for chemical  analyses at the  high, medium, and  low  test concentrations.
 If  the sample is  used for  more  than  one  daily renewal  of test  solutions, the
volume must  be  increased proportionately.

 11.10.1.2.4   Tests  should  begin as  soon  as possible,  preferably within 24  h of
 sample collection.   The  maximum holding  time  following retrieval  of  the  sample
 from the sampling device should not  exceed 36 h for off-site  toxicity  tests
 unless permission is granted  by the  permitting  authority.   In no  case  should
 the sample be used for the first  time  in a test more than  72  h  after sample
 collection (see  Section  8,  Effluent  and  Receiving  Water  Sampling,  Sample
 Handling,  and Sample Preparation  for Toxicity Tests).

 11.10.1.2.5  Just prior to test initiation  (approximately  1  h)  the  temperature
 of sufficient quantity of the sample to  make the test solutions  should be

                                       69

-------
 adjusted to the test temperature and maintained at that temperature during the
•addition of dilution water.

 11.10.1.2.6  The DO of the test solutions should be checked prior to the test
 initiation.  If any of the solutions are supersaturated with oxygen or any
 solution has a DO concentration below 4.0 mg/L, all of the solutions and the
 control must be gently aerated.

 11.10.1.3  Dilution Water

 11.10.1.3.1  Dilution water may be uncontaminated receiving water, a standard
 synthetic (reconstituted) water, or some other uncontaminated natural  water
 (see Section 7, Dilution Water).

 11.10.2  START OF THE TEST                                             '

 11.10.2.1  Label  the test chambers with a marking pen.   Use of color-coded
 tape to identify  each treatment and replicate is helpful.   A minimum of five
 effluent concentrations and a control  are used for each effluent test.  : Each
 treatment (including the control)  should have four (minimum of three)
 replicates.

 11.10.2.2  Tests  performed in laboratories  that have  in-house fathead  minnow
 breeding cultures  should use  larvae less than 24 h old.  When eggs or  larvae
must be shipped to the  test site from  a remote location, it may be necessary
to use larvae  older than 24 h because  of the difficulty in  coordinating.test
organism shipments with field operations.   However, in  the  latter case,  the
larvae should  not  be more than 48  h old at  the start  of the test and should
all  be within  24  h of the same age.

11.10.2.3  Randomize the position  of test chambers  at the beginning  of  the
test (see Appendix A).   Maintain the chambers in this configuration  throughout
the  test.   Preparation  of a position chart may be  helpful.

11.10.2.4 The  larvae are pooled and placed  one  or  two  at a time into each
randomly arranged  test  chamber or  intermediate container in sequential  order,
until  each chamber  contains 15 (minimum of  10)  larvae,  for  a  total of
60 larvae (minimum  of 30)  for each  concentration  (see Appendix  A).   The  test
organisms should come from a  pool of larvae  consisting  of at  least three
separate spawnings.  The  amount  of water added to the chambers  when
transferring the larvae  should be kept  to a  minimum to  avoid  unnecessary
dilution  of the test concentrations.

11.10.2.4.1  The chambers  may be placed  on a  light table to facilitate
counting the larvae.

11.10.3  LIGHT, PHOTOPERIOD,  AND TEMPERATURE

11.10.3.1  The light quality  and intensity should be at ambient  laboratory
levels, which is approximately 10-20 nE/nr/s, or 50 to 100  foot candles '
(ft-c), with a photoperiod  of  16 h of light and 8 h of darkness.  The water
temperature in the test chambers should be maintained  at 25 ฑ 1ฐC.

                                      70                                ;

-------
11.10.4  DISSOLVED OXYGEN (DO) CONCENTRATION

11.10.4.1  Aeration may affect the toxicity of effluents and should be used
only as a last resort to maintain satisfactory DO concentrations.  The DO
concentrations should be measured in the new solutions at the start of the
test (Day 0) and before daily renewal of the test solutions on subsequent
days.  The DO concentrations should not fall below 4.0 mg/L (see Section 8,
Effluent and Receiving Water Sampling, Sample Handling, and Sample Preparation
for Toxicity Tests).  If it is necessary to aerate, all concentrations and the
control should be aerated.  The aeration rate should not exceed 100
bubbles/min, using a pipet with an orifice of approximately 1.5 mm, such as a
1-mL, KIMAXฎ serological pipet, No. 37033, or equivalent.  Care should be
taken to ensure that turbulence resulting from aeration does not cause undue
physical stress to the fish.                              ',

11.10.5  FEEDING

11.10.5.1  The fish in each test chamber are fed 0.1 g (approximately 700 to
1000) of a concentrated suspension of newly hatched (less than 24-h old) brine
shrimp nauplii three times daily at 4-h intervals or, as a minimum, 0.15 g are
fed twice daily at an interval of 6 h.  Equal amounts of riauplii must be added
to each replicate chamber to reduce variability in larval weight.  Sufficient
numbers of nauplii should be provided to assure that some remain alive in the
test chambers at the next feeding, but not in excessive amounts which will
result in depletion of DO below acceptable levels  (below 4.0 mg/L).

11.10.5.2  The feeding schedule will depend on when the test solutions are
renewed.  If the test is initiated after 12:00 PM, the larvae may be fed only
once the first day.  On following days, the larvae normally would be fed at
the beginning of the work day, at least 2 h before test solution renewal, and
at the end of the work day, after test solution renewal.  However, if the test
solutions are changed at the beginning of the work day, the first feeding
would be after test solution renewal in the morning, and the remaining
feeding(s) would be at the appropriate intervals.  The larvae are not fed
during the final 12 h of the test.                        ,

11.10.5.3  The nauplii should  be rinsed with freshwater to remove salinity
before use  (see USEPA, 1993b).  At feeding time pipette about 5 mL (5 g) of
concentrated newly hatched brine shrimp nauplii into a 120 mesh nylon net or
plastic cup with nylon mesh bottom.  Slowly run freshwater through the net or
rinse by immersing the cup in  a container of fresh water several times.
Resuspend the brine shrimp in  10 mL  of fresh water in  a 30 ml beaker or simply
set the cup of washed brine shrimp in  % inch of fresh water so that the cup
contains about 10 mL of water.  Allow the container to set for a minute or two
to allow dead nauplii and empty cysts to settle or float to the surface before
collecting the brine shrimp from just below the surface  in a pipette for
feeding.  Distribute 2 drops  (0.1 g) of the brine  shrimp to each test chamber.
If the survival rate in any test chamber falls below 50%, reduce the feeding
in that chamber to  1 drop of  brine shrimp at each  subsequent feeding.
                                       71

-------
 11.10.6  OBSERVATIONS DURING THE TEST

 11.10.6.1  Routine Chemical  and Physical  Determinations               •

 11.10.6.1.1  DO is measured  at the beginning  and  end  of  each  24-h  exposure
 period in at least one test  chamber at  each test  concentration  and in the
 control.                                                              I

 11.10.6.1.2  Temperature  and pH are measured  at the end  of  each  24-h exposure
 period in at least one test  chamber at  each test  concentration  and in the
 control.   Temperature should also be monitored continuously or  observed  and
 recorded  daily for at least  two locations  in  the  environmental  control system
 or  the samples.   Temperature should be  measured in a  sufficient  number of test
 vessels at least at the end  of the test to determine  the temperature variation
 in  the environmental  chamber.

 11.10.6.1.3  The pH is measured in the  effluent sample each day  before new
 test  solutions are made.

 11.10.6.1.4  Conductivity, alkalinity and  hardness are measured  in each  new
 sample (100% effluent or  receiving water)  and in  the  control.

 11.10.6.1.5  Record all the  measurements on the data  sheet  (Figure 1).

 11.10.6.2  Routine Biological  Observations

 11.10.6.2.1  The number of live larvae  in  each test chamber are  recorded daily
 (Figure 2),  and  the dead  larvae are discarded.

 11.10.6.2.2  Protect  the  larvae from unnecessary  disturbance during the  test
 by  carrying out  the daily test  observations, solution renewals,  and removal of
 dead  larvae,  carefully.   Make  sure  the  larvae remain  immersed during the
 performance of these  operations.

 11.10.7   DAILY CLEANING OF TEST CHAMBERS                              :

 11.10.7.1   Before  the  daily  renewal  of  test solutions, uneaten and dead
Artemia,  dead  fish  larvae, and  other  debris are removed  from the bottom of the
test  chambers with  a  siphon  hose.   Alternately,  a large  pipet (50 mL) fitted
with  a rubber  bulb  can  be used.   Because of their small  size during the first
few days  of the  tests,  larvae are easily drawn into the  siphon tube or pipet
when  cleaning the test  chambers.  By  placing the test chambers on  a light box,
inadvertent  removal of  larvae can be  greatly reduced because they can be more
easily seen.   If the water siphoned from the test chambers  is collected in a
white plastic tray, the larvae  caught up in the siphon can be retrieved and
returned  to the chambers.   Any  incidence of removal  of live larvae from the
test chambers during cleaning,  and  subsequent return to the chambers, should
be noted  in the records.
                                      72

-------
Discharger:
Location: _
Analyst:
Dates:
Day i
Cone:
Temp.
D.O. Initial
Final
pH Initial
Final
Al kal initv
Hardness
Conductivity
Chi orine

1










2










3










4










5










6










7





.




Remarks










Day
Cone:
Temp.
D.O. Initial
Final
pH Initial
Final
Al kal initv 	
Hardness
Conductivity
Chlorine

1










2










3










4










5










6










7










Remarks










                                     Day
Cone:
Temp. 	
D.O. Initial
Final
pH Initial
Final
Al kal initv 	
Hardness
Conductivity
Chlorine

1










2










3










4










5










6










7


i






!
Remarks










   Figure 1.  Data form for the fathead minnow, Pimephales promelas,  larval
              survival and growth test.  Routine chemical and  physical
              determinations.
                                      73

-------
Discharger:
Location:
Analyst:
Dates:
                                    Day
Cone:
Terno.
D.O. Initial
Final
oH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

1










2










3










4










5










6










7










Remarks










                                   Day
Cone:
Terno.
D.O. Initial
Final
oH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

1










2










3










4










5










6










7










Remarks










                                   Day
Cone:
Terno.
D.O. Initial
Final
pH Initial •
Final
Alkalinity
Hardness
Conductivity
Chlorine

1










2










3










4










5










6










7










Remarks










 Figure 1.  Data form for the fathead minnow, Pimephales promelas,  larval
            survival and growth test.   Routine chemical and physical
            determinations  (CONTINUED).                             :
                                   74

-------
Discharger:
Location:
        Dates:
        Analyst:
                         No. Surviving Organisms
 Cone:   Rep.
Day
No.
Control :



Cone:



Cone:



Cone:



Cone:



Cone:



Cone:



Cone:



1
































2
































3
































4
































5
































6
































7
































Remarks
































Comments
    Figure 2.  Mortality data for the fathead minnow, Pimephales promelas,
               larval survival and growth test.
                                    75

-------
11.10.8  TEST SOLUTION RENEWAL

11.10.8.1  Freshly prepared solutions are used to renew the tests daily
immediately after cleaning the test chambers.  For on-site toxicity studies,
fresh effluent or receiving water samples should be collected daily, and no
more than 24 h should elapse between collection of the samples and their use
in the tests (see Section 8, Effluent and Receiving Water Sampling, Sample
Holding, and Sample Preparation for Toxicity Tests).  For off-site tejsts, a
minimum of three samples are collected, preferably on days one, three, and
five.  Maintain the samples in the refrigerator at 4ฐC until used.

11.10.8.2  For test solution renewal, the water level in each chamber is
lowered to a depth of 7 to 10 mm, which leaves 15 to 20% of the test solution.
New test solution (250 mL) should be added slowly by pouring down the side of
the test chamber to avoid excessive turbulence and possible injury to the
larvae.

11.10.9  TERMINATION OF THE TEST                                     !

11.10.9.1  The test is terminated after seven days of exposure.  At test
termination, dead larvae are removed and discarded.  The surviving larvae in
each test chamber (replicate) are counted and immediately prepared as a group
for dry weight determination, or are preserved as a group in 70% ethanol or
4% formalin.  Preserved organisms are dried and weighed within 7 days.  For
safety, formalin should be used under a hood.

11.10.9.2  For immediate drying and weighing, place live larvae onto !a 500 /j,m
mesh screen in a large beaker to wash away debris that might contribute to the
dry weight.  Each group of larvae is rinsed with deionized water to remove
food particles, transferred to a tared weighing boat that has been properly
labeled, and dried at 60ฐC, for 24 h or at 100ฐC for a minimum of 6 h.
Immediately upon removal from the drying oven, the weighing boats are placed
in a dessicator until weighed, to prevent the absorption of moisture from the
air.  All weights should be measured to the nearest 0.01 mg and recorded on
data sheets (Figure 3).  Subtract tare weight to determine the dry weight of
the larvae in each replicate.  For each test chamber, divide the final dry
weight by the number of original larvae in the test chamber to determine the
average individual dry weight and record on the data sheet (Figure 3)j.  For
the controls, also calculate the mean weight per surviving fish in the test
chamber to evaluate if weights met test acceptability criteria (See
Section 11.11).  Average weights should be expressed to the nearest 0.001 mg.

11.10.9.3  Prepare a summary table as illustrated in Figure 4.       .

11.11  SUMMARY OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA    i

11.11.1  A summary of test conditions and test acceptability criteria! is
presented in Table 1.
                                      76

-------
                                                                                1


..
0
0
 0)










lง


















?i^
co •*- co
O 01 — '
y) o flj
•— o c
Q -1 <
0)
o at
 i- E
•M O> CD *-*
O 3--
t—

*J c o
O) TO CO ^^
m 2 > 01
cu t_ E
X t- co ^
c_ to — >
Q -H
"o ^
ซฃ • 1_ O)
•I-1 03 E
D) •*-* v^
'3
a. o

                                                     S-
                                                     10
                                                     I
                                                     O

                                                     Q.
                                                        CO
                                                     O)

                                                     •ป->  tC
                                                      ns
                                                     13 i—
                                                         03
                                                      05 >
                                                     CO
                                                      S-
                                                      13
                                                      O>
77

-------
Discharger:

Location:
 Test Dates:

	  Analyst:
TREATMENT
NO. LIVE LARVAE
SURVIVAL
101 "\
(A)
MEAN DRY WGT OF
LARVAE (MG)
ฑ SD
TEMPERATURE
RANGE (ฐC)
DISSOLVED
OXYGEN RANGE
(MG/L)
HARDNESS
CONDUCTIVITY
CONTROL







































i
'
1





COMMENTS:
Figure 4.   Summary data for the fathead minnow, Pimephales promelas,
            larval survival and growth test.
                                    78

-------
   TABLE 1.  SUMMARY OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA
             FOR FATHEAD MINNOW, PIMEPHALES PROMELAS, LARVAL SURVIVAL AND
             GROWTH TOXICITY TESTS WITH EFFLUENTS AND RECEIVING WATERS
 1. Test type:                   Static renewal         :

 2. Temperature (ฐC):            25 ฑ 1ฐC

 3. Light quality:               Ambient laboratory illumination

 4. Light intensity:             10-20 ME/m2/s  (50-100 ft-c)(ambient
                                 laboratory levels)

 5. Photoperiod:                 16 h light, 8 h darkness

 6. Test chamber size:           500 mL (minimum)
                                                        !

 7. Test solution volume:        250 mL (minimum)

 8. Renewal  of test                                     ;
     solutions:                  Daily                  I

 9. Age of test organisms:       Newly hatched larvae less than 24 h old.
                                 If shipped, not more than 48 h old, 24 h
                                 range in age

10. No. larvae per test chamber: 15 (minimum of 10)     '

11. No. replicate chambers
     per concentration:          4 (minimum of 3)

12. No. larvae per
     concentration:              60 (minimum of 30)     ;

13. Source of food:              Newly hatched Artemia nauplii  (less than
                                 24 h old)

14. Feeding  regime:              Feed 0.1 g newly hatched (less than 24-h
                                 old) brine shrimp nauplii three times
                                 daily at 4-h intervals or,  as  a minimum,
                                 0.15 g twice daily,  6 h between feedings
                                 (at the beginning of  the work day prior to
                                 renewal, and at the  end of the work day
                                 following  renewal).   Sufficient nauplii
                                 are added  to provide an excess.  Larvae
                                 fish are not fed during the final  12 h of
                                 the test
                                    79

-------
   TABLE 1.  SUMMARY OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA
             FOR FATHEAD MINNOW, PIMEPHALES PROMELAS, LARVAL SURVIVAL AND
             GROWTH TOXICITY TESTS WITH EFFLUENTS AND RECEIVING WATERS
             (CONTINUED)                                           '
15. Cleaning:


16. Aeration:



17. Dilution water:
18. Test concentrations:



19. Dilution factor


20. Test duration:

21. Endpoints:

22. Test acceptability
     criteria:
23. Sampling
     requirements:
Siphon daily, immediately before test
solution renewal

None, unless DO concentration falls below
4.0 mg/L.  Rate should not exceed
100 bubbles/min

Uncontaminated source of receiving or
other natural water, synthetic water
prepared using MILLIPORE MILLI-Qฎ or
equivalent deionized water and reagent
grade chemicals, or DMW (see Section 7,
Dilution Water)

Effluents: Minimum of 5 and a.control
Receiving  Water:   100% receiving  water  or
minimum of 5 and a control

Effluents:  > 0.5                    ;
Receiving waters:   None or > 0.5     !

7 days

Survival and growth (weight)
80% or greater survival in controls; ;
average dry weight per surviving organism
in control  chambers equals or exceeds:0.25 mg
For on-site tests, samples collected
daily, and used within 24 h of the time
they are removed from the sampling device;
For off-site tests, a minimum of
three samples collected on days one,
three and five with a maximum holding time
of 36 h before first use (see Section 8,
24. Sample volume required:   2.5 L/day
                                    80

-------
 11.12  ACCEPTABILITY OF TEST  RESULTS

 11.12.1  For the test results to  be acceptable,  survival  in  the  controls must
 be at least 80%.  The average dry weight  per  surviving  control larvae  at the
 end of the test should equal  or exceed 0.25 mg.

 11.13  DATA ANALYSIS
11.13.1  GENERAL

11.13.1.1  Tabulate and summarize the data.
growth response data  is shown  in Table  2.
A sample set of survival and
    TABLE 2.  SUMMARY OF SURVIVAL AND GROWTH DATA  FOR  FATHEAD MINNOW,
              PIMEPHALES PROMELAS, LARVAE EXPOSED  TO A REFERENCE TOXICANT
              FOR SEVEN DAYS
NaPCP
Cone.
(M9/L)
0
32
64
128
256
512
Proportion of
Survival in Replicate
Chambers
A B C D
1.0
0.8
0.9
0.9
0.7
0.4
1.0
0.8
1.0
0.9
0.9
0.3
0,9
1.0
1.0
0.8
1.0
0.4
0.9
0.8
1.0
1.0
0.5
0.2
Mean
Prop.
Surv
0
0
0
0
0
0
.95
.85
.975
.90
.775
.325
Avg
Rep!
A
0.711
0.517
0.602
0.566
0.455
0.143
Dry Wgt (mg) In Mean
icate Chambers Dry Wgt
B C D (mg)
0.662
0.501
0.669
0.612
0.502
0.163
0.646
0,723
0.694
0,410
0.606
0.195
0.690
0.560
0.676
0.672
0.508
0.099
0.677
0.575
0.660
0.565
0.454
0.150
     Four replicates of 10 larvae each.
11.13.1.2  The endpoints of toxicity tests using the fathead minnow,
Pimephales promelas, larvae are based on the adverse effects on survival and
growth.  The LC50, the IC25, and the IC50 are calculated using point
estimation techniques (see Section 9, Chronic Toxicity Test Endpoints and Data
Analysis).  LOEC and NOEC values for survival and growth are obtained using a
hypothesis testing approach such as Dunnett's Procedure (Dunnett, 1955) or
Steel's Many-one Rank Test (Steel, 1959; Miller, 1981) (see Section 9).
Separate analyses are performed for the estimation of the LOEC and NOEC
endpoints and for the estimation of the LC50, IC25 and IC50.  Concentrations

                                      81                  i

-------
at which there is no survival in any of the test chambers are excluded from
the statistical analysis of the NOEC and LOEC for survival and growth, but
included in the estimation of the LC50, IC25, and IC50.  See the Appendices
for examples of the manual computations, and examples of data input and
program output.

11.13.1.3  The statistical tests described here must be used with a knowledge
of the assumptions upon which the tests are contingent.  Tests for normality
and homogeneity of variance are included in Appendix B.  The assistance of a
statistician is recommended for analysts who are not proficient in statistics.

11.13.2  EXAMPLE OF ANALYSIS OF FATHEAD MINNOW, PIMEPHALES PROMELAS,
         SURVIVAL DATA

11.13.2.1  Formal statistical analysis of the survival data is outlined in
Figures 5 and 6.  The response used in the analysis is the proportion of
animals surviving in each test or control chamber.  Separate analyses are
performed for the estimation of the NOEC and LOEC endpoints and for the
estimation of the LC50, EC50, and 1C endpoints.  Concentrations at which there
is no survival in any of the test chambers are excluded from statistical
analysis of the NOEC and LOEC, but included  in the estimation of the  1C, EC,
and LC endpoints.

11.13.2.2  For the case of equal numbers of  replicates across all
concentrations and the control, the evaluation of the NOEC and LOEC endpoints
is made via a parametric test, Dunnett's Procedure, or a  nonparametric test,
Steel's Many-one Rank Test, on the arc  sine  square root transformed data.
Underlying assumptions of Dunnett's Procedure, normality  and homogeneity of
variance, are formally tested.  The test for normality is the Shapiro-Wilk's
Test, and Bartlett's Test is used to test for homogeneity of variance.  If
either of these tests fails, the nonparametric test, Steel's Many-one Rank
Test, is used to determine the NOEC and  LOEC endpoints.   If the assumptions of
Dunnett's Procedure are met, the endpoints are estimated  by the parametric
procedure.                                                           ',

11.13.2.3   If  unequal numbers of replicates  occur among the concentration
levels tested, there are  parametric and  nonparametric  alternative  analyses.
The parametric analysis is a t test with the Bonferroni adjustment (see
Appendix D).   The Wilcoxon Rank Sum Test with the Bonferroni adjustment is the
nonparametric  alternative  (see Appendix F).

11.13.2.4   Probit Analysis  (Finney, 1971;  see Appendix  I)  is used  to  estimate
the concentration that  causes a specified  percent decrease  in survival  from
the control.   In this analysis, the total  mortality data  from all  test
replicates  at  a  given concentration are combined.   If  the data  do  not fit  the
Probit  analysis, the Spearman-Karber Method, the Trimmed  Spearman-Karber
Method,  or  the Graphical  Method may be used  (see Appendices  I-L).
                                       82

-------
             STATISTICAL ANALYSIS OF FATHEAD MINNOW LARVAL
                        SURVIVAL AND GROWTH TEST

                      SURVIVAL HYPOTHESIS TESTING
SURVIVAL DATA
PROPORTION SURVIVING
i
r
                                  ARC SINE
                              TRANSFORMATION
                             SHAPIRO-WIUCSTEST
                                                NON-NORMAL DISTRIBUTION
                   NORMAL DISTRIBUTION
         HOMOGENEOUS
           VARIANCE
                               BARTLETTSTEST
HETEROGENEOUS
   VARIANCE
                                                    J
1
EQUAL NUME
REHLICA!
NO
^
IER OF
ES?
YES
T TP^T WITH
"^ 1 YVI 1 [I r\t IklKIL 1 1 IQ
BONFERRONI uu TptT &
ADJUSTMENT lt&l





i
EQ

YES
J
STEEL'S MANY-ONE
RANK TEST


t
ENDPOINT ESTIMATES
NOEC, LOEC
JAL NUMBER OF
REPLICATE-S?

1 NO
WILOOXON RANK SUM
TESiTWITH
BONFEHROIMI ADJUSTMENT


'

Figure 5.    Flowchart for statistical analysis of the fathead minnow, Pimephales
            promelas, larval  survival data by hypothesis testing.

                                    83

-------
          STATISTICAL ANALYSIS OF FATHEAD MINNOW LARVAL
                     SURVIVAL AND GROWTH TEST
                      SURVIVAL POINT ESTIMATION
      MORTALITY DATA
         #DEAD
       TWO OR MORE
    PARTIAL MORTALITIES?
 NO
             YES
      IS PROSIT MODEL
       APPROPRIATE?
    (SIGNIFICANT X2 TEST)
NO
ONE OR MORE
PARTIAL MORTALITIES?
NO
^
             YES
              IYES
  GRAPHICAL METHOD
       LC50
      PROBIT METHOD
     ZERO MORTALITY IN THE
     LOWEST EFFLUENT CONG.
    AND 100% MORTALITY IN THE
    HIGHEST EFFLUENT CONG.?
NO
                                      YES
                              SPEARMAN-KARBER
                                  METHOD
                              TRIMMED SPEARMAN
                               KARBER METHOD
                                LC50 AND 95%
                                CONFIDENCE
                                  INTERVAL
Figure 6.   Flowchart for statistical analysis of the fathead minnow, Pimephales
           promelas, larval  survival  data by point estimation.       '(

                                   84

-------
11.13.2.5  Example of Analysis of Survival Data

11.13.2.5.1  This example uses the survival data from the Fathead Minnow
Larval Survival and Growth Test (Table 2).  The proportion surviving in each
replicate must first be transformed by the arc sine square root transformation
procedure described in Appendix B.  The raw and transformed data, means and
variances of the transformed observations at each toxicant concentration and
control are listed in Table 3.  A plot of the survival proportions is provided
in Figure 7.

       TABLE 3.  FATHEAD MINNOW, PIMEPHALES PROMELAS, SURVIVAL DATA
            Replicate
Control
                                         NaPCP Concentration (ttq/L)
32
64
128
256
512

RAW


ARC SINE
TRANS-
FORMED

Mean(Y,)
s?
^
A
B
C
D
A
B
C
D



1
- 1
0
0
1
1
1
1
1
0
1
.0
.0
.9
.9
.412
.412
.249
.249
.330
.0088

0
0
1
0
1
1
1
1
1
0
2
.8
.8
.0
.8
.107
.107
.412
.107
.183
.0232

0.9
1.0
1.0
1.0
1.249
1.412
1.412
1,412
1.371
0.0066
3
0
0
0
1
1
1
1
1
1
0
4
.9
.9
.8
.0
.249
.249
.107
.412
.254
.0155

0.
0.
1.
0.
0.
1.
1.
0.
1.
0.
5
7
9
0
5
991
249
412
785
109
0768

0.4
0.3
0.4
0.2
0.685
0.580
0.685
0.464
0.604
0.0111
6
11.13.2.6  Test for Normality

11.13.2.6.1  The first step of the test for normality is to center the
observations by subtracting the mean of all observations within a
concentration from each observation in that concentration.  The centered
observations are summarized in Table 4.

11.13.2.6.2  Calculate the denominator, D, of the statistic:
                                D =
                                              2
    Where:  X,- = the ith centered observation

            X  = the overall mean of the centered observations

            n  = the total number of centered observations
                                      85

-------
o
*  ' *
O       00
        d
                                                    evi

                                                    LO
                                                    (O
                                                        oi
                                                       ^i
                                                       UJ
                                                       I

                                                       i
                                                       Q-
                                                       3
                                                       UJ
                                                       D-
                                                       o
                                                       o
                                                           CO


                                                            0)
                                                   _Q
                                                   as
                                                   s-
                                                   o
                                                   o.
                                                   o
                                                   s-
                                                   Q.
                                                   3
                                                           •fj
                                                            o
                                                            OJ
                                                            CD
                                                   -O
                                                  O
                                                  o
                   NOIldOdOdd IVAIAUnS
                            86

-------
         TABLE 4.  CENTERED OBSERVATIONS FOR SHAPIRO-MILK'S EXAMPLE
NaPCP Concentration fua/L)
Replicate
A
B
C
D
Control
0.082
0.082
-0.081
-0.081
32
-0.076
-0.076
0.229
-0.076
64
-0.122
0.041
0.041
0.041
128
-0.005
-0.005
-0.147
0.158
256
-0.118
0.140
0.303
-0.324
512
0.081
-0.024
0.081
-0.140
11.13.2.6.3  For this set of data:   n = 24               ;
                           X = — (0.000) =0.000
                                24                       ;
                            D = 0.4265
11.13.2.6.4  Order the centered observations  from smallest  to  largest
                            XC1)        \
                                                         \
where XO) denotes the ith ordered observation.   The  ordered observations
for this example are listed in Table 5.

    TABLE 5.  ORDERED CENTERED OBSERVATIONS FOR THE  SHAPIRO-WILK'S  EXAMPLE
i
1
2
3
4
5
6
7
8
9
10
11
12
x
-0.324
-0.147
-0.140
-0.122
-0.118
-0.081
-0.081
-0.076
-0.076
-0.076
-0.024
-0.005
i
13
14 '
15
16
17
18
19
20
21
22
23
24
xci)
-0.005
0.041
0.041
0.041
0.081
0.081
0.082
0.082
0.140
0.158
0.229
0.303
                                      87

-------
11.13.2.6.5  From Table 4, Appendix B, for the number of observations, n,
obtain the coefficients a,,  a2,  ...  ak where k is n/2 if n is even and (n-l)/2
if n is odd.  For the data in this example, n = 24 and k = 12.  The a,- values
are listed in Table 6.

        TABLE 6.  COEFFICIENTS AND DIFFERENCES FOR SHAPIRO-MILK'S EXAMPLE


           •        ^             v(r)~i+1)lv(i)
           i        a,-            X      - X
1
2
3
4
5
6
7
8
9
10
11
12
0.4493
0.3098
0.2554
0.2145
0.1807
0.1512
0.1245
0.0997
0.0764
0.0539
0.0321
0.'0107
0.627
0.376
0.298
0.262
0.200
0.163
0.162
0.157
0.117
0.117
0.065
0.000
X(24, .
Y<23)
A -
y{22)
A, -
Y<21>
A . ~
XC20) _
x(19) _
x<18> -
x(17) _
y(16)
ซ<15)
x!ซ! :
XC13> _
x<1>;
x(2)
x<3)
x"'
X
X2!
X *
x(ป
X
X
X
x<12>
11.13.2.6.6  Compute the test statistic, W, as follows:
                             1   ฃ                   2
                             D   i=i *


The differences x1> - X(i) are listed in Table 6.  For the data in this
example,
                        W=	(0.6444)2 = 0.974
                             0.4265

11.13.2.6.7  The decision rule for this test is to compare W as calculated in
Section 13.2.6.6 to a critical value found in Table 6, Appendix B.  If the
computed W is less than the critical value, conclude that the data are not
normally distributed.  For the data in this example, the critical value at a
significance level of 0.01 and n = 24 observations is 0.884.  Since M = 0.974
is greater than the critical value, conclude that the data are normally
distributed.

11.13.2.7  Test for Homogeneity of Variance

11.13.2.7.1  The test used to examine whether the variation in mean proportion
surviving is the same across all toxicant concentrations including the
control, is Bartlett's Test (Snedecor and Cochran, 1980).  The test statistic
is as follows:

                                      88

-------
         B=
Where:  V1-  = degrees  of freedom for each  toxicant  concentration  and
              control, V,  =  (n,  - 1)                       !

        nf  = the number of replicates  for concentration  i

        In = loge

        i  = 1, 2, ..., p where p is the number of concentrations
               including the control                      ;
           _   i-1
                    ,
                i-i
         C =
11.13.2.7.2  For the data in this example (see Table 3),  all  toxicant
concentrations including the control  have the same number of replicates
(n, = 4 for all  i).   Thus,  V, - 3 for all i.

11.13.2.7.3  Bartlett's statistic is  therefore:

                 B -  [(18)1*2(0.0236)  -3 ฃlrz(Si) 1/1.1296
                                          2=1

                   =  [18(-3.7465)  -  3(-24.7516)3/1.1296

                   =  6.8178/1.1296                       i

                   =  6.036

11.13.2.7.4  B is approximately distributed as chi-square with p -  1  degrees
of freedom, when the variances are in fact  the same.  Therefore, the
appropriate critical value for this test (from a table of chi-square
distribution), at a significance level  of 0.01 with five  degrees of freedom,
is 15.086.  Since B = 6.036 is less than the critical  value of 15.086,
conclude that the variances are not different.            j

11.13.2.8  Dunnett's Procedure

11.13.2.8.1  To obtain an estimate of the pooled variance for the Dunnett's
Procedure, construct an ANOVA table as described in Table 7.
                                      89

-------
                              TABLE 7.  ANOVA TABLE
Source
Between
Within
Total
df
P - 1
N - p
N - 1
Sum of Squares
(SS)
SSB
SSW
SST
Mean Square(MS)
(SS/df)
Sg = SSB/(p-l)
S* = SSH/(N,p)

  Where:  p  - number toxicant concentrations including the control

          N  - total number of observations n.,  + n2  ... +  np

          n,-  = number of observations in concentration i
   SSB = Y,Tl/nฑ-G2/N            Between Sum of Squares
          i-l


   SST - ฃ T,Yij-G2/N            Total Sum of Squares


   SSW = SST-SSB                  Within Sum of Squares


    G   -   the grand total of all sample observations,G =  ฃ Tฑ


    T,-  =   the total of the replicate measurements for concentration i

    Y,--  -   the jth observation for concentration i (represents the proportion
            surviving for toxicant concentration i in test chamber j)

11.13.2.8.2  For the data in this example:

    n, - n2  - n3 = n4 = n5  = n6 = 4                                    ;

    N  = 24

    TI = YH + Y12  + Y13 + Y14 = 5.322                                  :
    T2 = Y21 + Y22  + Y23 + Y24 - 4.733
    T3 - Y31 + Y32  + Y33 + Y34 = 5.485
    T4 - Y41 + Y42  H- Y43 + Y44 = 5.017


                                      90

-------
    T5 = Y51 + Y52 + Y53 + Y54 = 4.437
    T6 = Y61 + Y62 + Y63 + Y64 = 2.414

    G = T, + T2 + T3  + T4  + T5 + T6 = 27.408
  SSB  =  f,Ti/ni-G2/N
      = _1_(131.495)  -  (27.  408)2  =  1.574
         4                 24
  SST =
         i-1.7=1  ~"                                          !

       = 33.300 - (27.408)2  == 2.000                      |
                      24

  SSW = SST-SSB   = 2.000  -  1.574 = 0.4260


   S2   = SSB/(p-l)  = 1.574/(6~1)  = 0.3150

   S2   = SSW/(N-p)  = 0.426/(24-6) = 0.024                 ,



11.13.2.8.3  Summarize these calculations in the ANOVA table  (Table 8)
                                                          i

              TABLE 8.  ANOVA TABLE FOR DUNNETT'S PROCEDURE EXAMPLE
      Source         df        Sum of Squares        Mean Square(MS)
                                    (SS)                 (SS/df)
      Between        5            1.574                  0.315

      Within        18            0.426                  0.024



      Total         23            2.002
                                      91

-------
11.13.2.8.4  To perform the individual  comparisons,  calculate the t statistic
for each concentration, and control combination as follows:
Where:  Y
            = mean proportion surviving for concentration  i

        Y1  - mean proportion surviving for the  control

        Sw  = square root of the within mean square

        n.,  ~ number of replicates for the control

        n,-  - number of replicates for concentration  i.

11.13.2.8.5  Table 9 includes the calculated t values for each concentration
and control combination.  In this example, comparing  the 32 ng/l concentration
with the control the calculation is as follows:
                              (1.330-1.183)
                                                  =1.341
                           [0.155^(1/4) + (1/4)]
                      TABLE 9.  CALCULATED T VALUES
       NaPCP Concentration
32
64
128
256
512
2
3
4
5
6
1.341
-0.374
0.693
2.016
6.624
11.13.2.8.6  Since the purpose of this test is to detect a significant
reduction in proportion surviving, a one-sided test is appropriate.  The
critical value for this one-sided test is found in Table 5, Appendix C.   For
an overall alpha level of 0.05, 18 degrees,of freedom for error and five
concentrations (excluding the control) the critical value is 2.41.  The mean
proportion surviving for concentration i is considered significantly less than
the mean proportion surviving for the control if t,- is greater than the
critical value.  Since t, is greater than 2.41,  the 512 M9/L concentration  has
significantly lower survival than the control.  Hence the NOEC and the LOEC
for survival are 256 /zg/L and 512 /jg/L, respectively.

                                      92

-------
11.13.2.8.7  To quantify the sensitivity of the test, the minimum significant
difference (MSD) that can be detected statistically may be calculated.
                          MSD =  d fl^d/nj + (1/73)
  Where: d = the critical value for Dunnett's procedure
        Sw = the square root of the within mean square
        n  = the common number of replicates at each concentration
             (this assumes equal replication at each concentration)
        n, = the number of replicates in the control .
11.13.2.8.8  In this example:
                      MSD =  2.41(0. 155)v/(l/4) + (1/4)
                           = 2.41 (0.155)(0.707)
                           = 0.264                       I
11.13.2.8.9  The MSD (0.264) is in transformed units.  To determine the MSD in
terms of percent survival, carry out the following conversion.

    1. Subtract the MSD from the transformed control mean;
                            1.330 - 0.264 = 1.066        !
    2. Obtain the untransformed values for the control  mean and the difference
       calculated in 1.
                         [Sine ( 1.330)  ]* = 0.943
                         [Sine ( 1.066)  ]2 = 0.766
    3. The untransformed MSD (MSD )  is determined by subtracting  the
       untransformed values from 2.                      !   '
                        MSDU -  0.943  - 0.766 =  0.177
11.13.2.8.10  Therefore, for this set of data,  the  minimum  difference in mean
proportion surviving between the control and any toxicant; concentration that
can be detected as statistically significant is 0.177.
11.13.2.8.11  This represents a decrease in survival of 19% from the  control.
                                      93

-------
11.13.2.9  Calculation of the LC50
11.13.2.9.1  The data used for the Probit Analysis is summarized in Table 10.
To perform the Probit Analysis, run the USEPA Probit Analysis Program.
An example of the program input and output is supplied in Appendix I. ;
                       TABLE 10.  DATA FOR PROBIT ANALYSIS
                                         NaPCP Concentration
                         Control
32
64
128
256
512
Number Dead
Number Exposed
2
40
6
40
1
40
4
40
9
40
27
40
11.13.2.9.2  For this example, the chi-square test for heterogeneity was not
significant, thus Probit Analysis appears appropriate for this data.

11.13.2.9.3  Figure 8 shows the output data for the Probit Analysis of the
data in Table 10 using the USEPA Probit Program.

11.13.3  EXAMPLE OF ANALYSIS OF FATHEAD MINNOW, PIMEPHALES PROMELAS, GROWTH
         DATA

11.13.3.1  Formal statistical analysis of the growth data is outlined in
Figure 9.  The response used in the statistical analysis is mean weight per
replicate.  An 1C estimate can be calculated for the growth data via a point
estimation technique (see Section 9, Chronic Toxicity Test Endpoints and Data
Analysis).  Hypothesis testing can be used to obtain the NOEC for growth.
Concentrations above the NOEC for survival are excluded from the hypothesis
test for growth effects.

11.13.3.2  The statistical analysis using hypothesis tests consists of a
parametric test, Dunnett's Procedure, and a nonparametric test, Steel's
Many-one Rank Test.  The underlying assumptions of the Dunnett's Procedure,
normality and homogeneity of variance, are formally tested.  The test for
normality is the Shapiro-Wilk's Test and Bartlett's Test is used to test for
homogeneity of variance.  If either of these tests fails, the nonparametric
test, Steel's Many-one Rank Test, is used to determine the NOEC and LOEC
endpoints.  If the assumptions of Dunnett's Procedure are met, the endpoints
are determined by the parametric test.

11.13.3.3  Additionally, if unequal numbers of replicates occur among the
concentration levels tested there are parametric and nonparametric alternative
analyses.  The parametric analysis is a t test with the Bonferroni adjustment
(see Appendix D).  The Wilcoxon Rank Sum Test with the Bonferroni adjustment
is the nonparametric alternative (see Appendix F).

                                      94

-------
        Probit Analysis of Fathead Minnow Larval  Survival  Data


Cone.
Control
32.0000
64.0000
128.0000
256.0000
512.0000

Number
Exposed
40
40
40
40
40
40

Number
Resp.
2
6
1
4
9
27
Observed
Proportion
Responding
0.0500
0.1500
0.0250
0.1000
0.2250
0.6750
                                                               Proportion
                                                               Responding
                                                              Adjusted for
                                                                Controls

                                                                  0.0000
                                                                  0.0779
                                                                  -.0577
                                                                  0.0237
                                                                  0.1593
                                                                  0.6474
Chi - Square for Heterogeneity (calculated)
Chi - Square for Heterogeneity
        (Tabular value at 0.05 level)
                                4.522

                                7.815
          Probit Analysis of Fathead Minnow Larval Survival Data

               Estimated LC/EC Values and Confidence Limits
Point
LC/EC  1.00
LC/EC 50.00
Exposure
  Cone.

 127.637
 422.696
 Lower           Upper
 95% Confidence Limits
 34.590
345.730
195.433
531.024
Figure 8.  Output for USEPA  Probit Analysis Program, Version  1.5
                                     95

-------
             STATISTICAL ANALYSIS OF FATHEAD MINNOW LARVAL
                        SURVIVAL AND GROWTH TEST

                                  GROWTH
                                 GROWTH DATA
                               MEAN DRY WEIGHT
    POINT ESTIMATION
   HYPOTHESIS TESTING
(EXCLUDING CONCENTRATIONS
 ABOVE NOEC FOR SURVIVAL)
    ENDPOINT ESTIMATE
        IC25, IC50
   SHAPIRO-MILK'S TEST
                   NORMAL DISTRIBUTION
                       NON-NORMAL DISTRIBUTION
         HOMOGENEOUS
           VARIANCE
                               BARTLETTSTEST
                     J
                             HETEROGENEOUS
                                VARIANCE
                           I
^
EQUAL NUME
REPUCAT
NO
1
JEROF
ES?
YES
T-TESTWITH DUNNETrs
BONFERRONI DUNNETTS
ADJUSTMENT lfctsl






\
EQ

YES
r
STEEL'S MANY-ONE
RANK TEST


*
ENDPOINT ESTIMATES
NOEC, LOEC
UAL NUMBER OF
REPLICATES?
I NO
WILCOXON RANK SUM
TEST WITH
BONFERRONI ADJUSTMENT




Figure 9.    Flowchart for  statistical analysis of  fathead minnow, Pimephales
            promelas, larval growth  data.

                                     96

-------
11.13.3.4  The data, mean and variance of the observations; at each
concentration including the control are listed in Table 11.  A plot of the
weight data for each treatment is provided in Figure 10.  Since there is
significant mortality in the 512 vg/L concentration, its effect on growth is
not considered.

         TABLE 11.  FATHEAD MINNOW, PIMEPHALES PROMELAS, GROWTH DATA
  Replicate    Control
             NaPCP Concentration  C/zg/L)

           32     64      128     256    512
A
B
C
D
0.711
0.662
. 0.646
0.690
0.517
0.501
0.723
0.560
0.602
0.669
0.694
0.676
0.566
0.612
0.410
0.672
0.455
0.502
0.606
0.254
„
-
-

MearKY,)
Si
i
0.677
0.00084
1
0.525 0.660 0.624 0.580
0.01032 0.00162 0.01256 0.0218
2 3 4 5
6
 11.13.3.5  Test for Normality

 11.13.3.5.1  The first step of the test for normality is to center the
 observations by subtracting the mean of all the observations within a
 concentration from each observation in that concentration.  The centered
 observations are summarized in Table 12.

            TABLE 12.  CENTERED OBSERVATIONS FOR SHAPIRO-MILK'S EXAMPLE
    Replicate
                                          NaPCP Concentration (tto/L)
Control
32
64
                                                           128
256
A
B
C
D
0.034
-0.015
-0.031
0.013
-0.058
-0.074
0.148
-0.015
-0.058
0.009
0.034
0.016
0.001
0.047
-0.155
0.107
0.001
0.048
0.152
-0.200
                                       97

-------
O
CO
o
CO

CD
o
LO
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 II I

                      O      O
                      CM      T-
                        •       a       ^
                      o      o
                                                                       CD
                                                                       LO
                                                                       CVI
                                                                    —  o
                  O
                  CO
 o       o      o

(Bui) J.H9I3M NV3IN




             98
                                                                               o>
                                                                               o

                                                                               en
                                                                               cO
                                                                               =5
                                                                               I/)
                                                                           HI
                                                                               8
                                                                               ex.
                                                          HI


                                                          oi
                                                          /v**  *O

                                                      Sg |
                                                          X -S

                                                         .8-:-
                                                          z  ฐ
                                                          HI  ฃ

                                                          S 'E
                                                          _. "O
                                                         •< -J  ra
                                                         :5 5
                                                          O "^
                                                      CVI  GO <^
                                                                               nj
                                                                                 01
                                                                               
 O S-
i— O
D_ <4-
                                                                              O)
                                                                              cn

-------
11.13.3.5.2  Calculate the denominator, D, of the test statistic:
                                    i-i
    Where: X,- = the ith centered observation

           X  = the overall  mean of the centered observations

           n  = the total  number of centered observations:

For this set of data,            n = 20

                                 X = _1_ (0.004) = 0.000
                                     20                  i

                                 D = 0.1414

11.13.3.5.3  Order the centered observations from smallest to largest

                  X(1)  
-0.200
-0.155
-0.074
-0.058
-0.058
-0.031
-0.015
-0.015
0.001
0.001
i
11
12
13
14
15
16
17
18
19
20
X(D
0.009
. 0.013
0.016
0.034
0.034
0..047
0..048
0.107
0.148
i 0.152
 11.13.3.5.4  From Table 4,  Appendix B,  for the number of observations,  n,
 obtain the coefficients a,,  a2, ..., ak where k is n/2 if n is even and
 (n-l)/2 if n is odd.   For the data in  this example,  n = 20 and k = 10.   The a,-
 values are listed in  Table 14.
                                       99

-------
     TABLE  14.   COEFFICIENTS AND  DIFFERENCES  FOR  SHAPIRO-WILK'S  EXAMPLE


          i        a,-            X(n'i+1) - X(0
1
2
3
4
5
6
7
8
9
10
0.4734
0.3211
0.2565
0.2085
0.1686
0.1334
0.1013
0.0711
0.0422
0.0140
0..352
0.303
0.131
0.106
0.105
0.065
0.049
0.031
0.012
0.008
v(20)
A
x<19)
xcia>
XC17)
X(16>
X(15>
X(H)
X(13)
x<12)

- Xm
x<2)
v(3)
- x<4)
- x<5)
- x(6)
- x(7)
- x(8)
- x(9)
- x(10)
11.13.3.5.5  Compute the test statistic, W, 'as follows:
                         W =
                              D
                                                                      I

the differences x(n"i+1) - Xci)  are  listed  in  Table  14.   For  this  set  of data:

                     W =      1       (0.3666)2 = 0.9505
                           0.1414

11.13.3.5.6  The decision rule for this test is to compare W with the critical
value found in Table 6, Appendix B.  If the computed W is less than the
critical value, conclude that the data are not normally distributed.  For this
example, the critical value at a significance level of 0.01 and 20
observations (n) is 0.868.  Since W = 0.959 is greater than the critical
value, the conclusion of the test is that the data are normally distributed.

11.13.3,6  Test for Homogeneity of Variance

11.13.3.6.1  The test used to examine whether the variation in mean dry weight
is the same across all toxicant concentrations including the control, is
Bartlett's Test (Snedecor and Cochran, 1980).  The test statistic is as
fol1ows:
                 [ (   Vฑ) ln~S2 -   'Vฑ in
                                      100

-------
   Where:  V,-   =   degrees  of freedom  for  each toxicant  concentration and
                 control,   V,- =  (n,- - 1)

          n,-   =   the  number of  replicates  for  concentration  i.

          In   =   loge

           i   =   1,  2,  ..., p where  p is the number of concentrations
                 including the  control
                  i=l
                              1=1        i=l
11.13.3.6.2  For the data in this example,  (see Table 11)  all  toxicant
concentrations including the control  have the same number  of replicates
(n,  = 4 for all  i).   Thus,  V,- = 3 for all i.

11.13.3.6.3  Bartlett's statistic is therefore:


        B=  [(15) In (0.00947) -3 ฃ In (S?)]/1.133

          =  [15(-5.9145) - 3(-26.2842]/1.133

          =  8.8911/1.133

          =  7.847

11.13.3.6.4  B is approximately distributed as chi-square  with p - 1 degrees
of freedom, when the variances are in fact the same.  Therefore, the
appropriate critical value for this test, at a significance level of 0.01 with
four degrees of freedom, is 13.277.  Since B = 7.847 is less than the critical
value of 13.277, conclude that the variances are not different.
                                      101

-------
11.13.3.7  Dunnett's Procedure
11.13.3.7.1  To obtain an estimate of the pooled variance for the Dunnett's
Procedure, construct an ANOVA table as described in Table 15.
                          TABLE 15.  ANOVA TABLE
Source
Between
Within
Total
df
p - 1
N - p
N -, 1
Sum of Squares
(SS)
SSB
SSW
SST
Mean Square(MS)
(SS/df)
SB = SSB/(p-l)
Sy = SSW/(N-p)

Where:      p  - number toxicant concentrations including the control
            N  = total  number of observations n,, +  n2 ... + n
            n,-  - number of observations in concentration i
          SSB =   r/j^-GV-W     Between Sum of Squares

         SST =  ฃ ^yfi-G2/N      Total  Sum of Squares
                jt-i;?=i
          SSW = SST-SSB     .      Within Sum of Squares
           G  =  the  grand  total  of all  sample observations,  G = & T-
                                                                 i=l   :
          T,-  -  the  total  of the replicate measurements for  concentration i
         YJJ  =  the  jth  observation for concentration  i (represents the mean
                 dry  weight of the fish  for toxicant concentration  i in test
                 chamber  j)
                                     102

-------
11.13.3.7.2  For the data in this example:
                     n.,  = n2  =  n3 = n4 = n5 = 4
                          20
                     T  - V  -i- Y   4- V
                     T2 - Y2; + "ll + Yฃ
                     T  _ Y  >v   i v
                      !;  !!
                     Y14 = 2.709
                     Y,, = 2.301
Tc = Y
      51
                               Y52 + Y53
33  +  Y34 = 2.641
„  +  Y44 = 2.260
     Y54 = 1.817
                      6 = T, + T2 + T3 + T4 + T5 = 11.728
                   SSB = f,Ti/nฑ-G2/N
                          1=1
                        =  I (28.017) - (11.728)'
                          4               20
                                0.1270
                   SST =
                        =  7.146 - (11.728r  = 0.2687
                                     20

                   SSW = SST-SSB   = 0.2687  - 0.1270  = 0.1417

                    S2  = SSB/(p-l)  = 0.1270/(5-l)   = 0.0318

                    S2  = SSW/(N-p)  = 0.041/(20-5)  = 0.0094

11.13.3.7.3  Summarize these calculations in the ANOVA  table  (Table  16)
          TABLE 16.  ANOVA TABLE FOR DUNNETT'S PROCEDURE  EXAMPLE
Source
Between
Within
df
4
15
Sum of Squares
(SS)
0.1270
0.1417
Mean Square(MS)
(SS/df)
0.0318
0.0094
     Total
 19
 0.2687
                                      103

-------
11.13.3.7.4  To perform the individual comparisons, calculate the t statistic
for each concentration, and control combination as follows:
Where: Y
       n
mean dry weight for toxicant concentration i

mean dry weight for the control

square root of the within mean square

number of replicates for the control

number of replicates for concentration i.
11.13.3.7.5  Table 17 includes the calculated t values for each concentration
and control combination.  In this example, comparing the 32 jug/L concentration
with the control the calculation is as follows:
                              (0.677-0.575)
                          [0 . 097^(1/4) + (1/4)]
                                                  = 1.487
                       TABLE 17.  CALCULATED T VALUES
                NaPCP
            Concentration
32
64
128
256
2
3
4
5
1.487
0.248
1.632
3.251
11.13.3.7.6  Since the purpose of this test is to detect a significant
reduction in mean weight, a one-sided test is appropriate.  The critical  value
for this one-sided test is found in Table 5, Appendix C.  For an overall  alpha
level of 0.05, 15 degrees of freedom for error and four concentrations
(excluding the control) the critical value is 2.36.  The mean weight for
concentration "i" is considered significantly less than the mean weight for
the control if t, is greater than, the critical  value.   Since t4  and t5 are
greater than 2.36, the 128 /Ltg/L and 256 /*g/L concentrations have significantly
lower growth than the control.  Hence the NOEC and the LOEC for growth are
128 ng/l and 256 ng/L, respectively.
                                      104

-------
11.13.3.7.7  To quantify the sensitivity of the test, the minimum significant
difference (MSD) that can be statistically detected may be calculated:,

                          MSD = d S^CL/nJ + (l/rz)


Where: d  = the critical value for the Dunnett's Procedure

       Sw = the square root of the within mean square

       n  = the common number of replicates at each concentration
            (this assumes equal replication at each concentration)

       n., = the number of replicates in the control.

11.13.3.7.8  In this example:                            ,
                      MSD = 2.36(0.052)^(1/4) + (1/4)


                           = 2.36 (0.097)  (0.707)

                           = 0.162

 11.13.3.7.9   Therefore,  for this set  of data,  the minimum  difference  that  can
 be  detected  as  statistically significant  is  0.162 mg.

 11.13.3.7.10 This  represents a 24% reduction  in mean  weight  from the control.

 11.13.3.8 Calculation  of the 1C

 11.13.3.8.1   The growth data in Table 2 modified to  be mean weights  per
 original  number of  fish are utilized  in this example.   As  seen in Table 2  and
 Figure 11, the  observed means are not monotonically  non-increasing with
 respect to concentration (the mean response  for each higher concentration  is
 not less than or equal  to the mean response  for the  previous  concentration,
 and the responses between concentrations  do  not follow a linear trend).
 Therefore, the  means are smoothed prior to calculating the 1C.  In the
 following discussion,  the observed means  are represented, by Y1 and the
 smoothed means  by M,-.

 11.13.3.8.2   Starting with the control mean, Yt = 0.677, we see that  Y, >  Y2.
 Set M, - Yr  Comparing  Y2 to Y3, Y2 < Y3.

 11.13.3.8.3   Calculate the smoothed means:

                   M2 =  M3 - (Y2 + Y3)/2 =  0.618          ;

 11.13.3.8.4  For the remaining observed means,  M3 > Y4 :>  Y,5 > Y6. Thus,  M4
 becomes Y4,  M5  becomes  Y5,  etc., for  the remaining concentrations.  Table  18
 contains the smoothed means, and Figure 11 provides  a plot of the smoothed
 concentration response curve.

                                       105                >

-------
                                   CM

                                   it)
                                  -O
I  ' I
                 1  I  ' I  ' I  ' I
              ง

                                              o
                                              o
ฉ  o  o  o  o o o o o o" o o Q p  o  o
              (Bui) SSVWOI9 NV3W
                                                        OS

                                                        I
                                           O
                                           c
                                           C
                                          -o
                                          n}
                                          O)
*ฃ.
• , w? LJLJ
I! — ITX1

Q_
O
DC
q
	 i


l*%
& z
LU
0_
5'
•cฃ
,^J
Q
O
co
-co



01
c:
(0
OJ
ฃ
-a
OJ
JC
4->
O
o •
E 00
CO •— 1
-o -a
(XJ (O
•-CM
CO
C CO
rO O)
O) r—
E-Q
T3 1—
0)
> C
O
CO fO
JD +->
o m
73
"^U Q
s-
3 O)
                                            c
-------
              TABLE  18.   FATHEAD MINNOW, PIMEPHALES PROHELAS.
                         MEAN  GROWTH  RESPONSE AFTER SMOOTHIN&
NaPCP
Cone
(pg/L)
Control
32
64
128
256
512
i
1
2
3
4
5
6
Response
means, Y.-
(mgj '
0.677
0.575
0.660
0.565
0.454
0.150
Smoothed
means. M.-
(>ng) '
0.677
0.618
0.1518
0.565
0.454
0.150
11.13.3.8.5  An IC25 and an IC50 can  be  estimated using the Linear
Interpolation Method.   A 25% reduction  in weight, compared to the controls,
would result in a mean dry weight of  0.508 mg, where M,,(l - p/100) = 0.677(1 -
25/100).  A 50% reduction in weight,  compared to the controls, would result in
a mean weight of 0.339 mg, where M.(l -  p/100) = 0.677(1 --50/100).   Examining
the smoothed means and their associated  concentrations  (Table 18), the
response 0.508 mg is bracketed by C4 = 128 ^g/L and C.  = 2,56  //g/L.   For the
50% reduction (0.339 mg), the response  (0.339 /xg) is "bracketed by C5 =
256 /zg/L and C6 = 512  A*g/L.                              !
11.13.3.8.6  Using the equation in Section  4.2  from Appendix M, the estimate
of the IC25 is calculated as follows:                     ;


      icp = c^U^U
     IC25 = 128+[0.677 (1-25/100) -0.565]    (256-128)
                                             (0.454-0.565,1

          = 194

11.13.3.8.7  Using the equation in Section  4.2  of  Appendix M the estimate of
the  IC50 is calculated as follows:.


      ICp = C                            ~
     IC50 = 256+ [0.677 (1-50/100) -0.454]    (512-2561)
                                             (0.150-0 .454)

          = 353 fj.g/1



                                     107

-------
11.13.3.8.8  When the ICPIN program was used to analyze this set of data,
requesting 80 resamples, the estimate of the IC25 was 193.9503 //g/L.  The
empirical 95% confidence interval for the true mean was (54.9278 /*g/L,
340.6617 A*g/L).  The computer program output for the IC25 for this data set is
shown in Figure 12.

11.13.3.8.9  When the ICPIN program was used to analyze this set of data for
the IC50, requesting 80 resamples, the estimate of the IC50 was 353.2884 M9/L-
The empirical 95% confidence interval for the true mean was 208.4723 //g/L and
418.5276 fig/L.  The computer program output is shown in Figure 13.
11.14  PRECISION AND ACCURACY

11.14.1  PRECISION

11.14.1.1  Single-Laboratory Precision

11.14.1.1.1  Information on the single-laboratory precision of the fathead
minnow larval survival and growth test is presented in Table 19.  The range of
NOECs was only two concentration intervals, indicating good precision.


  TABLE 19.  PRECISION OF THE FATHEAD MINNOW, PIMEPHALES PROMELAS, LARVAL
             SURVIVAL AND GROWTH TEST, USING NAPCP AS A REFERENCE TOXICANT1'2


Test
1
2
3
4
5
n:
Mean:

NOEC
(/*9/L)
256
128
256
128
128
5
, NA

LOEC
(M9/L)
512
256
512
256
256
5
NA
Chronic
Value
U9/L)
362
181
362
181
181
5
253.4
     From Pickering,  1988.
     For a discussion of the precision of data from chronic toxicity tests,
     (see Section 4,  Quality Assurance).
                                      108

-------
Cone. ID


Cone. Tested
Response
Response
Response
Response
1
2
3
4
0
0
0
0
1
0
.711
.662
.646
.690


0.
0.
0.
0.
2
32
517
501
723
560


0.
0.
0.
0.
3
64
602
669
694
676
4
128
0.566
0.612
0.410
0.672


0
0
0
"0
5
256
.,455
.502
..606
..254
6
512
0.143
0.163
0.195
0.099
*** Inhibition Concentration Percentage Estimate ***
Toxicant/Effluent: NaPCP
Test Start Date: Example   Test Ending Date:
Test Species: Fathead minnows
Test Duration:        .     7-d
DATA FILE: fhmanual.icp
OUTPUT FILE: fhmanual.i25
Cone.
ID
1
2
3
4
5
6
Number
Replicates
4
4
4
4
4
4
Concentration
ug/i
0.000
32.000
64.000
128.000
256.000
512.000
Response
Means
0.677
0.575
0.660
0.565
0.454
0.150
Stcl. Pooled
Dev. Response Means
0.029
0.102
0.040
0.112
0.148
0.040
0.677
0.618
0.618
0.565
0.454
0.150
The Linear Interpolation Estimate:   193.9503   Entered:? Value: 25

Number of Resamplings:   80
The Bootstrap Estimates Mean: 186.4935 Standard Deviation:    52.6094
Original Confidence Limits:   Lower:   107.0613 Upper:  , 285.6449
Expanded Confidence Limits:   Lower:    54.9278 Upper:   340.6617
Resampling time in Seconds:     1.81  Random Seed: 1272173518
              Figure 12.  ICPIN program output for the IC25.
                                    109

-------
Cone. ID

Cone. Tested
Response
Response
Response
Response
1
2
3
4
1
0
0.711
0.662
0.646
0.690
2
32
0.517
0.501
0.723
0.560
3
64
0.602
0.669
0.694
0.676
4
128
0.566
0.612
0.410
0.672
5
256
0.455
0.502
0.606
0.254
6
512
0.143
0.163
0.195
0.099
*** Inhibition Concentration Percentage Estimate ***
Toxicant/Effluent: NaPCP
Test Start Date: Example   Test Ending Date:
Test Species: Fathead minnows
Test Duration:             7-d
DATA FILE: fhmanual.icp
OUTPUT FILE: fhmanual.i50
Cone.
ID
1
2
3
4
5
6
Number
Replicates
4
4
4
4
4
4
Concentration
ug/l
0.000
32.000
64.000
128.000
256.000
512.000
Response
Means
0.677
0.575
0.660
0.565
0.454
0.150
Std. Pooled
Dev. Response Means
0.029
0.102
0.040
0.112
0.148
0.040
0.677
0.618
0.618
0.565
0.454
0.150
The Linear Interpolation Estimate:   353.2884   Entered P Value: 50

Number of Resamplings:   80
The Bootstrap Estimates Mean: 345.1108 Standard Deviation:    37.0938
Original Confidence Limits:   Lower:   262.7783 Upper:   394.0629
Expanded Confidence Limits:   Lower:   208.4723 Upper:   418.5276
Resampling time in Seconds:     1.87  Random Seed: 1126354766
              Figure 13.  ICPIN program output for the IC50.
                                    110

-------
11.14.1.2  Multilaboratory Precision

11.14.1.2.1  A multilaboratory study of Method 1000.0 described in the first
edition of this manual (USEPA, 1985e), was performed using seven blind samples
over an eight month period (DeGraeve et. al., 1988).  In this study, each of
the 10 participating laboratories was to conduct two tests simultaneous with
each sample, each test having two replicates of 10 larvae for each of five
concentrations and the control.  Of the 140 tests planned^ 135 were completed.
Only nine of the 135 tests failed to meet the acceptance criterion of 80%
survival in the controls.  Of the 126 acceptable survival NOECs reported, an
average of 41% were median values, and 89% were within one concentration
interval of the median (Table 20).  For the growth (weight) NOECs, an average
of 32% were at the median, and 84% were within one concentration interval of
the median (Table 21).  Using point estimate techniques, the precision (CV) of
the IC50 was 19.5% for the survival data and 19.8% for the growth data.  If
the mean weight acceptance criterion of 0.25 mg for the surviving control
larvae, which is included in this revised edition of the method, had applied
to the test results of the interlaboratory study, one third of the 135 tests
would have failed to meet the test criteria (Norberg-King, personal
communication and 1989 memorandum; DeGraeve et al., 1991),

11.14.2  ACCURACY

11.14.2.1  The accuracy of toxicity tests cannot be determined.
                                     Ill

-------
   TABLE 20.  COMBINED FREQUENCY DISTRIBUTION FOR SURVIVAL NOECs
              FOR ALL LABORATORIES1
                                    NOEC Frequency (%)Distribution
Sample
Tests with Two Reps
 Median  ฑ lz>23
Tests with Four Reps
  Median  ฑ lz  > 25
1.
2.
3.
4.
5.
6.
7.
Sodium Pentachlorophenate (A) 35
Sodium Pentachlorophenate (B) 42
Potassium
Potassium
Refinery
Refinery
Dichromate (A)
Dichromate (B)
Effluent 301
Effluent 401
Utility Waste 501
47
41
26
37
56
53
42
47
41
68
53
33
12
16
6
18
6
10
11
57
56
75
50
78
56
56
29
44
25
50
22
44
33
14
0
0
0
0
0
11
From DeGraeve et al., 1988.
Percent of values within one concentration intervals of the median.
Percent of values within two or more concentrations intervals of the
median.
                                 112

-------
TABLE 21. COMBINED FREQUENCY DISTRIBUTION FOR WEIGHT NOECs FOR ALL
          LABORATORIES1
NOEC Freauencv
Tests with Two
Sample Median ฑ I2
1.
2.
3.
4.
5.
6.
7.
Sodium Pentachlorophenate (A)
Sodium Pentachlorophenate (B)
Potassium Dichromate (A)
Potassium Dichromate (B)
Refinery Effluent 301
Refinery Effluent 401
Utility Waste 501
59
37
35
12
35
37
11
41
63
47
47
53
47
61
Reps
>23
0
0
18
41
12
16
28
(%) Distribution
Tests with Four Reps
Median ฑ I2 > 23
57
22
88
i 63
75
I 33
33
I
43
45
0
25
25
56
56
0
33
12
12
0
11
11
   From DeGraeve et al.,  1988.
   Percent of values within one concentration intervals of the median.
   Percent of values within two or more concentrations intervals of the
   median.
                                    113

-------
                                  SECTION 12

                                  TEST  METHOD

                     FATHEAD MINNOW, PIHEPHALES PRONELAS,
                EMBRYO-LARVAL SURVIVAL AND TERATOGENICITY TEST
                                 METHOD 1001.0
12.1  SCOPE AND APPLICATION

12.1.1  This method estimates the chronic toxicity of whole effluents and
receiving water to the fathead minnow, Pimephales promelas, using embryos in a
seven-day, static renewal test.  The effects include the synergistic,
antagonistic, and additive effects of all the chemical, physical, and
biological components which adversely affect the physiological and biochemical
functions of the test organisms.  The test is useful in screening for
teratogens because organisms are exposed during embryonic development.

12.1.2  Daily observations on mortality make it possible to also calculate the
acute toxicity for desired exposure periods (i.e., 24-h, 48-h, 96-h LCBOs).

12.1.3  Detection limits of the toxicity of an effluent or pure substance are
organism dependent.

12.1.4  Brief excursions in toxicity may not be detected using 24-h composite
samples.  Also, because of the long sample collection period involved in
composite sampling, and because the test chambers are not sealed, highly
degradable and highly volatile toxicants, in the source may not be detected in
the test.

12.1.5  This test method is commonly used in one of two forms: (1) a
definitive test, consisting of a minimum of five effluent concentrations and a
control, and (2) a receiving water test(s), consisting of one or more
receiving water concentrations and a control.

12.2  SUMMARY OF METHOD

12.2.1  Fathead minnow, Pimephales promelas, embryos are exposed in a static
renewal system to different concentrations of effluent or to receiving water
for seven days, starting shortly after fertilization of the eggs.  Test
results are based on the total frequency of both mortality and gross
morphological deformities (terata).

12.3  INTERFERENCES

12.3.1  Toxic substances may be introduced by contaminants in dilution water,
glassware, sample hardware, and testing equipment (see Section 5, Facilities,
Equipment, and Supplies).
                                      114

-------
                         ฐI  loV.5issolved ^en  (DO),  high  concentrations  of
 toxic sbsUnces dlSSฐlved  sollds' and extremes  of PH  may mask the presence of


 12.3.3  Improper effluent sampling and sample handling may  adversely affect
 test results (see Section 8, Effluent and Receiving Water Sampling, Sample
 Handling, and Sample Preparation for Toxicity Tests).               ->amj>ie

 ll'*-*  Pathogenic and/or predatory organisms in the dilution water and
 effluent may affect test organism survival and confound test results.

 12.4  SAFETY

 12.4.1  See Section 3,  Health and Safety.                 I
                                                           i
 12.5  APPARATUS AND EQUIPMENT


         Mathead "!1nnow and brine ^rimp culture units -- See Section  11

         3J5hฐW'/7rPfa7^/rTla^  Urva1  Surv1val  and Growth  Test>  and
       .9?3b-  UTฐ test  effluent toxicity on-site or in  the laboratory,
 sufficient numbers  of newly  fertilized  eggs  must be available,  preferably from
 a  laboratory fathead  minnow  culture unit.   If necessary,  embryos  can  be
 shipped  in well  oxygenated water in insulated containers.   In cases where
 shipping is  necessary,  up  to 48-h old embryos may be  used  for the test.


 J™i!h?m!!niPl!;rf "" aut??atlc Ampler,  preferably with  sample cooling
 capability,  that can  collect a  24-h composite sample  of 5  L  or more.
                col?t?ine,rrs  "  tor  Cample  shipment  and  storage  (see  Section  8,
 for Toxcny Teltl)!"9        SamDlin9> Sample  Handling,  and Sample Preparation
                      Chamber  ฐr ea.uivalerit  facility with temperature control


12.5.5  Water purification system -- MILLIPORE MILLI-Qฎ, deionized water or
equivalent  (see Section 5, Facilities, Equipment, and Supplies)

12.5.6  Balance -- analytical, capable of accurately weighing to 0.00001 g.
 on     ^        wei?nts.' Class's -- for checking performance of balance
Weights should bracket the expected weights of material to :be weighed

12.5.8  Test chambers -- four (minimum of three) borosilicate glass or
fcP ฐsable  non-toxic plastic labware, per test solution, such as:  500-mL
beakers; 100 mm x 15 mm or 100 mm x 20 mm glass or disposable polystyrene
SH ^heS; 01V2^ ฐDi stackable "Carolina" culture dishes   The chambers
should be covered with safety glass plates or sheet plastic during the test to

aVd                                                                  ™
                                     115

-------
12 5 9  Dissecting microscope, or long focal  length magnifying lens,  hand  or
sfend supported  - for examining embryos and larvae in the test chambers.
12.5.10  Light box, microscope lamp, or flashlight - for illuminating
chambers during examination and observation of embryos and larvae.
12 5 11  Volumetric flasks and graduated cylinders -- Class A, borosilicate
glass or non-toxic plastic labware, 10-1000 ml, for making test solutions.
12.5.12  Volumetric pipets -- Class A, 1-100 ml.
12.5.13  Serological pipets -- 1-10 ml, graduated.
12.5.14  Pipet bulbs and fillers  - PROPIPETฎ, or  equivalent.
12  5  15  Droppers, and glass  tubing with fire  polished edges,  2-mm ID—  for
transferring  embryos, and  4-mm  ID -  for transferring larvae.
12  5  16  Wash bottles -- for  washing  embryos  from  substrates  and containers
and for  rinsing  small glassware  and instrument electrodes  and probes.;
12.5.17  Thermometers, glass  or electronic,  laboratory grade  - for  measuring
water temperatures.
 12.5.18  Bulb-thermograph  or  electronic-chart type thermometers  --for
continuously recording  temperature.
 12 5 19  Thermometer,  National  Bureau of Standards Certified  (see  USEPA Method
 170.i, USEPA 1979b)  --  to calibrate laboratory thermometers.          ;
 12.5.20  Meters, pH, DO, and specific conductivity - for routine  physical and
 chemical measurements.
 12.6  REAGENTS AND CONSUMABLE MATERIALS
 1? 6 1  Samole containers -- for  sample shipment  and storage  (see Section 8,
 Effluent and Receiving Water Sampling, Sample Handling and Sample Preparation
 for Toxicity Tests).
  12.6.2  Data sheets (one  set per test) -- for recording data.
  12.6.3  Tape, colored --  for labelling test  chambers.
  12.6.4  Markers,  waterproof  --  for marking  containers,  etc.
  12 6 5  Reagents for hardness  and alkalinity tests  -- see USEPA Methods'130.2
  and  310.1,  USEPA 1979b.
  12 6.6  Membranes and  filling  solutions  for DO probe (see USEPA Method 360.1,
  USEPA 1979b), or reagents -- for modified Winkler analysis.
                                        116

-------
12.6.7  Standard pH buffers, pH 4, pH 7, and pH 10  (or as per instructions of
instrument manufacturer) -- for instrument calibration (see USEPA Method
150.1, USEPA 1979b).

12.6.8  Specific conductivity standards -- see USEPA Method 120.1, USEPA
1979b.

12.6.9  Laboratory quality control samples and standards -- for calibration of
the above methods.

12.6.10  Reference toxicant solutions -- see Section 4, Quality Assurance.

12.6.11  Reagent water — defined as distilled or deionized water which does
not contain substances which are toxic to the test  organisms (see Section 5,
Facilities, Equipment, and Supplies).

12.6.12  Effluent, receiving water, and dilutiorf water --' see Section 7,
Dilution Water; and Section 8, Effluent and Receiving Water Sampling, Sample
Handling, and Sample Preparation for Toxicity Tests.

12.6.13  TEST ORGANISMS, FATHEAD MINNOWS, PIMEPHALES PROMELAS

12.6.13.1  Fathead minnow embryos, less than 36-h old, are used for the test.
The test is conducted with four (minimum of three)  test chambers at each
toxicant concentration and control.  Fifteen (minimum of ten) embryos are
placed; in each replicate test chamber.  Thus 60 (minimum of 30) embryos are
exposed at each test concentration and 360 (minimum of 180) embryos would be
needed for a test consisting of five effluent concentrations and a control.

12.6.13.2  Sources of Organisms

12.,6.13.2.1  It is recommended that the embryos be  obtained from inhouse
cultures or other local sources if at all possible, because it is often
difficult to ship the embryos so that they will be  less than 36 h old for
beginning the test.  Receipt of embryos via Express Mail, air express, or
other carrier, from a reliable outside source is an acceptable alternative,
but they must not be over 48 h old when used to begin the test.
                                                         i
12.6.13.2.2  Culturing methods for fathead minnows, Pimephales promelas, are
described in Section 6, Section 11 and in USEPA, 1993b.

12.6.13.2.3  Fish obtained from outside sources (see Section 5, Facilities,
Equipment, and Supplies) such as commercial  biological supply nouses for use
as brood stock should be guaranteed to be (1) of the correct species, (2)
disease free, (3) in the requested age range, and (4) in good condition.  This
can be done by providing the record of the date on which the eggs were laid
and hatched, and information on the sensitivity of the contemporary fish to
reference toxicants.

12.6.13.3  Obtaining Embryos for Toxicity Tests from InhoUse Cultures.

12.6.13.3.1  Spawning substrates with the newly-spawned,  fertilized embryos
are removed from the spawning tanks or ponds, and the embryos are separated

                                      117

-------
from the spawning substrate by using the index finger and rolling the embryos
gently with a circular movement of the finger (see Gast and Brungs, 1973).
The embryos are then combined and washed from the spawning substrate onto a
400 urn NITEXฎ screen, sprayed with a stream of deionized water to remove
detritus and food particles, and back-washed with dilution water into a
crystallizing dish for microscopic examination.  Damaged and infertile eggs
are discarded.

12.6.13.3.2  The embryos from three or more spawns are pooled in a single
container to provide a sufficient number to conduct the tests.  These embryos
may be used immediately to start a test inhouse or may be transported for use
at a remote location.  When transportation is required, embryos should be
taken from the substrates within 12 h of spawning.  This permits off-site
tests to be started with less than 36-h old embryos.  Embryos should be
transported or shipped in clean, opaque, insulated containers, in well aerated
or oxygenated fresh culture or dilution water, and should be protected from
extremes of temperature and any other stressful conditions during transport.
Instantaneous changes of water temperature when embryos are transferred from
culture unit water to test dilution water, or from transport container water
to on-site test dilution water, should be less than 2ฐC.  Sudden changes  in
pH, dissolved ions, osmotic strength, and DO should be avoided.

12.7  EFFLUENT AND RECEIVING WATER COLLECTION, PRESERVATION, AND STORAGE

12.7.1  See Section 8, Effluent and Receiving Water Sampling, Sample Handling,
and Sample Preparation for Toxicity Tests.

12.8  CALIBRATION AND STANDARDIZATION

12.8.1  See Section 4, Quality Assurance.

12.9  QUALITY CONTROL

12.9.1  See Section 4, Quality Assurance.

12.10  TEST PROCEDURES

12.10.1  TEST SOLUTIONS

12.10.1.1  Receiving Waters

12.10.1.1.1  The sampling point is determined  by  the objectives of the test.
Receiving water toxicity is determined with samples used directly  as collected
or after samples are passed through a 60 urn NITEXฎ filter and compared without
dilution, against  a control.  Using four replicate chambers per test, each
containing 100 ml, and 400 ml for chemical analysis, would require
approximately one  liter, or more, of sample per test day.

12.10.1.2  Effluents

12.10.1.2.1  The selection of the effluent test concentrations should be  based
on the objectives  of the study.  A dilution factor of  0.5 is  commonly used.
A dilution factor  of 0.5 provides precision of ฑ  100%,  and testing of

                                      118

-------
concentrations between 6.25% and 100% effluent using only five effluent
concentrations (6.25%, 12.5%, 25%, 50%, and 100%).  Improvements in precision
decline rapidly if the dilution factor is increased beyond 0..5 and precision
declines rapidly if a smaller dilution factor is used.  Therefore, USEPA
recommends the use of the > 0.5 dilution factor.

12.10.1.2.2  If the effluent is known or suspected to be highly toxic, a lower
range of effluent concentrations should be used (such as 25%, 12.5%, 6.25%,
3.12%, and 1.56%).  If a high rate of mortality is observed during the first
1 to 2 h of the test, additional dilutions should be added at the lower range
of effluent concentrations.

12.10.1.2.3  The volume of effluent required for daily renewal of four
replicates per concentration, each containing 100 ml of test solution, is
1.5 L.  Sufficient test solution (approximately 1000 ml) iis prepared at each
effluent concentration to provide 400 ml additional volume for chemical
analyses.  If the sample is used for more than one daily renewal of test
solutions, the volume must be increased proportionately.

12.10.1.2.4  Tests should begin as soon as possible, preferably within 24 h of
sample collection.  The maximum holding time following retrieval of the sample
from the sampling device should not exceed 36 h for the off-site toxicity
tests unless permission is granted by the permitting authority.  In no case
should the sample be used in a test more than 72 h after sample collection
(see Section 8, Effluent and Receiving Water Sampling, Sample Handling, and
Sample Preparation for Toxicity Tests).                   '

12.10.1.2.5  Just prior to test initiation (approximately 1 h) the temperature
of sufficient quantity of the sample to make the test solutions should be
adjusted to the test temperature and maintained at that temperature during the
addition of dilution water.

12.10.1.2.6  The DO of the test solutions should be checked prior to test
initiation.  If any of the solutions are supersaturated with oxygen or any
solution has a DO below 4.0 mg/L, all of the solutions and the control must be
gently aerated.                                           ;

12.10.1.3  Dilution Water

12.10.1.3.1  Dilution water may be uncontaminated receiving water, a standard
synthetic (reconstituted) water, or some other uncontamineited natural  water
(see Section 7, Dilution Water).

12.10.1.3.2  If the hardness of the test solutions (including the control)
does not equal or exceed 25 mg/L as CaC03,  it  may  be necessary to adjust  the
hardness by adding reagents for synthetic softwater as listed in Table 3,
Section 7.  In this case parallel tests should be conducted, one with  the
hardness adjusted and one unadjusted.
                                      119

-------
12.10.2  START OF THE TEST

12.10.2.1  Label the test chambers with a ma'rking pen and use color-coded tape
to identify each treatment and replicate.  A minimum of five effluent
concentrations and a control are used for each effluent test.  Each treatment
(including the control) should have four (minimum of three) replicates.

12.10.2.2  Tests performed in laboratories that have inhouse fathead minnow
breeding cultures must initiate tests with embryos less than 36 h old.  When
the embryos must be shipped to the test site from a remote location, it may be
necessary to use embryos older than 36 h because of the difficulty of
coordinating test organism shipments with field operations.  However, in the
latter case, the embryos must not be more than 48 h old at the start of the
test and should all be within 24 h of the same age.

12.10.2.3  Randomize the position of the test chambers at the beginning of the
test (see Appendix A).  Maintain the chambers in this configuration throughout
the test.  Preparation of a position chart may be helpful.

12.10.2.4  The test organisms should come from a pool of embryos consisting of
at least three separate spawnings.  Gently agitate and mix the embryos to be
used in the test in a large container so that eggs from different spawns are
thoroughly mixed.

12.10.2.5  Using a small bore (2 mm ID) glass tube, the embryos are placed one
or two at a time into each randomly arranged test chamber or intermediate
container in sequential order, until each chamber contains 15 (minimum of 10)
embryos, for a total of 60 (minimum of 30) embryos for each concentration
(see Appendix A).  The amount of water added to the chambers when transferring
the embryos to the compartments should be kept to a minimum to avoid
unnecessary dilution of the test concentrations.

12.10.2.6  After the embryos have been distributed to each test chamber,
examine and count them.  Remove and discard damaged or infertile eggs and
replace with new undamaged embryos.  Placing the test chambers on a light
table may facilitate examining and counting the embryos.

12.10.3  LIGHT, PHOTOPERIOD AND TEMPERATURE

12.10.3.1  The light quality and intensity should be at ambient laboratory
levels, which is approximately 10-20 /iE/m/s,  or 50 to 100 foot candles
(ft-c), with a photoperiod of 16 h of light and 8 h of darkness.  The water
temperature in the test chambers should be maintained at 25 ฑ 1ฐC.

12.10.4  DISSOLVED OXYGEN (DO) CONCENTRATION

12.10.4.1  Aeration may affect the toxicity of effluents and should be used
only as a last resort to maintain satisfactory DO concentrations.  The DO
concentrations should be measured in the new solutions at the start of the
test (Day 0) and before daily renewal of the new solutions on subsequent days.
The DO concentrations should not fall below 4.0 mg/l. (see Section 8, Effluent
and Receiving Water Sampling, Sample Handling, and Sample Preparation1 for

                                      120

-------
Toxicity Tests).  If it is necessary to aerate, all concentrations and the
control should be aerated.  The aeration rate should not exceed 100
bubbles/min, using a pipet with an orifice of approximately 1.5 mm, such as a
1-mL KIMAXฎ serological Pipet No. 37033, or equivalent.  Care should be taken
to ensure that turbulence resulting from the aeration does; not cause undue
physical stress to the embryos.

12.10.5  FEEDING

12.10.5.1  Feeding is not required.                       <

12.10.6  OBSERVATIONS DURING THE TEST                   .  |     '

12.10.6.1  Routine Chemical and Physical Determinations

12.10.6.1.1  DO is measured at the beginning and end of each 24-h exposure
period in at least one test chamber at each test concentrations and in the
control.

12.10.6.1.2  Temperature and pH are measured at the end of each 24-h exposure
period in at least one test chamber at each test concentration and in the
control.  Temperature should also be monitored continuously or observed and
recorded daily for at least two locations in the environmental control system
or the samples.  Temperature should be measured in a sufficient number of test
vessels, at least at the end of the test, to determine temperature variation
in the environmental chamber.

12.10.6.1.3  The pH is measured in the effluent sample each clay before new,
test solutions are made.

12.10.6.1.4  Conductivity, alkalinity and hardness are measured in each new
sample (100% effluent or receiving water) and in the control.

12.10.6.2  Record all the measurements on the data sheet (Figure 1).

12.10.6.3  Routine Biological Observations

12.10.6.3.1  At the end of the first 24 h of exposure, before renewing the
test solutions,'examine the embryos.  Remove the dead embryos (milky colored
and opaque) and record the number (Figure 2).  If the rate of mortality
(including those with fungal infection) exceeds 20% in the control chambers,
or if excessive non-concentration-related mortality occurs, terminate the test
and start a new test with new embryos.                    -
                                                          i
12.10.6.3.2  At 25ฐC, hatching may begin on the fourth day.  After hatching
begins, count the number of dead and live embryos and the number of hatched,
dead, live, and deformed larvae, daily.  Deformed larvae are those with gross
morphological abnormalities such as lack of appendages, lack of fusiform shape
(non-distinct mass), lack of mobility, a colored, beating heart in an opaque
mass, or other characteristics that preclude survival.  Count and remove dead
embryos and larvae as previously discussed and record the numbers for all of
the test observations (Figure 2).  Upon hatching, deformed larvae are counted
as dead.
                                      121

-------
Discharger:
Location: _
Analyst:
Dates:
Dav
Control :
Temp.
D.O. Initial
Final
oH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

1










2










3










4










5










6










7










Remarks










Dav
Cone:
Temo.
D.O. Initial
Final
pH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

1










2










3










4










5










6










7










Remarks










Dav
Cone:
Terno.
D.O. Initial
Final
pH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

1










2










3










4










5










6










7










Remarks










Figure  1.  Data form for  the  fathead minnow, Pimephales pro/ne/as,  embryo-
           larval  survival and  teratogenicity test.  Routine  chemical  and
           physical  determinations.
                                    122

-------
Discharger:
Location:
Analyst:
Dates:
Dav
Control :
Temp .
D.O. Initial
Final
pH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

1










2










3










4










5










6










7










Remarks










Dav
Cone:
Temp.
D.O. Initial
Final
pH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

1










2










3










4










5










6










7










Remarks










Dav
Cone:
Temo.
D.O. Initial
Final
pH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

1










2










3










4










5










6










7










Remarks










 Figure  1.  Data form for  the  fathead minnow, Pimephales promelas,  embryo-
           larval  survival and  teratogenicity test.   Routine  chemical  and
           physical  determinations (CONTINUED).
                                    123

-------
Discharger:

Location:
Test Dates:.

Analyst: 	
Cone: Rep. Condition of
No. Embryos/larvae
Control: 1 Live/dead
Terata
2 Live/dead
Terata
3 Live/dead
Terata
4 Live/dead
Terata
Treatment: 1 Live/dead
Terata
2 Live/dead
Terata
3 Live/dead
Terata
4 Live/dead
Terata
Treatment: 1 Live/dead
Terata
2 Live/dead
Terata
3 Live/dead
Terata
4 Live/dead
Terata
Treatment: 1 Live/dead
Terata
2 Live/dead
Terata
3 Live/dead
Terata
4 Live/dead
Terata
Dav '
1
































2
































3
































4
































5
































6
































7
































   Figure 2.   Data form for the fathead minnow, Pimephales promelas,
              embryo-larval survival  and teratogenicity test. Survival
              and terata data.
                                    124

-------
Discharger:

Location:
Test Dates:.

 Analyst: _
Cone: Rep. Condition of
No. Embrvos/1 arvae
Treatment: 1 Live/dead
Terata
2 Live/dead
Terata
3 Live/dead
Terata
4 Live/dead
Terata
Treatment: 1 Live/dead
Terata
2 Live/dead
Terata
3 Live/dead
Terata
4 Live/dead
Terata
Dav
1 2 3 45 67


































































































Comments:
      Figure  2.   Data  form for the  fathead  minnow,  Pimephales promelas,
                 embryo-larval  survival  and teratogenicity test.  Survival
                 and terata data (CONTINUED).
                                    125

-------
12.10.6.3.3  Protect the embryos and larvae from unnecessary disturbance
during the test by carrying out the daily test observations, solution
renewals, and removal of dead organisms carefully.  Make sure that the test
organisms remain immersed during the performance of the above operations.

12.10.7  DAILY CLEANING OF TEST CHAMBERS

12.10.7.1  Since feeding is not required, test chambers are not cleaned daily
unless accumulation of particulate matter at the bottom of the chambers causes
a problem.

12.10.8  TEST SOLUTION RENEWAL

12.10.8.1  Freshly prepared solutions are used to renew the tests daily.  For
on-site toxicity studies, fresh effluent or receiving water samples should be
collected daily, and no more than 24 h should elapse between collection of the
samples and their use in the tests (see Section 8, Effluent and Receiving
Water Sampling, Sample Handling and Sample Preparation for Toxicity Tests).
For off-site tests, a minimum of three samples are collected, preferably on
days one, three, and five.  Maintain the samples in the refrigerator at 4ฐC
until used.

12.10.8.2  The test solutions are renewed immediately after removing dead
embryos and/or larvae.  During the daily renewal process, the water level in
each chamber is lowered to a depth of 7 to 10 mm, which leaves 15 to 20% of
the test solution.  New test solution should be added slowly by pouring down
the side of the test chamber to avoid excessive turbulence and possible injury
to the embryos or larvae.

12.10.9  TERMINATION OF THE TEST

12.10.9.1  The test is terminated after seven days of exposure.  Count the
number of surviving, dead, and deformed larvae, and record the numbers of each
(Figure 2).  The deformed larvae are treated as dead in the analysis of the
data.  Keep a separate record of the total number and percent of deformed
larvae for use in reporting the teratogenicity of the test solution.

12.10.9.2  Prepare a summary of the data as illustrated in Figure 3.

12.11  SUMMARY OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA

12.11.1  A summary of test conditions and test acceptability criteria is
presented in Table 1.

12.12  ACCEPTABILITY OF TEST RESULTS

12.12.1  For the test results to be acceptable, survival  in the controls must
be at least 80%.
                                      126

-------
Discharger:

Location:
 Test  Dates:

	 Analyst:
Treatment
No. dead embryos
and larvae
No. terata
Total mortality
(dead and
deformed)
Total mortality
(%)
Terata (%)
Hatch (%)
Control

































i







Comments:
    Figure  3.   Summary data for the fathead minnow,  Pimephales promelas,
               embryo-larval  survival  and teratogenicity test.
                                    127

-------
TABLE 1.  SUMMARY OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA FOR
          FATHEAD MINNOW, PIMEPHALES PROMELAS, EMBRYO-LARVAL SURVIVAL AND
          TERATOGENICITY TOXICITY TESTS WITH EFFLUENTS AND RECEIVING WATERS
 1. Test type:
 2. Temperature:
 3. Light quality:
 4. Light intensity:
 5. Photoperiod:
 6. Test chamber size:
 7. Test solution volume:
 8. Renewal of test solutions:
 9. Age of test organisms:
Static renewal
25 ฑ 1ฐC
Ambient laboratory illumination
10-20 //E/m2/s or 50-100 f.t-.c. (ambient
laboratory levels)
16 h light, 8 h dark            ;
150 mL (Minimum)
70 mL (Minimum)
Daily
Less than 36-h old embryos (Maximum of
48-h if shipped)
10. No. embryos per test chamber:   15 (minimum of 10)
11. No. replicate test
     chambers per concentration:
12. No. embryos per concentration:
13. Feeding regime:
14. Aeration:
15. Dilution water:
4 (minimum of 3)
60 (minimum of 30)
Feeding not required
None unless DO falls below 4.0 mg/L
Uncontaminated source of receiving or
other natural water, synthetic water
prepared using MILLIPORE MILLI-Q* or
equivalent deionized water and reagent
grade chemicals or DMW (see Section
7, Dilution Water).  The hardness  of the
test solutions should equal or exceed
25 mg/L (CaC03)  to ensure hatching
success
                                    128

-------

 TABLE 1.   SUMMARY  OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA FOR
           FATHEAD MINNOW,  PIMEPHALES PROMELAS,  EMBRYO-LARVAL SURVIVAL AND
           TERATOGENICITY TOXICITY TESTS WITH EFFLUENTS AND RECEIVING
           WATERS (CONTINUED)
16.  Test concentrations:



17.  Dilution factor:


18.  Test duration:

19.  Endpoint:


20.  Test acceptability criteria:

21.  Sampling requirements:
22. Sample volume required:
Effluents:  Minimum of 5 and a control
Receiving waters:   100% receiving water
or minimum of 5 and a control
Effluents:  > 0.5
Receiving waters:

7 days
None, or > 0.5
Combined mortality (dead and deformed
organisms)

80% or greater survival in controls

For on-site tests, samples collected
daily and used within 24 h of the time
they  are  removed from  the  sampling
device.   For  off-site  tests a minimum
of three samples  collected on days one,
three, and five with a maximum holding
time  of 36  h before  first use  (see
Section 8, Effluent and Receiving Water
Sampling, Sample  Handling,  and Sample
Preparation  for   Toxicity  Tests,  and
Subsection 8.5.4)

1.5  to  2.5   L/day  depending  on  the
volume of test solutions used
                                    129

-------
12.13  DATA ANALYSIS

12.13.1  GENERAL

12.13.1.1  Tabulate and summarize the data (Figure 3).

12.13.1.2  The endpoints of this toxicity test are based on total mortality,
combined number of dead embryos, and dead and deformed larvae.  The EC1 is
calculated using Probit Analysis (Finney, 1971; see Appendix I).  Separate
analyses are performed for the estimation of LOEC and NOEC endpoints and for
the estimation of the EC1 endpoint.  Concentrations at which there is no
survival in any of the test chambers are excluded from the statistical
analysis of the NOEC and LOEC, but included in the estimation of the EC1
endpoint.  See the Appendices for examples of the manual computations and
examples of data input and output for the computer programs.

12.13.1.3  The statistical tests described here must be used with a knowledge
of the assumptions upon which the tests are contingent.  The assistance of a
statistician is recommended for analysts who are not proficient in statistics.

12.13.2  EXAMPLE OF ANALYSIS OF  FATHEAD MINNOW  EMBRYO-LARVAL SURVIVAL AND
         TERATOGENICITY DATA

12.13.2.1  Formal statistical analysis of the total mortality data is outlined
on the flowchart in Figure 4.  The response used in the analysis is the total
mortality proportion in each test or control chamber.  Separate analyses are
performed for the estimation of the NOEC and LOEC endpoints and for the
estimation of the EC endpoint.  Concentrations at which there is 100% total
mortality in all of the test chambers are excluded from statistical analysis
of the NOEC and LOEC, but included in the estimation of the EC1 endpoint.

12.13.2.2  For the case of equal numbers of replicates across all
concentrations and the control, the evaluation of the NOEC and LOEC endpoints
is made  via a parametric test, Dunnett's Procedure, or a nonparametric test,
Steel's  Many-one Rank Test, on the arc sine square root transformed data.
Underlying assumptions of Dunnett's Procedure, normality and homogeneity of
variance, are formally tested.  The test for normality is the Shapiro-Milk's
Test, and Bartlett's Test is used to test for homogeneity of variance.  If
either of these tests fails, the nonparametric test, Steel's Many-one Rank
Test, is used to determine the NOEC and LOEC endpoints.  If the assumptions of
Dunnett's Procedure are met, the endpoints are estimated by the parametric
procedure.

12.13.2.3  If unequal numbers of replicates occur among the concentration
levels tested, there are parametric and nonparametric alternative analyses.
The parametric analysis is a t test with the Bonferroni adjustment
(see Appendix D).  The Wilcoxon Rank Sum Test with the Bonferroni adjustment
is the nonparametric alternative (see Appendix F).

12.13.2.4  Probit Analysis (Finney, 1971) is used to estimate the
concentration that causes a specified percent decrease in survival from the
control.  In this analysis, the total mortality data from all test replicates
at a given concentration are combined.
                                      130

-------
       STATISTICAL ANALYSIS OF FATHEAD MINNOW EMBRYO-LARVAL
                  SURVIVAL, AND TERATOGENICITY TEST
                             TOTAL MORTALITY
                       TOTAL NUMBER OF DEAD EMBRYOS,
                     DEAD LARVAE, AND DEFORMED LARVAE
PROBIT ANALYSIS
1
r
                                 ARC SINE
                             TRANSFORMATION
    ENDPOINT ESTIMATE
          EC1
    SHAPIRO-WILK-STEST
                                                NON-NORMAL DISTRIBUTION
                  NORMAL DISTRIBUTION
       HOMOGENEOUS
          VARIANCE
                              BARTLETTS TEST
                    I
                              HETEROGENEOUS
                                 VARIANCE
             EQUAL NUMBER OF
               REPLICATES?
          NO
YES
    T-TESTWITH
    BONFERRONI
    ADJUSTMENT
                      EQUAL NUMBER OF
                        REPLICATES?
YES
                                                             NO
        STEEL'S MANY-ONE
           RANK TEST
                                   T
            WILCOXON RANK SUM
                 TEST WITH
           BONFERRONI ADJUSTMENT
                            ENDPOINT ESTIMATES
                                NOEC, LOEC
Figure  4.  Flowchart  for  statistical  analysis  of fathead  minnow,  Pimephales
          promelas, embryo-larval data.
                                    131

-------
12.13.2.5  The data for this example are listed in Table 2.  Total mortality,
expressed as a proportion (combined total number of dead embryos, dead larvae
and deformed larvae divided by the number of embryos at start of test), is the
response of interest.  The total mortality proportion in each replicate must
first be transformed by the arc sine square root transformation procedure
      TABLE 2.  DATA FROM FATHEAD MINNOW, PIMEPHALES PROMELAS,
                EMBRYO-LARVAL TOXICITY TEST WITH GROUND WATER EFFLUENT
  Effluent   No.
   Cone.   Eggs at
   (%)      Start
Dead at Test
Termination
  No.     %
Deformed at Test
Termination
  No.      %
Dead + Deformed
at Termination
 No.      %
Control



3.125



6.25



12.5



25.0



50.0



10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
0
2
0
1
0
0
0
1
0
0
0
0
0
0
0
1
1
2
2
1
4
3
5
3
0
20
0
10
0
0
0
10
0
0
0
0
0
0
0
10
10
20
20
10
40
30
50
30
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
9
8
8
4
6
7
5
7
0
0
0
0
0
10
0
0
0
0
0
10
0
0
0
0
90
80
80
40
60
70
50
70
0
2
0
1
0
1
0
1
0
0
0
1
0
0
0
1
10
10
10
5
10
10
10
10
0
20
0
10
0
10
0
10
0
0
0
10
0
0
0
10
100
100
100
50
100
100
100
100-
                                     132

-------
described in Appendix B.  The raw and transformed data, means and
variances of the transformed observations at each effluent concentration and
control are listed in Table 3.   A plot of the data is provided in Figure 5.
Since there is 100% total mortality in replicates for the 50.0% concentration,
it is not included in this statistical analysis and is considered a
qualitative mortality effect.
    TABLE 3.  FATHEAD MINNOW, PIMEPHALES PROMELAS, EMBRYO-LARVAL TOTAL
              MORTALITY DATA
           Replicate   Control
                                         Effluent Concentration (%)
3.125    6.25
12.5
25.0   50.0
RAW



A
B
C
D
0.00
0.20
0.00
0.10
0.00
0.10
0.00
0.10
0.00
0.00
0.00
0.10
0.00
0.00
0.00
0.10
1.00
1.00
1.00
0.50
1.00
1.00
1.00
1.00
ARC SINE
TRANS-
FORMED

Mean(Y,)
s?
i
A
B
C
D



0.159
0.464
0.159
0.322
0.-276
0.022
1
0.159
0.322
0.159
0.322
0.241
0.009
2
0.159
0.159
0.159
0.322
0.200
0.007
3
0.159
0.159
0.159
0.322
0.200
0.007
4
1.412
1.412
1.412
0.785
1.255
0.098
5
12.13.2.6  Test for Normality

12.13.2.6.1  The first step of the test for normality is to center the
observations by subtracting the mean of all observations within a
concentration from each observation in that concentration.   The centered
observations are summarized in Table 4.

12.13.2.6.2  Calculate the denominator, D,  of the statistic:

                                D = t (X, -X)2
                                    i=i               '

          Where:   Xf  = the ith centered  observation

                   X  = the overall mean  of the centered observations

                   n  = the total  number  of centered  observations
                                     133

-------
o
   ii
I  fel
     m
                    *> *
                                        *   *
                                       *'*
         cq
         o
CD

O
T
 iq
 o
CO
o
i—'—r
CVJ    T-;

O    O
                                                   O>
                                                      O
                                                      IE
                               LU
                            10  O

                               O
                               O
                                   111

                                   u_
                                   u_
                                   LU

                               -co
                                                               o
                                                              •a
                                                               >>
                                                               s-
                                                               o
                                                               ITS
                                                               +J
                                                               o
                             as

                             cu
                             €
                             O

                             Q.
                             
                             
                             -c

                             g-
                                                               3
                                                               O
                               -o
q
o
                                                                 0)
                                                               +j
                                                               o a>
                                                               r— .C
                                                               a. +j
                                                               01
                                                               en
                     ivioi do
                              134

-------
         TABLE 4.  CENTERED OBSERVATIONS FOR SHAPIRO-WILK'S EXAMPLE

Replicate
A
B
C
D

Control
-0.117
0.188
-0.117
0.046

3.125
-0.082
0.081
0.081
-0.082
Effluent
6.25
-0.041
-0.041
-0.041
0.122
Concentration (%)
12.5
-0.041
-0.041
-0.041
0.122
25.0 50.0
0.157
0.157
0.157
-0.470
12.13.2.6.3  For this set of data,    n = 20
                                      X =
                                            1
(-0.003)  =0.000
                                           20

                                      D = 0.4261

12.13.2.6.4  Order the centered observations from smallest to largest
                          <  ... 
-------
 12.13.2.6.5  From Table 4,  Appendix B,  for the number of observations,  n,
 obtain the coefficients a1?  a2, ..., ak where k is n/2 if n is even and
 (n-l)/2 if n is odd.   For'the''data'in "this example,'n = 20 and'k
 values  are listed in  Table 6.
                                                 10.   The a,-
     TABLE 6.  COEFFICIENTS AND DIFFERENCES FOR THE SHAPIRO-WILK'S EXAMPLE
                                                _ w(
1
2
3
4
5
6
7
8
9
10
0.4734
0.3211
0.2565
0.2085
0.1686
0.1334
0.1013
. 0.0711
0.0422
0.0140
0.658
0.274
0.274
0.239
0.204
0.163
0.122
0.122
0.087
0.000
Y(20)
A
X(19)
x<18>
X(17)
X(16)
X(15>
X(14>
x<ซ)
X(12)
X(11)
- x(1)
- x<2)
- x(3)
- XC4)
- x(5)
- x<6)
- x<7)
- x(8)
- x<9>
- xฐฐ>
12.13.2.6.6  Compute the test statistic, W, as follows:
                         W = -
                             D
The differences x(n"l>1> - X(i) are listed  in Table 6.  For the data  in this
example,
                               (0.6004)2
W
                          1
                       0.4261

                     = 0.846

12.13.2.6.7  The decision rule for this test is to compare W as calculated in
Section 13.2.6.6 to a critical value found in Table 6, Appendix B.  If the
computed W is less than the critical value, conclude that the data are not
normally distributed.  For the data in this example, the critical value at a
significance level of 0.01 and n = 20 observations is 0.868.  Since W = 0.846
is less than the critical value, conclude that the data are not normally
distributed.
                                     136

-------
12.13.2.6.8  Since the data do not meet the assumption of normality, Steel's
Many-one Rank Test will be used to analyze the total mortality data.

12.13.2.7  Steel's Many-one Rank Test

12.13.2.7.1  For each control and concentration combination, combine the data
and arrange the observations in order of size from smallest to largest.
Assign the ranks (1, 2, ..., 8) to the ordered observations with a rank of
1 assigned to the smallest observation, rank of 2 assigned to the next larger
observation, etc.  If ties occur when ranking, assign the average rank to each
tied observation.

12.13.2.7.2  An example of assigning ranks to the combined data for the
control and 3.125% effluent concentration is given in Table 7.  This ranking
procedure is repeated for each control/concentration combination.  The
complete set of rankings is summarized in Table 8.  The control group ranks
are next summed for each effluent concentration pairing, as shown in Table 9.

  TABLE 7.  ASSIGNING RANKS TO THE CONTROL AND 3.125% EFFLUENT CONCENTRATION
            FOR STEEL'S MANY-ONE RANK TEST

Rank

2.5
2.5
2.5
2.5
6
6
6
8
Transformed
Proportion
Mortality
0.159
0.159
0.159
0.159
0.322
0.322
0.322
0.464
Ef f 1 uent
Concentration
(%)
Control
Control
3.125
3.125
Control
3.125
3.125
Control
 12.13.2.7.3   For this  example,  we  want  to  determine  if the  total  mortality  in
 any of the effluent concentrations is significantly  higher  than  the  total
 mortality in the control.   If this occurs,  the  rank  sum of  the control  would
 be significantly less  than the rank sum at that concentration.   Thus we are
 only concerned with comparing the  control  rank  sum for each pairing  with the
 various effluent concentrations with some  "minimum"  or critical  rank sum, at
 or below which the concentration total  mortality would be considered
 significantly greater  than the control. At a signficance level  of 0.05, the
 minimum rank sum in a  test with four concentrations  (excluding the control)
 and four replicates per concentration  is 10 (see Table 5, Appendix E).

 12.13.2.7.4   Since the control  rank sum for the 25.0% effluent concentration
 pairing is equal to the critical value, the total proportion mortality  in the
 25.0% concentration is considered  significantly greater than that in the

                                      137

-------
            TABLE 8.  TABLE OF RANKS FOR STEEL'S MANY-ONE RANK TEST
                                          Effluent  Concentration  (%)
   Repl.     Control
3.125
6.25
12.5
25.0
A
B
C
D
0.159 (2.5,3,3,1.5)
0.464 (8,8,8,4)
0.159 (2.5,3,3,1.5)
0.322 (6,6.5,6.5,3)
0.159 (2.5)
0.322 (6)
0.159 (2.5)
0.322 (6)
0.159 (3)
0.159 (3)
0.159 (3)
0.322 (3)
0.159 (3)
0.159 (3)
0.159 (3)
0.159 (3)
1.412 (7)
1.412 (7)
1.412 (7)
0.785 (5)
                            TABLE 9.   RANK SUMS
                      Effluent
                  Concentration  (%)
              Control
              Rank Sum
                        3.125
                        6.25
                        12.5
                        25.0
                19
                20.5
                20.5
                10
control.  Since no other rank sums are less than or equal to the critical
value, no other concentrations have signficantly higher total proportion
mortality than the control.  Hence the NOEC is 12.5% and the LOEC is 25.0%.

12.13.2.8  Calculation of the LC50

12.13.2.8.1  The data used for the Probit Analysis is summarized in Table 10,
To perform the Probit Analysis, run the USEPA Probit Analysis Program.  An
example of the program input and output is supplied in Appendix I.

12.13.2.8.2  For this example, the chi-square test for heterogeneity was not
significant.  Thus Probit Analysis appears appropriate for this data.

12.13.2.8.3  Figure 6 shows the output data for the Probit Analysis of the
data from Table 10 using the USEPA Probit Program.
                                      138

-------
                     TABLE 10.  DATA FOR PROBIT ANALYSIS
                                         Effluent Concentration (%)
                      Control        3.125    6.25     12.5     25.0     50.0
    Number Dead            3           1        0        1        6      15
    Number Exposed        40          40       40       40       40      40
12.14  PRECISION AND ACCURACY

12.14.1  PRECISION

12.14.1.1  Single-laboratory Precision

12.14.1.1.1  Data shown in Tables 11 and 12 indicate that; the precision of the
embryo-larval survival and teratogenicity test, expressed as the relative
standard deviation (or coefficient of variation, CV) of the LCI values, was
62% for cadmium (Table 11) and 41% for Diquat (Table 12).

12.14.1.1.2  Precision data are also available from four embryo-larval
survival and teratogenicity tests on trickling filter pilot plant effluent
(Table 13).  Although the data could not be analyzed by Probit Analysis, the
NOECs and LOECs obtained using Dunnett's Procedure were the same for all four
tests, 7% and 11% effluent, respectively, indicating maximum precision in
terms of the test design.

12.14.1.2  Multilaboratory Precision

12.14.1.2.1  Data on the multilaboratory precision of this test are not yet
available.

12.14.2  ACCURACY

12.14.2.1  The accuracy of toxicity tests cannot be determined.
                                      139

-------
                        USEPA PROBIT ANALYSIS PROGRAM
                      USED FOR CALCULATING LC/EC VALUES
                                 Version 1.5
          Probit Analysis of Fathead Minnow Embryo-Larval Survival
                           and  Teratogenicity Data
       Cone.
      Control
       0.5000
         .0000
         .0000
         .0000
1,
2.
4.
       8.0000
 Number
Exposed

   20
   20
   20
   20
   20
   20
Number
Resp.

   2
   2
   1
   4
  16
  20
  Observed
Proportion
Responding

  0.1000
  0.1000
  0.0500
  0.2000
  0.8000
  1.0000
  Proportion
  Responding
Adjusted for
   Controls

    0.0000
    0.0174
    -.0372
    0.1265
    0.7816
    1.0000
Chi - Square for Heterogeneity (calculated)     =   0.441
Chi - Square for Heterogeneity (tabular value)  =  7.815
          Probit Analysis of Fathead Minnow Embryo-Larval Survival
                           and  Teratogenicity  Data
      Estimated LC/EC Values and Confidence Limits
Point
LC/EC  1.00
LC/EC 50.00
            Exposure
             Cone.

             1.346
             3.018
                Lower           Upper
                95% Confidence Limits
                0.453
                2.268
                    1.922
                    3.672
          Figure 6.  Output for USEPA Probit Program, Version 1.5.
                                    140

-------
TABLE 11.
PRECISION OF THE FATHEAD MINNOW, PIMEPHALES PROMELAS,
EMBRYO-LARVAL SURVIVAL AND TERAJOGENICITY TEST, USING
ranMTiiM flc A Dctrcocwrr TnvirAMT1'2
           CADMIUM AS A REFERENCE
Test
1
2
3
4
5
N
Mean
SD
CV(%)
LCI3
(mg/L)
0.014
0.006
0.005
0.003
0.006
5
0.0068
0.0042
62
95% Confidence NOEC4
Limits (mg/L)
0.009 - 0.018 0.012
0.003 - 0.010 0.012
0.003 - 0.009 0.013
0.002 - 0.004 0.011
0.003 - 0.009 0.012
5
NA
NA
Tests conducted by Drs. Wesley Birge and Jeffrey Black, University
of Kentucky, Lexington, under a cooperative agreement with the
Bioassessment and Ecotoxicology Branch, EMSL, USEIPA, Cincinnati, OH.
Cadmium chloride was used as the reference toxicant.  The nominal
concentrations, expressed as cadmium (mg/L), were: 0.01, 0.032,
0.100, 0.320, and 1.000.  The dilution water was reconstituted water
with a hardness of 100 mg/L as calcium carbonate, and a pH of 7.8.
Determined by Probit Analysis.
Highest no-observed-effect concentration determined by independent
statistical analysis (2X2 Chi-square Fisher's Exact Test).  NOEC
range of 0.011 - 0.013 represents a difference of one exposure
concentration.
                              141

-------
TABLE 12.  PRECISION OF THE FATHEAD MINNOW, PIMEPHALES
         •  PROMELAS, EMBRYO-LARVAL, SURVIVAL AND
           TERATOGENICITY TOXICITYTEST, USING DIQUAT
           AS A REFERENCE TOXICANT1'2
Test
1
2
3
4
5
N
Mean
SD
CV(%)
LCI3
(mg/L)
0.58
2.31
1.50
1.71
1.43
5
1.51
0.62
41.3
95% Confidence
Limits
0.32 - 0.86
--4
1.05 - 1.87 ;
1.24 - 2.09
0.93 - 1.83

     Tests conducted by  Drs.  Wesley Birge and Jeffrey Black,  Uni-
     versity of Kentucky, Lexington, under a  cooperative  agreement
     with the Bioassessment and Ecotoxicology Branch,  EMSL,  USEPA,
     Cincinnati, OH.
     The Diquat concentrations were determined by chemical  analysis.
     The dilution water was reconstituted water with  a hardness  of
     100 mg/L as calcium carbonate, and a pH of 7.8.
     Determined by Probit Analysis.
     Cannot be calculated.
                             142

-------
TABLE 13.  PRECISION OF THE FATHEAD MINNOW, PIMEPHALES PROMELAS,
           EMBRYO-LARVAL SURVIVAL AND TERATOGENICITY STATIC-RENEWAL
           TEST CONDUCTED WITH TRICKLING FILTER EFFLUENT1'^'3
Test
No.
1
2
3
4
NOEC
(% Effluent)
7
7
7
7
LOEC
(% Effluent)
11
: 11
11
; n
        Data provided by Timothy Neiheisel,  Bioassessrnent and
        Ecotoxicology Branch, EMSL, USEPA, Cincinnati, OH.
        Effluent  concentrations  used:   3,  5,  7,  11  and 16%.
        Maximum precision achieved in  terms  of NOEC-LOEC interval
        For a discussion of the  precision  of data from chronic
        toxicity  tests (see Section 4,  Quality Assurance).
                               143

-------
                                  SECTION 13

                                  TEST METHOD

                         DAPHNID, CERIODAPHNIA DUBIA,
                        SURVIVAL AND REPRODUCTION TEST
                                 METHOD 1002.0
13.1  SCOPE AND APPLICATION

13.1.1  This method measures the chronic toxicity of effluents and receiving
water to the daphnid, Cen'odaphm'a dubia, using less than 24 h old neonates
during a three-brood (seven-day), static renewal test.  The effects include
the synergistic, antagonistic, and additive effects of all the chemical,
physical, and biological components which adversely affect the physiological
and biochemical functions of the test organisms.

13.1.2  Daily observations on mortality make it possible to also calculate
acute toxicity for desired exposure periods (i.e., 24-h, 48-h, and 96-h
LCBOs).

13.1.3  Detection limits of the toxicity of an effluent or pure substance are
organism dependent.

13.1.4  Brief excursions in toxicity may not be detected using 24-h composite
samples.  Also, because of the long sample collection period involved in
composite sampling, and because the test chambers are not sealed, highly
degradable or highly volatile toxicants in the source may not be detected in
the test.

13.1.5  This test method is commonly used in one of two forms: (1) a
definitive test, consisting of a minimum of five effluent concentrations and a
control, and (2) a receiving water test(s), consisting of one or more
receiving water concentrations and a control.

13.2  SUMMARY OF METHOD

13.2.1  Cen'odaphm'a dubia are exposed in a static renewal system to different
concentrations of effluent, or to receiving water, until 60% of surviving
control organisms have three broods of offspring.  Test results are based on
survival and reproduction.  If the test is conducted as described, the
surviving control organisms should produce 15 or more young in three broods.
If these criteria are not met at the end of 8 days, the test must be repeated.

13.3  INTERFERENCES

13.3.1  Toxic substances may be introduced by contaminants in dilution water,
glassware, sample hardware, and testing equipment (see Section 5, Facilities,
Equipment, and Supplies).
                                      144

-------
13.3.2  Improper effluent sampling and handling may adversely affect test
results (see Section 8, Effluent and Receiving Water Sampling, Sample
Handling,  and Sample Preparation for Toxicity Tests).

13.3.3  Pathogenic and/or predatory organisms in the dilution water and
effluent may affect test organism survival and confound test results.

13.3.4  The amount and type of natural food in the effluent or dilution water
may confound test results.

13.3.5  Food added during the test may sequester metals and other toxic
substances and confound test results.  Daily renewal of solutions, however,
will reduce the probability of reduction of toxicity caused by feeding.

13.4  SAFETY

13.4.1  See Section 3, Health and Safety.

13.5  APPARATUS AND EQUIPMENT

13.5.1  Ceriodaphnia and algal culture units -- See Ceriodaphnia and algal
culturing methods below and algal culturing methods in Section 14 and USEPA,
1993b.

13.5.2  Samplers -- automatic sampler, preferably with sample cooling
capability, capable of collecting a 24-h composite sample of 5 L or more.

13.5.3  Sample containers -- for sample shipment and storage (see Section 8,
Effluent and Receiving Water Sampling, Sample Handling, and Sample Preparation
for Toxicity Tests).

13.5.4  Environmental chambers, incubators, or equivalent facilities with
temperature control (25 ฑ 1ฐC).                ,

13.5.5  Water purification system -- MILLIPORE MILLI-Qฎ, deionized water or
equivalent (see Section 5, Facilities, Equipment, and Supplies).

13.5.6  Balance -- analytical, capable of accurately weighing 0.00001 g.

13.5.7  Reference weights, Class S -- for checking performance of balance.
Weights should bracket the expected weights of the material to be weighed.

13.5.8  Test chambers --10 test chambers are required for each concentration
and control.  Test chambers such as 30-mL borosilicate glass beakers or
disposable polystyrene cups are recommended because they will fit in the
viewing field of most stereoscopic microscopes.  The glass beakers and plastic
cups are rinsed thoroughly with dilution water before use.  To avoid potential
contamination from the air and excessive evaporation of the test solutions
during the test, the test vessels should be covered with safety glass plates
or  sheet plastic (6 mm thick).

13.5.9  Mechanical shaker or magnetic stir plates --for algal cultures.

                                      145

-------
13.5.10  Light meter -- with a range of 0-200 /iE/m2/s (0-1000 ft-c).

13.5.11  Fluorometer (optional) -- equipped with chlorophyll detection light
source, filters, and photomultiplier tube (Turner Model 110 or equivalent).

13.5.12  UV-VIS spectrophotometer (optional) -- capable of accommodating 1-5
cm cuvettes.

13.5.13  Cuvettes for spectrophotometer -- 1-5 cm light path.

13.5.14  Electronic particle counter (optional) --  Coulter Counter,  ZBI, or
equivalent, with mean cell (particle) volume determination.

13.5.15  Microscope with 10X, 45X, and 100X objective lenses, 10X ocular
lenses, mechanical stage, substage condenser, and light source (inverted or
conventional microscope) -- for determining sex and verifying identification.

13.5.16  Dissecting microscope, stereoscopic, with zoom objective,
magnification to 50X -- for examining and counting the neonates in the test
vessels.

13.5.17  Counting chamber -- Sedgwick-Rafter, Palmer-Maioney, or
hemocytometer.

13.5.18  Centrifuge (optional) -- plankton, or with swing-out buckets having a
capacity of 15-100 ml.

13.5.19  Centrifuge tubes -- 15-100 ml, screw-cap.

13.5.20  Filtering apparatus -- for membrane and/or glass fiber filters.

13.5.21  Racks (boards) -- to hold test chambers.  It is convenient to use a
piece of styrofoam insulation board, 50 cm x 30 cm x 2.5 cm (20 in x 12 in x 1
in), drilled to hold 60 test chambers, in six rows of 10 (see Figure 1).

13.5.22  Light box -- for illuminating organisms during examination.

13.5.23  Volumetric flasks and graduated cylinders -- class A, borosilicate
glass or non-toxic plastic labware, 10-1000 mL, for culture work and
preparation of test solutions.

13.5.24  Pipettors, adjustable volume repeating dispensers -- for feeding.
Pipettors such as the Gilson REPETMANฎ, Eppendorf, Oxford, or equivalent,
provide a rapid and accurate means of dispensing small volumes (0.1 mL) of
food to large numbers of test chambers.

13.5.25  Volumetric pipets -- class A, 1-100 mL.

13.5.26  Serological pipets -- 1-10 mL, graduated.

13.5.27  Pipet bulbs and fillers -- PROPIPETฎ, or equivalent.


                                      146

-------
13.5.28  Disposable polyethylene pipets, droppers, and glass tubing with
fire-polished edges * > 2mm ID -- for transferring organisms.
13.5.29  Wash bottles -- for rinsing small glassware and instrument electrodes
and probes.
13.5.30  Thermometer, glass or electronic, laboratory grade, -- for measuring
water temperatures.
13.5.31  Bulb-thermograph or electronic-chart type thermometers -- for
continuously recording temperature.
13.5.32  Thermometer, National Bureau of Standards Certified (see USEPA Method
170.1, USEPA 1979b) --to calibrate laboratory thermometers.
13.5.33  Meters, DO, pH, and specific conductivity -- for routine physical and
chemical measurements.
13.6  REAGENTS AND CONSUMABLE MATERIALS
13.6.1  Sample containers -- for sample shipment and storage (see Section 8,
Effluent and Receiving Water Sampling, Sample Handling, and Sample Preparation
for Toxicity Tests).
13.6.2  Data sheets (one set per test) -- for recording the data.
13.6.3  Vials, marked -- for preserving specimens for verification (optional).
13.6.4  Tape, colored -- for labeling test vessels.
13.6.5  Markers, waterproof -- for marking containers.
13.6.6  Reagents for hardness and alkalinity tests -- see USEPA Methods 130.2
and 310.1, USEPA, 1979b.
13.6.7  Buffers, pH 4, pH 7, and pH 10 (or as per instructions of instrument  .
manufacturer) -- for instrument calibration check (see USEPA Method 150.1,
USEPA, 1979b).
13.6.8  Specific conductivity standards -- see USEPA Method 120.1, USEPA,
1979b.
13.6.9  Membranes and filling solutions for DO probe (see USEPA Method 360.1*
USEPA, 1979b), or reagents -- for modified Winkler analysis.
13.6.10  Laboratory quality control samples and standards -- for calibration
of the above methods.
13.6.11  Reference toxicant solutions -- see Section 4, Quality Assurance.
                                      147

-------
13.6.12  Reagent water -- defined as distilled or deionized water that does
not contain substances which are toxic to the test organisms (see Section 5,
Facilities, Equipment, and Supplies).

13.6.13  Effluent, surface water, and dilution water -- see Section 7,
Dilution Water; and Section 8, Effluent and Receiving Water Sampling, Sample
Handling, and Sample Preparation for Toxicity Tests.

13.6.14  Trout chow, yeast, and CEROPHYLฎ food (or substitute food) --for
feeding the cultures and test organisms.

13.6.14.1  Digested trout chow, or substitute flake food (TETRAMINฎ, BIORILฎ,
or equivalent), is prepared as follows:

   1. Preparation of trout chow or substitute flake food requires one week.
      Use starter or No. ltpellets prepared according to current U.S. Fish and
      Wildlife Service specifications.  Suppliers of trout chow include
      Zeigler Bros., Inc., P.O. Box 95, Gardners, PA, 17324 (717-780-9009);
      Glencoe Mills, 1011 Elliott, Glencoe, MN, 55336 (612-864-3181); and
      Murray Elevators, 118 West 4800 South, Murray, UT 84107 (800-521-9092).
   2. Add 5.0 g of trout chow pellets or substitute flake food to 1 L.of
      MILLI-Qฎ water.  Mix well in a blender and pour into a 2-L separatory
      funnel.  Digest prior to use by aerating continuously from the bottom of
      the vessel for one week at ambient laboratory temperature.  Water lost
      due to evaporation is replaced during digestion.  Because of the
      offensive odor usually produced during digestion, the vessel should be
      placed in a fume hood or other isolated, ventilated area.
   3. At the end of digestion period, place in a refrigerator and allow to
      settle for a minimum of 1 h.  Filter the supernatant through a fine mesh
      screen (i.e., NITEXฎ 110 mesh).  Combine with equal volumes of
      supernatant from CEROPHYLLฎ and yeast preparations (below).  The
      supernatant can be used fresh, or frozen until use.  Discard the
      sediment.

13.6.14.2  Yeast is prepared as follows:

   1. Add 5.0 g of dry yeast, such as FLEISCHMANN'Sฎ Yeast, Lake State Kosher
      Certified Yeast, or equivalent, to 1 L of MILLI-Qฎ water.
   2. Stir with a magnetic stirrer, shake vigorously by hand, or mix with a
      blender at low speed, until the yeast is well dispersed.
   3. Combine the yeast suspension immediately (do not allow to settle) with
      equal volumes of supernatant from the trout chow (above) and CEROPHYLLฎ
      preparations (below).  Discard excess material.

13.6.14.3  CEROPHYLLฎ is prepared as follows:

   1. Place 5.0 g of dried, powdered, cereal or alfalfa leaves, or rabbit
      pellets, in a blender.  Cereal leaves are available as "CEREAL LEAVES,"
      from Sigma Chemical Company, P.O. Box 14508, St. Louis, MO 63178;
      800-325-3010; or as CEROPHYLLฎ, from Ward's Natural Science
      Establishment, Inc., P.O. Box 92912, Rochester, NY 14692-9012;
                                      148

-------
       716-359-2502.  Dried, powdered, alfalfa leaves may be obtained from
       health food stores, and rabbit pellets are available at pet shops.
    2. Add 1 L of MILLI-Qฎ water.
    3. Mix in a blender at high speed for 5 min, or stir overnight at medium
       speed on a magnetic stir plate.
    4. If a blender is used to suspend the .material, place in a refrigerator
       overnight to settle.  If a magnetic stirrer is used, allow to settle for
       1 h.  Decant the supernatant and combine with equal volumes of
       supernatant from trout chow and yeast preparations (above).  Discard
       excess material.

 13.6.14.4  Combined yeast-cerophyl-trout chow (YCT) is mixed as follows:

    1. Thoroughly mix equal (approximately 300 ml)  volumes of the three foods
       as described above.
 ,   2. Place aliquots of the mixture in small  (50 ml to 100 ml)  screw-cap
       plastic bottles and freeze until  needed.
    3. Freshly prepared food can  be used immediately,  or it can  be frozen until
       needed.   Thawed food is stored in the refrigerator between feedings,  and
       is used for a maximum of two weeks.   Do not  store frozen  over three
       months.
    4. It is advisable to measure the dry weight of solids in each batch  ;of YCT
       before use.   The food should contain 1.7-1.9 g  solids/I.   Cultures or
       test solutions should contain 12-13  mg  solids/L.

 13.6.15  Algal  food --  for feeding the  cultures and test organisms.

 13.6.15.1   Algal  Culture Medium  is prepared as  follows:

    1.  Prepare  (five)  stock nutrient solutions using reagent  grade chemicals  as
       described  in  Table 1.
    2.  Add  1  ml of  each  stock solution,  in  the order listed  in Table  1, to
       approximately 900  ml of  MILLI-Qฎ water.   Mix  well  after the addition  of
       each  solution.   Dilute to  1  L,  mix well.  The final  concentration  of
       macronutrients  and  micronutrients  in  the  culture medium is  qiven in
       Tab!e  2.                                                                .
    3.  Immediately  filter  the medium through a 0.45 urn pore diameter membrane
       at a  vacuum of  not  more  than  380 mm  (15 in.) mercury, or at  a pressure
       of not more than one-half atmosphere  (8 psi).  Wash the filter with 500
      mL deionized water  prior to  use.
    4.  If the filtration  is carried  out with sterile apparatus, filtered medium
      can be used immediately, and  no further sterilization steps  are required
      before the inoculation of the medium.  The medium can also be sterilized
      by autoclaving  after it  is placed in the culture vessels.
    5. Unused sterile medium  should not be stored more than one week prior to
      use, because there may be substantial loss of water by evaporation.

13.6.15.2  Algal Cultures

13.6.15.2.1  See Section 6, Test  Organisms, for information on sources of
 starter  cultures of Selenastrum capricornutum, 5. minutum, and Chlamydomonas
reinhardti.

                                      149

-------
TABLE  1.   NUTRIENT STOCK SOLUTIONS FOR MAINTAINING ALGAL STOCK CULTURES
STOCK
SOLUTION
             COMPOUND
 AMOUNT DISSOLVED IN
500 mL MILLI-Qฎ WATER
1. MACRONUTRIENTS

   A.



   B.

   C.

   D.

2. MICRONUTRIENTS
               MgCl2.6H20
               CaCl2'2H20
               NaN03

               MgS04-7H20

               K2HP04

               NaHCO,
                   H,BO,
                   MnCl2.4H20
                   ZnCl2
                   FeCl,.6H20
                   CoCl2.6H20
                   Na2MoO,.2H20
                   CuCl2.2H20
                   Na2EDTA-2H20
                   Na2Se04
       6.08  g
       2.20  g
      12.75  g

       7.35  g

       0.522 g

       7.50  g
                                                 92.8
                                                208.0
                                                  1.64
             mg
             mg
                                                 79.9
             mg
             mg_
        0.714 ing,
        3.63  mg
        0.006 mg
      150.0   mg
        1.196 mg5
                                                    Add 1 mL of this
                                                          Add  1  mL of

                                                          Add 1 mL
ZnCl2 - Weigh out 164 mg and dilute to 100 mL.
solution to Stock 2, micronutrients.
CoCl2-6H20  -  Weigh  out 71.4 mg and dilute to 100 mL.
this solution to Stock 2, micronutrients.
NaJloO/^H^  - Weigh  out  36.6  mg  and  dilute  to 10 mL.
of this solution to  Stock 2,  micronutrients.
CuCV2H?0  -  Weigh  out 60.0 mg and dilute to 1000 mL.   Take 1 mL of
this solution and dilute to 10 mL.   Take  1  mL of the  second
dilution and add to  Stock 2,  micronutrients.                  ^   .
Na2Se04 - Weigh  out 119.6 mg and dilute to 100 mL.  Add 1 mL or this
solution to  Stock 2,  micronutrients.
                                     150

-------
TABLE 2.  FINAL CONCENTRATION OF MACRONUTRIENTS AND MICRONUTRIENTS
          IN THE CULTURE MEDIUM
MACRONUTRIENT
NaN03
MgCT2.6H20
CaCl2-2H20
MgS04-7H20
K2HP04
NaHC03


MICRONUTRIEN.T
H3B03
MnCT2.4H20
ZnC12
CoCl2ซ6H20
CuCl2-2H20
Na2Mo04-2H20
FeCl3.6H20
Na2EDTA-2H20
Na2Se04
CONCENTRATION
fma/U
25.5
12.2
4.41
14.7
1.04
15.0


CONCENTRATION
fua/n
185.0
416.0
3.27
1.43
0.012
7.26
160.0
300.0
2.39
ELEMENT
N \
\
Mg
Ca !
i
S
P
Na
K
C i
ELEMENT
B
Mn
Zn
Co
Cu
Mo
Fe
--
Se
CONCENTRATION
(ma/L)
4.20
2.90
1.20
1.91
0.186
11.0
0.469
2.14
CONCENTRATION
lua/L}
32.5
115.0
1.57
0.354
0.004
2.88
33.1
	
0.91
                              151

-------
13.6.15.2.2  Two types of algal cultures are maintained:  "stock"  cultures,  and
"food" cultures.

13.6.15.2.2.1  Establishing and Maintaining Stock Cultures of Algae:

   1  Upon receipt of the "starter" culture (usually about 10 ml),  a stock
    ' culture is initiated by aseptically transferring one milliliter to each
      of several 250-mL culture flasks containing 100 ml algal culture medium
      (prepared as described above).  The remainder of the starter culture can
      be held in reserve for up to six months in a refrigerator (in the dark)
      at 4 ฐ C.
   2  The stock cultures are used as a source of algae to initiate "food"
      cultures for Ceriodaphnia dubia toxicity tests.  The volume of stock
      culture maintained at any one time will depend on the amount of algal
     • food required for the Ceriodaphnia dubia cultures and tests,  stock
      culture volume may be rapidly "scaled up" to several liters, if.
      necessary, using 4-L serum bottles or similar vessels,  each containing 6
      I of growth medium.                                            . .  .   .
   3  Culture temperature is not critical.  Stock cultures may be maintained
      at 25ฐC in environmental chambers with cultures of other organisms if
      the illumination is adequate  (continuous "cool-white" fluorescent
      lighting  of approximately 86 ฑ 8.6 ^E/m/s, or 400 ft-c).
   4. Cultures  are mixed twice daily by hand.
   5. Stock  cultures  can be held in the refrigerator until used to start
      "food" cultures, or can  be transferred to new medium weekly.
      One-to-three milliliters of  7-day old algal stock culture, containing
      approximately  1.5 X  106  cells/ml, are transferred to each 100 ml of
      fresh  culture  medium.  The inoculum  should provide  an  initial  cell
      density of  approximately 10,000-30,000 cells/ml  in  the  new stock
      cultures.   Aseptic techniques  should be  used  in  maintaining  the  stocK
      algal  cultures,  and  care should  be exercised  to  avoid  contamination  by
      other  microorganisms.                                  _•-
    6 Stock  cultures should  be examined microscopically weekly,  at transfer,
      for  microbial  contamination.   Reserve quantities of culture  organisms
      can  be maintained  for  6-12 months if stored  in  the  dark at  4ฐC.   It  is
       advisable to  prepare new stock cultures  from  "starter"  cultures  obtained
       from established outside sources of  organisms (see  Section  6,  Test
       Organisms)  every four  to six months.

 13.6.15.2.2.2   Establishing  and Maintaining "Food"  Cultures of Algae:

    1  "Food" cultures are started seven days prior to use for Ceriodaphnia
       dubia cultures and tests.   Approximately 20 ml of 7-day-old algaljtock
       culture (described in the previous paragraph),  containing 1.5 X 10
       cells/ml, are added to each liter of fresh algal culture medium (i.e.,
       3 L of medium in a 4-L bottle, or 18 L in a 20-L bottle).  The inoculum
       should provide an initial cell density of approximately 30,000 cells/ml.
       Aseptic techniques should be used in preparing and maintaining the
       cultures, and care should be exercised to avoid contamination by other
       microorganisms.  However, sterility of food cultures is not as critical
       as in stock cultures because the food cultures are terminated in
                                       152

-------
       7-10  days.   A one-month supply of algal  food can be grown at one time,
       and stored  in the refrigerator.
    2.  Food  cultures may be maintained  at 25ฐC  in environmental  chambers with
       the algal stock cultures or cultures of  other organisms if the
       illumination is adequate (continuous "cool-white"  fluorescent lighting
       of approximately 86  ฑ 8.6 //E/m/s  or 400 ft-c).
    3.  Cultures .are mixed continuously  on a magnetic stir plate  (with a medium
       size  stir bar)  or in a moderately aerated  separatory funnel,  or are
       mixed twice  daily by hand.   If the cultures are  placed  on a magnetic
       stir  plate,  heat generated by  the stirrer  might  elevate the culture
       temperature  several  degrees.   Caution should be  exercised to prevent the
       culture temperature  from rising  more than  2-3ฐC.

 13.6.15.2.3 Preparing Algal  Concentrate for Use as Ceriodaphnia dubia Food:

    1.  An algal concentrate containing  3,0 to 3.5 X 107 cells/ml  is  prepared
       from  food cultures by centrifuging the algae with  a plankton  or
       bucket-type  centrifuge,  or by  allowing the cultures to  settle in a
       refrigerator for at  least three  weeks and  siphoning off the supernatant.
    2.  The cell density (cells/mL)  in the concentrate is  measured with an
       electronic particle  counter, microscope  and hemocytometer,  fluorometer,
       or spectrophotometer (see Section  14,  Green Alga,  Selenastrum
       capn'cornutum Growth Test),  and  used to  determine  the dilution  (or
       further concentration)  required  to achieve  a final  cell count of 3.0 to
       3.5 X loVmL.
    3.  Assuming a cell  density  of approximately 1.5 X 106 cells/ml in the
       algal  food cultures  at  7  days, and 100%  recovery in  the concentration
       process, a 3-L,  7-10 day  culture will  provide 4.5  X  109 algal cells.
       This  number  of  cells would  provide approximately 150 ml of algal  cell
       concentrate  (1500  feedings  at 0.1  ml/feeding)  for  use as  food.   This
       would  be enough  algal  food  for four Ceriodaphm'a dubia  tests.
    4.  Algal  concentrate  may be  stored  in  the refrigerator  for one month.

 13.6.15.3   Food Quality

 13.6.15.3.1  USEPA  recommends  Fleishmann'sฎ yeast,  Cerophyllฎ, trout  chow, and
Selenastrum  capn'cornutum  as the  preferred  Ceriodaphnia dubia food
combination.  This  recommendation is based  on extensive data  developed  by many
laboratories which  indicated high Ceriodaphnia dubia survival and reproduction
 in  culturing and testing.   The  use of  substitute  food(s)  is acceptable  only
after  side-by-side  tests are conducted to determine that the  quality of  the
substitute food(s)  is  equal to  the USEPA  recommended food combination based on
survival and reproduction  of Ceriodaphnia dubia.
                                                          \
13.6.15.3.2  The quality of food prepared with newly acquired supplies of
yeast, trout chow, dried cereal leaves,  algae,  and/or any substitute food(s)
should be determined in  side-by-side comparisons  of Ceriodaphnia dubia
survival and reproduction,   using the new  food and  food of known, acceptable
quality, over a seven-day  period in control medium.
                                      153

-------
13.6.16  TEST ORGANISMS, DAPHNIDS, CERIODAPHNIA DUBIA

13.6.16.1  Cultures of test organisms should be started at lea:st three weeks
before the brood animals are needed, to ensure an adequate supply of neonates
for the test.  Only a few individuals are needed to start a culture because of
their prolific reproduction.

13.6.16.2  Neonates used for toxicity tests must be obtained from individually
cultured organisms.  Mass cultures may be maintained, however, to serve as a
reserve source of organisms for use in initiating individual cultures and in
case of loss of individual cultures.

13.6.16.3  Starter animals may be obtained from commercial sources and may be
shipped in polyethylene bottles.  Approximately 40 animals and 3 ml of food
are placed in a 1-L bottle filled full with culture water for shipment.
Animals received from an outside source should be transferred to new culture
media gradually over a period of 1-2 days to avoid mass mortality.

13.6.16.4  It is best to start the cultures with one animal, which is
sacrificed after producing young, mounted on a microscope slide, and retained
as a permanent slide mount to facilitate identification and permit future
reference.  The species identification of the stock culture should be verified
by preparing slide mounts, regardless of the number of animals used to start
the culture.  The following procedure is recommended for making slide mounts
of Cen'odaphm'a dubia (modified from Beckett and Lewis, 1982):

    1.   Pi pet the animal onto a watch glass.
    2.   Reduce the water volume by withdrawing excess water with the pi pet.
    3.   Add a few drops of carbonated water.(club soda or seltzer water) or
         70% ethanol to relax the specimen so that the post-abdomen is
         extended.   (Optional:  with practice, extension of the postabdomen
         may be accomplished by putting pressure on the cover slip).
    4.   Place a small  amount  (one to three drops) of mounting medium on a
         glass microscope slide.  The recommended mounting medium is CMCP-9/10
         Medium , prepared by mixing two parts of CMCP-9 with one part of
         CMCP-10 stained with enough acid fucKsin dye to color the mixture  a
         light pink.  For more viscosity and faster drying, CMC-10 stained
         with acid fuchsin  may be used.
    5.   Using forceps  or a pipet,  transfer the animal to the drop of mounting
         medium on the  microscope slide.
    6.   Cover with  a 12 mm round cover slip and exert minimum pressure to
         remove any  air bubbles trapped under the cover  slip.  Slightly more
         pressure will  extend  the postabdomen.
    7.   Allow mounting medium to dry.
    8.   Make slide  permanent  by  placing varnish around  the edges of  the
         covers!ip.
          CMCP-9,  CMCP-10 and Acid Fuchsin are available from Polysciences, Inc.,
          Paul  Valley  Industrial  Park,  Harrington,  PA,  18976,  215-343-6484.
          Neonates from mass cultures  are  not  to be used directly  in  toxicity
          tests (see Subsection 13.10.2.3).

                                       154

-------
    9.   Identify to species (see Pennak, 1978; Pennak, 1989:; and Berner,
         1986).
   10.   Label with waterproof ink or diamond pencil.
   11.   Store for permanent record.

13.6.16.5  Mass Culture

13.6.16.5.1  Mass cultures are used only as a "backup" reservoir of organisms.

13.6.16.5.2  One-liter or 2-1 glass beakers, crystallization dishes, "battery
jars," or aquaria may be used as culture vessels.  Vessels are commonly filled
to three-fourths capacity.  Cultures are fed daily.  Four or more cultures are
maintained in separate vessels and with overlapping ages to serve as back-up
in case one culture is lost due to accident or other unanticipated problems,
such as low DO concentrations or poor quality of food or laboratory water.

13.6.16.5.3  Mass cultures which will serve as a source of brood organisms for
individual culture should be maintained in good condition by frequent renewal
with new culture medium at least twice a week for two weeks.  At each renewal,
the adult survival is recorded, and the offspring and the old medium are
discarded.  After two weeks, the adults are also discarded, and the culture is
re-started with neonates in fresh medium.  Using this schedule, 1-L cultures
will produce 500 to 1000 neonate Cen'odaphm'a dubia each week.

13.'6.16.6  Individual Culture               -              ;

13.6.16.6.1  Individual cultures are used as the immediate source of neonates
for toxicity tests.

13.6.16.6.2  Individual organisms are cultured in 15 mL of culture medium in
30-mL (1 oz) plastic cups or 30-mL glass beakers.  One neonate is placed in
each cup.  It is convenient to place the cups in the same type of board used
for toxicity tests (see Figure 1).

13.6.16.6.3  Organisms are fed daily (see Subsection 13.6.16.9) and are
transferred to fresh medium a minimum of three times a week, typically on
Monday, Wednesday, and Friday.   On the transfer days, food is added to the new
medium immediately Before or after the organisms are transferred.

13.6.16.6.4  To provide cultures of overlapping ages, new boards are started
weekly, using neonates from adults which produce at least eight young in their
third or fourth brood.  These adults can be used as sources of neonates until
14 days of age.  A minimum of two boards are maintained concurrently to
provide backup supplies of organisms in case of problems.

13.6.16.6.5  Cultures which are properly maintained should produce at least
20 young per adult in three broods (seven days or less).   Typically,  60 adult
females (one board) will  produce more than the minimum number of neonates
(120) required for two tests.

13'.6.16.6.6  Records should be  maintained on the survival  of brood organisms
and number of offspring at each renewal.   Greater than 20% mortality of
                                                          i
                                     155

-------
                                              > ฃ>,P
                   JO.Q JOJO.Q.O.O jo 0,0
                   SO.O.CXCXO.O,0.,0
                    [CO) (CO) (<*,
                            co

                            
-------
adults, or less than an average of 20 young per female would  indicate
problems, such as poor quality of culture media or food.  Cultures that do not
meet these criteria should not be used as a source of test organisms.

13.6.16.7  Culture Medium

13.6.16.7.1  Moderately hard synthetic water prepared using MILLIPORE MILLI-Qฎ
or equivalent deionized water and reagent grade chemicals or  20% DMW is
recommended as a standard culture medium (see Section 7, Dilution Water).

13.6.16.8  Culture Conditions

13.6.16.8.1  The daphnid, Cenodaphnia dubia, should be cultured at a
temperature of 25 ฑ 1ฐC.

13.6.16.8.2  Day/night cycles prevailing in most laboratories will provide
adequate illumination for normal growth and reproduction.  A  photoperiod of
16-h of light and 8-h of darkness is recommended.  Light intensity should be
10-20 /iE/mz/s or 50 to 100 ft-c.

13.6.16.8.3  Clear, double-strength safety glass or 6 mm plastic panels are
placed on the culture vessels to exclude dust and dirt, and reduce
evaporation.

13.6.16.8.4  The organisms are delicate and should be handled as carefully and
as little as possible so that they are not unnecessarily stressed.  They are
transferred with a pipet of approximately 2-mm bore, taking care to release
the animals under the surface of the water.  Any organism that is injured
during handling should be discarded.

13.6.16.9  Food and Feeding

13.6.16.9.1  Feeding the proper amount of the right food is extremely
important in Ceriodaphnia dubia culturing.   The key is to provide sufficient
nutrition to support normal reproduction without adding excess food which may
reduce the toxicity of the test solutions,  clog the animal's filtering
apparatus, or greatly decrease the DO concentration and increase mortality.
A combination of Yeast, CEROPHYLLฎ,  and Trout chow (YCT), along with the
unicellular green alga, Selenastrum capn'cornutum,  will  provide suitable
nutrition if.fed daily.

13.6.16.9.2  Other algal  species (such as S.  minutum or Ch1amydomonas
reinhardti), other substitute food combinations (such as Flake Fish Food),  or
different feeding rates may be acceptable as  long as performance criteria are
met and side-by-side comparison tests confirm acceptable quality (see
Subsection 13.6.15.3).

13.6.16.9.3  Cultures should be fed  daily to  maintain the organisms in optimum
condition so as to provide maximum reproduction.   Stock cultures which are
stressed because they are not adequately fed  may produce low numbers of young,
large numbers of males, and/or ephippial  females.   Also,  their offspring may
produce few young when used in toxicity tests.

                                     157

-------
13.6.16.9.4  Feed as follows:

  1. If YCT is frozen, remove a bottle of food from the freezer 1 h before
     feeding time, and allow to thaw.
  2. YCT food mixture and algal concentrates should both be thoroughly mixed
     by shaking before dispensing.
  3. Mass cultures are fed daily at the rate of 7 ml YCT and 7 ml algae
     concentrate/L culture.
  4. Individual cultures are fed at the rate of 0.1 ml YCT and 0.1 ml algae
     concentrate per 15 ml culture.
  5. Return unused YCT food mixture and algae concentrate to the refrigerator.
     Do not re-freeze YCT.  Discard unused portion after two weeks.

13.6.16.10  It is recommended that chronic toxicity tests be performed monthly
with a reference toxicant.  Daphnid, Cen'odaphnia dubia, neonates less than'
24 h old, and all within 8 h of the same age are used to monitor the chronic
toxicity of the reference toxicant to the Cen'odaphnia dubia produced by the
culture unit (see Section 4, Quality Assurance).

13.6.16.11  Record Keeping

13.6.16.11.1  Records, kept in a bound notebook, include (1) source of
organisms used to start the cultures, (2) type of food and feeding times, (3)
dates culture were thinned and restarted, (4) rate of reproduction in
individual cultures,  (5) daily observations of the condition and behavior of
the organisms in the cultures, and  (6) dates and results of reference toxicant
tests performed  (see Section 4, Quality Assurance).

13.7  EFFLUENT AND RECEIVING WATER COLLECTION, PRESERVATION, AND STORAGE

13.7.1  See Section 8, Effluent and Receiving Water Sampling, Sample Handling,
and Sample Preparation for Toxicity Tests.

13.8  CALIBRATION AND STANDARDIZATION

13.8.1  See Section 4, Quality Assurance.

13.9  QUALITY CONTROL

13.9.1  See Section 4, Quality Assurance.

13.10  TEST PROCEDURES

13.10.1  TEST SOLUTIONS

13.10.1.1  Receiving  Waters

13.10.1.1.1  The  sampling  point is  determined by the  objectives  of the test.
Receiving water  toxicity  is determined with samples used.directly  as collected
or  after samples  are  passed through  a 60 urn NITEXฎ filter  and compared without
dilution, against a control.   For  a test consisting of  single receiving water
and control, approximately 600 ml  of sample would be  required for  each test,

                                      158

-------
assuming 10 replicates of 15 ml, and sufficient additional sample for chemical
analysis.

13.10.1.2  Effluents

13.10.1.2.1  The selection of the effluent test concentrations should be based
on the objectives of the study.  A dilution factor of 0.5 is commonly used.
A dilution factor of 0.5 provides precision of ฑ 100%, and testing of
concentrations between 6.25% and 100% effluent using only five effluent
concentrations (6.25%, 12.5%, 25%, 50%, and 100%).  Improvements in precision
decline rapidly if the dilution factor is increased beyond 0.5, and precision
declines rapidly if a smaller dilution factor is used.  Therefore, USEPA
recommends the use of the > 0-5 dilution factor.

13.10.1.2.2  If the effluent is known or suspected to be highly toxic, a lower
range of effluent concentrations should be used (such as 25%, 12.5%, 6.25%,
3.12%, and 1.56%).  If a high rate of mortality is observed during the first
1 to 2 h of the test, additional dilutions should be added at the lower range
of effluent concentrations.

13.10.1.2.3  The volume of effluent required for daily renewal of 10
replicates per concentration, each containing 15 ml of test solution, with a
dilution series of 0.5, is approximately 1 L/day.  A volume of 15 ml of test
solution is adequate for the organisms, and will provide a depth in which it
is possible to count the animals under a stereomicroscope with a minimum of
re-focusing.  Ten test chambers are used for each effluent dilution and for
the control.  Sufficient test solution (approximately 550 ml) is prepared at
each effluent concentration to provide 400 ml additional  volume for chemical
analyses at the high, medium, and low test concentrations.

13.10.1.2.4  Tests should begin as soon as possible,  preferably within 24 h of
sample collection.  The maximum holding time following retrieval  of the sample
from the sampling device should not exceed 36 h for off-site toxicity tests
unless permission is granted by the permitting authority.  In no case should
the sample be used in a test more than 72 h after sample collection (see
Section 8,  Effluent and Receiving Water Sampling, Sample Handling,  and Sample
Preparation for Toxicity Tests).

13.10.1.2.5  Just prior to test initiation (approximately one h)  the
temperature of sufficient quantity of the sample to make the test solutions
should be adjusted to the test temperature and maintained at that temperature
during the preparation of the test solutions.

13.10.1.2.6  The DO of the test solutions should be checked prior to test
initiation.  If any of the solutions are supersaturated with oxygen or any
solution has a DO concentration below 4.0 mg/L,  all  the solutions and the
control  must be gently aerated.
                                      159

-------
13.10.1.3  Dilution Water

13.10.1.3.1  Dilution water may be uncontaminated receiving water, a standard
synthetic (reconstituted) water, or some other uncontaminated natural water
(see Section 7, Dilution Water).

13.10.2  START OF THE TEST

13.10.2.1  Label the test chambers with a marking pen.  Use of color-coded
tape to identify each treatment and replicate is helpful.  A minimum of five
effluent concentrations and a control are used for each effluent test.  Each
treatment (including the control) must have ten replicates.

13.10.2.2  The test solutions can be randomly assigned to a board using a
template (Figure 1) or by using a table of random numbers (see Appendix A).
When using the randomized block design, test chambers are randomized only
once, at the beginning of the test.  A number of different templates should be
prepared, so that the same template is not used for every test.

13.10.2.3  Neonates less than 24 h old, and all within 8 h of the same age,
are required to begin the test.  The neonates are obtained from individual
cultures using brood boards, as described above (see Section 6,, Test
Organisms).  Neonates are taken only from adults that have eight or more young
in their third or subsequent broods.  These adults can be used as brood stock
until they are 14 days old.  If the neonates are held more than one or two
hours before using in the test, they should be fed (0.1 ml YCT and 0.1 ml
algal concentrate/15 ml of media).

13.10.2.4  Ten brood cups, each with 8 or more young, are randomly selected
from a brood board for use in setting up a test.  To start the test, one
neonate from the first brood cup is transferred to each of the six test
chambers in the first row on the test board (Figure 1).  One neonate from the
second brood cup is transferred to each of the six test chambers in the second
row on the test board.  This process is continued until each of the 60 test
chambers contains one neonate.                    .                   ,

13.10.2.4.1  The cups and test chambers may be placed on a light table to
facilitate counting the neonates.  However, care must be taken to avoid
temperature increase due to heat from the light table.

13.10.2.5  This blocking procedure allows the performance of each female to be
tracked.  If a female produces one weak offspring or male, the likelihood of
producing all weak offspring or all males is greater.  By using this known
parentage technique, poor performance of young from a given female can be
omitted from all concentrations.

13.10.3  LIGHT, PHOTOPERIOD, AND TEMPERATURE

13.10.3.1  The light quality and intensity should be at ambient laboratory
levels, approximately 10-20 nE/m/s, or 50 to 100 ft-c, with a photoperiod of
16 h of light and 8 h of darkness.
                                      160

-------
13.10.3.2  It is critical that the test water temperature be maintained at
25 ฑ 1ฐC to obtain three broods in seven days.

13.10.4  DISSOLVED OXYGEN (DO) CONCENTRATION

13.10.4.1  Aeration may affect the toxicity of effluents and should be used
only as a last resort to maintain satisfactory DO concentrations.  The DO
concentrations should be measured in the new solutions at the start of the
test (Day 0) and before daily renewal of the test solutions on subsequent
days.  The DO concentration should not fall below 4.0 mg/L (see Section 8,
Effluent and Receiving Water Sampling, Sample Handling, and Sample Preparation
for Toxicity Tests).  Aeration is generally not practical during the daphnid,
Cen'odaphm'a dubia, test.  If the DO in the effluent and/or dilution water is
low, aerate gently before preparing the test solutions.  The aeration rate
should not exceed 100 bubbles/min using a pipet with an orifice of
approximately 1.5 mm, such as a 1 ml KIMAXฎ serological pipet, No. 37033, or
equivalent.  Care should be taken to ensure that turbulence resulting from
aeration does not cause undue physical stress to the organisms.

13.10.5  FEEDING

13.10.5.1  The organisms are fed when the test is initiated, and daily
thereafter.  Food is added to the fresh medium immediately before or
immediately after the adults are transferred.  Each feeding consists of 0.1 ml
YCT and 0.1 mL Selenastrum capn'cornutum concentrate/15 ml test solution
(0.1 ml of algal concentrate containing 3.0-3.5 X 10  cells/ml will  provide
2-2.3 X 10  cells/ml in the test chamber).

13.10.5.2  The YCT and algal suspension can be added accurately to the test
chambers by using automatic pipettors, such as Gil son, Eppendorf, Oxford, or
equivalent.

13.10.6  OBSERVATIONS DURING THE TEST

13.10.6.1  Routine Chemical and Physical Determinations

13.10.6.1.1  DO is measured at the beginning and end of each 24-h exposure
period in at least one test chamber at each test concentration and in the
control.       •

13.10.6.1.2  Temperature and pH are measured at the end of each 24-h exposure
period in at least one test chamber at each test concentration and in the
control.  Temperature should be monitored continuously or observed and
recorded daily for at least two locations in the environmental control system
or the samples.  Temperature should be measured in sufficient number of test
vessels at least at the end of the test to determine the temperature variation
in the environmental chamber.

13.10.6.1.3  The pH is measured in the effluent sample each day before new
test solutions are made.
                                      161

-------
13.10.6.1.4  Conductivity, alkalinity and hardness are measured in each new
sample (100% effluent or receiving water) and in the control.

13.10.6.1.5  Record the data on data sheet (Figure 2).

13.10.6.2  Routine Biological Observations

13.10.6.2.1  Three or four broods are usually obtained in the controls in a
7-day test conducted at 25 ฑ 1ฐC.  A brood is a group of offspring released
from the female over a short period of time when the carapace is discarded
during molting.  In the controls, the first brood of two-to-five young is
usually released on the third or fourth day of the test.  Successive broods
are released every 30 to 36 h thereafter.  The second and third broods usually
consist of eight to 20 young each.  The total number of young produced by a
healthy control organism in three broods often exceeds 30 per female.

13.10.6.2.2  The release of a brood may be inadvertently interrupted during
the daily transfer of organisms to fresh test solutions, resulting in a split
in the brood count between two successive days.  For example, four neonates of
a brood of five might be released on Day 3, just prior to test solution
renewal, and the fifth released just after renewal, and counted on Day 4.
Partial broods, released over a two-day period, should be counted as one
brood.

13.10.6.2.3  Each day, the live adults are transferred to fresh test
solutions, and the numbers of live young are recorded (see data form,
Figure 3). The young can be counted with the aid of a stereomicroscope with
substage lighting.  Place the test chambers on a light box over a strip of
black tape to aid in counting the neonates.  The young are discarded after
counting.

13.10.6.2.4  Some of the effects caused by toxic substances  include, (1) a
reduction in the number of young produced, (2) young may develop in the brood
pouch of the adults, but may not be released during the exposure period, and
(3) partially or fully developed young may be released, but  are all dead at
the end of the 24-h period.  Such effects should be noted on the data sheets
(Figure 3).

13.10.6.2.5  Protect the daphnids, Ceriodaphm'a dubia, from  unnecessary
disturbance during the test by carrying out the daily test observations,
solution renewals, and transfer of females carefully.  Make  sure the females
remain immersed during the performance of these operations.

13.10.7  DAILY PREPARATION OF TEST CHAMBERS

13.10.7.1  The test is started (Day 0) with new disposable polystyrene cups or
precleaned 30-mL borosilicate glass beakers that are  labeled and color-coded
with tape.  Each following day, a new set of plastic  cups or precleaned glass
beakers is prepared, labeled,  and color-coded with tape similar to the
original set.  New solutions are placed  in the new set of test chambers, and
the test organisms are transferred from the original  test chambers to the new
ones with corresponding labels and color-codes.  Each day, previously used

                                      162

-------
Discharger: _
Location: 	
Template No.:
Analyst:
  Dates:
   Food:
                                    Day
Control :
Temp.
D.O. Initial
Final
pH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

0










1










2










3










4










5










6










7










Remarks










                                    Day
Cone:
Temp.
D.O. Initial
Final
pH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

0










1










2










3










4










5










6










7










Remarks










                                    Day
Cone:
Temp.
D.O. Initial
Final
pH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

0










1










2










3










4










5










6










7










Remarks










Figure 2.   Data form for the daphnid, Ceriodaphnia dubia,  survival  and
            reproduction test.  Routine chemical and physical
            determinations.
                                    163

-------
Discharger: _
Location: 	
Template No.:
Analyst:
Dates: _
Food:
                                    Day
Cone:
Temp.
D.O. Initial
Final
pH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

0










1










2










3










4










5










6










7










Remarks










                                    Day
Cone:
Temp.
D.O. Initial
Final
pH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

0










1










2










3










4










5










6










7










Remarks










                                    Day
Cone:
Temp.
D.O. Initial
Final
oH Initial
Final
Alkalinity
Hardness
Conductivity
Chlorine

0










1










2










3










4










5










6










7










Remarks










Figure 2.   Data  form  for the  daphnid,  Ceriodaphm'a dubia,  survival  and
            reproduction test.  Routine chemical and physical determinations
            (CONTINUED).
                                    164

-------
glass beakers are recleaned (see Section 5, Facilities, Equipment, and
Supplies) for the following day, and previously used plastic cups are
discarded.

13.10.8  TEST SOLUTION RENEWAL

13.10.8.1  Freshly prepared solutions are used to renew the test daily.  For
on-site toxicity studies, fresh effluent or receiving water samples should be
collected daily, and no more than 24 h should elapse between collection of the
samples and their use in the tests (see Section 8, Effluent and Receiving
Water Sampling, Sample Handling, and Sample Preparation for Toxicity Tests).
For off-site tests, a minimum of three samples are collected, preferably on
days one, three, and five.  No more than 36. h should elapse between collection
of the sample and the first use in the test.  Maintain the samples in the
refrigerator at 4ฐC until used.                 '

13.10.8.2  New test solutions are prepared daily, and the test organisms are
transferred to the freshly prepared solutions using a small-bore (2 mm) glass
or polyethylene dropper or pipet.  The animals are released under the surface
of the water so that air is not trapped under the carapace.  Organisms that
are dropped or injured are discarded.

13.10.9  TERMINATION OF THE TEST

13.10.9.1  Tests should be terminated when 60% of the control organisms have
produced their third brood, or at the end of 8 days, whichever occurs first.
Because of the rapid rate of development of Cen'odaphnia dubia, at test
termination all observations on organism survival and numbers of offspring
should be completed within two hours.  An extension of more than a few hours
in the test period would be a significant part of the brood production cycle
of the animals, and could result in additional broods.

13.10.9.2  Count the young, conduct required chemical measurements, and
complete the data sheets (Figure 3).

13.10.9.3  Any animal not producing young should be examined to determine if
it is a male (Berner, 1986).  In most cases, the animal will need to be placed
on a microscope slide before examining (see Subsection 13.6.16.4).

13.11  SUMMARY OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA

13.11.1  A summary of test conditions and test acceptability criteria is
presented in Table 3.

13.12  ACCEPTABILITY OF TEST RESULTS

13.12.1  For the test results to be acceptable, at least 80% of the control
organisms must survive, and 60% of surviving adults in the controls must have
had at least three broods, with an average total number of 15 or more
offspring per surviving adult.
                                      165

-------
Discharger:
Location:
Date Sample Collected:
Analyst:	
Test Start-Date/Time:
Test Stop -Date/Time:
Re
1
Cone.








Dav
1
2
3
4
5
6
7

Total
2










3










4










5










lica
6










te
7










8










9










10










Number
of
Younq










Number
of
Adults










Young
per
Adult










Replicate

Cone.








1
Dav
1
2
3
4
5
6
7

Total
2










3










4










5










6










7










8










9










10










Number
of
Younq










Number
of
Adults










Young
per
Adult










Re

Cone.








1
Dav
1
2
3
4
5
6
7

Total
2










3










4










plic
5










ate
6










7










8










9










10










Number
of
Younq










Number
of
Adults










Young
per
Adult










Figure 3.   Data form for the daphnid, Cen'odaphm'a dubia,  survival  and
            reproduction test.  Daily summary of data.
                                    166

-------
Discharger:
Location:
Date Sample Collected:
Analyst:	
Test Start-Date/Time:
Test Stop -Date/Time:
Re
1
Cone.









Dav
1
2
3
4
5
6
7

Total
?










3










4










R










ica
6










te
7










R










9










10










Number
of
Younq










Number
of
Adults










Young
per
Adult










Replicate

Cone.









1
Dav
1
2
3
4
5
6
7

Total
?










3










4










5










6










7










8










9










10










Number
of
Younq










Number
of
Adults










Young
per
Adult










Re

Gone.









1
Dav
1
2
3
4
5
6
7

Total
?










3










4










plic
5










ate
6










7










8










9










-10










Number
of
Younq










Number
of
Adults










Young
per
Adult










 Figure 3.    Data form for the daphnid,  Ceriodaphm'a  dubia,  survival  and
             reproduction test.   Daily summary of data (CONTINUED).
                                    167

-------
TABLE 3.  SUMMARY OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA FOR
          DAPHNID, CERIODAPHNIA DUBIA, SURVIVAL AND REPRODUCTION TOXICITY
          TESTS WITH EFFLUENTS AND RECEIVING WATERS
   1. Test type:
   2. Temperature (ฐC):
   3. Light quality:
   4. Light intensity:

   5. Photoperiod:
   6. Test chamber size:
   7. Test solution volume:
   8. Renewal  of test solutions:
   9. Age of test organisms:
  10.  No.  neonates per
       test chamber:
  11.  No.  replicate test
       chambers per concentration:
  12.  No.  neonates per
       test concentration:
  13.  Feeding regime:
  14.  Cleaning:

  15.  Aeration:
  16.  Dilution water:
Static renewal
25 ฑ  1ฐC
Ambient laboratory  illumination
10-20 /^E/m2/s, or 50-100 ft-c
(ambient laboratory levels)
16 h  light, 8 h dark
30 mL (minimum)
15 mL (minimum)
Daily
Less  than  24  h;   and  all  released
within a 8-h period
10
10
Feed  0.1  mL  each of  YCT  and  algal
suspension per test chamber daily
Use freshly cleaned  glass beakers or
new plastic cups daily
None
Uncontaminated source of receiving or
other natural  water,  synthetic  water
prepared using MILLIPORE  MILLI-Qฎ or
equivalent deionized  water and reagent
grade chemicals or DMW   (see Section
7, Dilution Water)
                                   168

-------
TABLE 3.  SUMMARY OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA
          FOR DAPHNID, CERIODAPHNIA DUBIA, SURVIVAL AND REPRODUCTION
          TOXICITY TESTS WITH EFFLUENTS AND RECEIVING WATERS (CONTINUED)
17. Test concentrations:



18. Dilution factor:


19. Test duration:



20. Endpoints:

21. Test acceptability criteria:
22. Sampling requirements:
Effluents: Minimum of 5 and a control
Receiving Water:  100% receiving water
or minimum of 5 and a control
Effluents:  > 0.5
Receiving Waters:
None or > 0.5
22. Sample volume required:
Until   60%   of  surviving   control
organisms have  three  broods (maximum
test duration 8 days)

Survival and reproduction

80% or greater survival  and an average
of  15  or  more  young  per  surviving
female  in the  control  solutions. 60%
of  surviving  control organisms  must
produce three broods.

For  on-site tests,  samples collected
daily,  and  used within  24 h  of the
time   they   are  removed   from   the
sampling device.  For off-site tests,
a minimum of  three  samples collected
on  days  one,  three, and  five  with a
maximum  holding time of  36  h  before
first use (see Section 8,  Effluent and
Receiving   Water   Sampling,   Sample
Handling, and  Sample  Preparation for
Toxicity Tests, Subsection 8.5.4)

1 L/day           i
                                  169

-------
13.13  DATA ANALYSIS  '

13.13.1  GENERAL

13.13.1.1  Tabulate and summarize the data.  A sample set of survival and
reproduction data is listed in Table 4.

  TABLE 4.    SUMMARY  OF  SURVIVAL  AND REPRODUCTION  DATA FOR  THE DAPHNID,
              CERIODAPHNIA DUBIA, EXPOSED TO AN EFFLUENT FOR SEVEN DAYS
  Effluent
  Concentration
      (X)
 No.  of Young  per Adult
	Replicate	
                                    10
No.
Live
Adults
Control
1.56
3.12
6.25
12.5
25.0
27
32
39
27
10
0
30
35
30
34
13
0
29
32
33
36
7
0
31
26
33
34
7
0
16
18
36
31
7
0
15
29
33
27
10
0
18
27
33
33
10
0
17
16
27
31
16
0
14
35
38
33
12
0
27
13
44
31
2
0
10
10
10
10
10
3
13.13.1.2  The endpoints of toxicity tests using the daphnid, Cen'odaphm'a
dubia, are based on the adverse effects on survival and reproduction.  The
LC50, the IC25, the IC50 and the EC50 are calculated using point estimation
techniques, and LOEC and NOEC values for survival and reproduction are
obtained using a hypothesis test approach such as Fisher's Exact Test (Finney,
1948; Pearson and Hartley, 1962), Dunnett's Procedure (Dunnett, 1955) or
Steel's Many-one Rank Test (Steel, 1959; Miller, 1981) (see Section 9, Chronic
Toxicity Test Endpoints and Data Analysis).  Separate analyses are performed
for the estimation of the LOEC and NOEC endpoints and for the estimation of
the LC50, IC25, IC50 and EC50.  Concentrations at which there is no survival
in any of the test chambers are excluded from the statistical analysis of the
NOEC and LOEC for reproduction, but included in the estimation of the LC50,
IC25, IC50, and EC50.   See the Appendices for examples of the manual
computations, program listings, and examples of data input and program output.

13.13.1.3  The statistical tests described here must be used with a knowledge
of the assumptions upon which the tests are contingent.  Tests for normality
and homogeneity of variance are included in Appendix B.  The assistance of a
statistician is recommended for analysts who are not proficient in statistics.
                                      170

-------
13.13.2  EXAMPLE OF ANALYSIS OF THE DAPHNID, CERIODAPHNIA DUBIA, SURVIVAL DATA

13.13.2.1  Formal statistical analysis of the survival data is outlined on the
flowchart in Figure 4.  The response used in the analysis,is the number of
animals surviving at each test concentration.  Separate analyses are performed
for the estimation of the NOEC and LOEC endpoints and for the estimation of
the EC50, LC50, IC25, or IC50 endpoints.  Concentrations at which there is no
survival in any of the test chambers are excluded from the statistical
analysis of the NOEC and LOEC, but included in the estimation of the LC, EC,
and 1C endpoints.

13.13.2.2  Fisher's Exact Test is used to determine the NOEC and LOEC
endpoints.  It provides a conservative test of the equality of any two
survival proportions assuming only the independence of responses from a
Bernoulli (binomial) population.  Additional information on Fisher's Exact
Test is provided in Appendix G.

13.13.2.3  Probit Analysis (Finney, 1971; Appendix I) is used to estimate the
concentration that causes a specified percent decrease in survival from the
control.  In this analysis, the total number dead at a given concentration is
the response.
                                                          j

13.13.2.4  Example of Analysis of Survival Data

13.13.2.4.1  The data in Table 4 will be used to illustrate the analysis of
survival data from the daphnid, Cen'odaphm'a dubia, Survival and Reproduction
Test.  As can be seen from the data in Table 4, there were no deaths in the
1.56%, 3.12%, 6.25%, and 12.5% concentrations.  These concentrations are
obviously not different from the control in terms of survival.  This leaves
only the 25% effluent concentration to be tested statistically for a
difference in survival from the control.

13.13.2.5  Fisher's Exact Test

13.13.2.5.1  The basis for Fisher's Exact Test is a 2x2 contingency table.
From the 2x2 table prepared by comparing the control  and the effluent
concentration, determine statistical  significance by look-ing up a value in  the
table provided in Appendix G (Table G.5).  However, to use this table the
contingency table must be arranged in the format illustrated in Table 5.

13.13.2.5.2  Arrange the table so that the total number of observations for
row one is greater than or equal to the total  for row two (A > B).  Categorize
a success such that the proportion of successes for row one is greater than or
equal to the proportion of successes  for row two (a/A > b/B).,   For these data,
a success may be 'alive' or 'dead' whichever causes a/A > b/B.  The test is
then conducted by looking up a value  in the table of significance levels of b
and comparing it to the b value given in the contingency table.   The table  of
significance levels of b is included  in Appendix G, Table G.5.  Enter
Table G.5 in the section for A, subsection for B, and the line for a.  If the
b value of the contingency table is equal to or less  than the  integer in the
column headed 0.05 in Table G.5, then the survival  proportion  for the effluent
concentration is significantly different from that of the control.  A dash  or

                                     171

-------
               STATISTICAL ANALYSIS OF CERIODAPHNIA
                 SURVIVAL AND REPRODUCTION TEST

                               SURVIVAL
      MORTALITY DATA
         #DEAD
       TWO OR MORE
    PARTIAL MORTALITIES?
       FISHER'S EXACT
           TEST
NO
IS PROBIT MODEL
APPROPRIATE?
(SIGNIFICANT X2 TEST)
^
YES
r
NO

ENDPOINT ESTIMATES
   NOEC, LOEC
ONE OR MORE
PARTIAL MORTALITIES?
i
YES
r
NO

                                                    GRAPHICAL METHOD
                                                          LC50
PROBIT METHOD L
M

H
	 fe
                            ZERO MORTALITY IN THE
                            LOWEST EFFLUENT CONG.
                           AND 100% MORTALITY IN THE
                           HIGHEST EFFLUENT CONC.?
                           NO
                                     YES
                              SPEARMAN-KARBER
                                  METHOD
                                    I
                                LC50AND95%
                                CONFIDENCE
                                 INTERVAL
ซ
TRIMMED SPEARMAN
KARBER METHOD


Figure 4.   Flowchart for statistical  analysis of the daphnid,  Ceriodaphm'a
           dubia, survival data.

                                  172

-------
                TABLE 5.  FORMAT OF THE 2x2 CONTINGENCY TABLE

Condition 1
Condition 2
Total
i
Number of
Successes Failures
a A - a
b B - b
a + b [(A+B) - a - b]
Number of
Observations
A
B
A + B
absence of entry in Table G.5 indicates that no contingency table in that
class is significant.

13.13.2.5.3  To compare the control and the effluent concentration of 25%,  the
appropriate contingency table for the test is given in Table 6.

13.13.2.5.4  Since 10/10 > 3/10, the category 'alive' is regarded as a
success.  For A = 10, B = 10 and, a = 10, under the column headed 0.05,  the
value from Table G.5 is b = 6.  Since the value of b (b = 3) from the
contingency table (Table 6), is less than the value of b (b = 6) from
Table G.5 in Appendix G, the test concludes that the proportion  surviving in
the 25% effluent concentration is significantly different from the control.
Thus the NOEC for survival is 12.5% and the LOEC is 25%.

            TABLE 6.  2x2 CONTINGENCY TABLE FOR CONTROL AND 25% EFFLUENT


                                 Number of               Number of
                                                         Observations
           	. .  .	Alive	Dead	

             Control           10             0              10

             25% Effluent       3             7              10


             Total             13             7              20
                                      173

-------
13.13.2.6  Calculation of the  LC50

13.13.2.6.1  The data used for the Trimmed Spearman-Karber Method are
summarized in Table 7.  To perform the Trimmed Spearman-Karber Method, run the
USEPA Trimmed Spearman-Karber  Program.  An example of the program input and
output  is supplied in Appendix J.
            TABLE 7.  DATA FOR TRIMMED SPEARMAN-KARBER ANALYSIS
                                       Effluent Concentration (%)
                         Control   1.56   3.12   6.25   12.5   25.0
Number Dead
Number Exposed
0
10
0
10
0
10
0
10
0
10
8
10
13.13.2.6.2  For this-example, with only one partial mortality, Trimmed
Spearman-Karber analysis appears appropriate for this data.

13.13.2.6.3  Figure 5 shows the output for the Trimmed Spearman-Karber
Analysis of the data in Table 7 using the USEPA Program.

13.13.3  EXAMPLE OF ANALYSIS OF THE DAPHNID, CERIODAPHNIA DUBIA, REPRODUCTION
         DATA

13.13.3.1  Formal statistical analysis of the reproduction data is outlined on
the flowchart in Figure 6.  The response used in the statistical analysis is
the number of young produced per adult female, which is determined by taking
the total number of young produced until either the time of death of the adult
or the end of the experiment, whichever comes first.  An animal that dies
before producing young, if it has not been identified as a male, would be
included in the analysis with zero entered as the number of young produced.
The subsequent calculation of the mean number of live young produced per adult
female for each toxicant concentration provides a combined measure of the
toxicant's effect on both mortality and reproduction.  An 1C estimate can be
calculated for the reproduction data using a point estimation technique (see
Section 9, Chronic Toxicity Test Endpoints and Data Analysis).  Hypothesis
testing can be used to obtain an NOEC for reproduction.  Concentrations above
the NOEC for survival are excluded from the hypothesis test for reproduction
effects.

13.13.3.2  The statistical analysis using hypothesis tests consists of a
parametric test, Dunnett's Procedure, and a nonparametric test, Steel's
Many-one Rank Test.  The underlying assumptions of the Dunnett's Procedure,
normality and homogeneity of variance, are formally tested using the Shapiro

                                      174

-------

TRIMMED SPEARMAN-KARBER METHOD.  VERSION 1.5
DATE:   1                  TEST NUMBER: 2
TOXICANT:  effluent
SPECIES:   ceriodaphnia dubia

RAW DATA:  Concentration
                                                      DURATION:
                                  7 Days
                .00
               1.25
               3.12
               6.25
              12.5
              25.0

  SPEARMAN-KARBER TRIM:

  SPEARMAN-KARBER ESTIMATES:
Number
Exposed
10
10
10
10
10
10
Mortal!
0
0
0
0
0
8
   20.41 %

LC50:          77.28
        95% CONFIDENCE LIMITS
        ARE NOT RELIABLE,:
 NOTE:   MORTALITY PROPORTIONS WERE NOT MONOTONICALLY INCREASING.
        ADJUSTMENTS WERE MADE PRIOR TO SPEARMAN-KARBER ESTIMATION.
         Figure 5.   Output for USEPA Trimmed  Spearman-Karber program.
                                     175

-------
                 STATISTICAL ANALYSIS OF CERIODAPHNIA
                    SURVIVAL AND REPRODUCTION TEST
                             REPRODUCTION DATA
                           NO. OF YOUNG PRODUCED
POINT ESTIMATION
1
F
                             HYPOTHESIS TESTING
                         (EXCLUDING CONCENTRATIONS
                          ABOVE NOEC FOR SURVIVAL)
   ENDPOINT ESTIMATE
       IC25, IC50
    SHAPIRO-WILK'S TEST
                        NON-NORMAL DISTRIBUTION
                   NORMAL DISTRIBUTION
        HOMOGENEOUS
          VARIANCE
                               BARTLETTSTEST
                              HETEROGENEOUS
                                 VARIANCE
              EQUAL NUMBER OF
                REPLICATES?
           NO
YES
     T-TESTWITH
     BONFERRONI
     ADJUSTMENT
                      EQUAL NUMBER OF
                        REPLICATES?
                                        YES
                                     NO
        STEEL'S MANY-ONE
           RANK TEST
                                    T
  WILCOXON RANK SUM
      TEST WITH
BONFERRONI ADJUSTMENT
                             ENDPOINT ESTIMATES
                                 NOEC, LOEC
Figure 6.    Flowchart for the statistical  analysis of the daphnid, Ceriodaphnia
            dubia, reproduction data.

                                    176

-------
               t              -?d  Bartlett/s Test fฐr homogeneity of variance.
               these tests  fails,  a nonparametric test, Steel's Many-one Rank
  rnH.       to de+um1n! the NOEC and LOEC-   If th* assumptions of Dunnett's
  Procedure are met, the endpoints are determined by the parametric test

 •13.13.3.3  Additionally, if unequal numbers of replicates occur amona the
  anSlvses  ^Kp^6"615 ?6?ted ^ are ^metric and nonparametr?c alterative
  analyses.  The parametric analysis is a t test with the Bonferroni adjustment
  (see Appendix D)   The Wilcoxon Rank Sum Test with the Boinferroni adjustment
  is the nonparametric alternative (see Appendix F) .                aajusiment

  13.13.3.4  The data, mean,  and variance of the observations at each
  concentration including the control are listed in Table 8.  A plot of the
  number of young per adult female for each concentration is pro! ded in
  Figure 7.  Since there is significant mortality in the 25% effluent
  concentration,  its effect on reproduction is not considered

        TABLE 8.   THE DAPHNID,  CERIODAPHNIA DUBIA, .REPRODUCTION DATA
   Replicate    Control
                                      Effluent Concentratinn  (%)
1.56
3.12
                                                      6.25
                              12.5
1
2
3
4

6
7
8
9
10
Mean(Y,.)
?.i
i
27
30
29
31
16
15
18
17
14
27
22.4
48.0
1
32
35
32
26
18
29
27
16
35
13
26.3
64.0
2
39
30
33
33
36
33
33
27
38
44
34.6
23.4
3
27
34
36
34
31
27
33
31
33
31
31.7
8.7
4
10
13
7
7
7
10
10
16
12
2
9.4
15.1
5
13.13.3.5  Test for Normality

13.13.3 5.1  The first step of the test for normality is to center the
observations by subtracting the mean of all  the observatio s witMn a
concentration from each observation in that concentration   T e  centered
observations are summarized in Table 9                          centered
                                     177

-------
*
                                                                                s-
                                                                                =3
                                                                                CO

                                                                                to

                                                                                I
                                                                                ns
                                                                                c
                                                                                •c
                                                                                Q.

                                                                                O
                                                                                C
Z
O
Z
UJ
                                                                           8
                                                                            u.
                                                                            LLI
                                                                                 •a
                                                                                 ns
                                                                                 O)
                                                                                 d)
      rC
      s-
      o>
      o.
      en
                                                                                  o  c
                                                                                     o
                                                                                  O) +->
                                                                                 JO  U
                                                                                  i-g
                                                                                  c  o
                                                                                 <4-  O.
                                                                                  O  O)
                                                                                     ro
                                                                    -O
                                                                                  O)
                                                                                  3
                               dO
                              178

-------
        TABLE 9.   CENTERED OBSERVATIONS  FOR  SHAPIRO-WILK'S  EXAMPLE
Effluent Concentration (%)
Repl i cate
1
2
3
4
5
6
7
8
9
10
Control
4.6
7.6
6.6
8.6
-6.4
-7.4
-4.4
-5.4
-8.4
4.6
1.56
5.7
8.7
5.7
-0.3
-8.3
2.7
0.7
-10.3
8.7
-13.3
3.12
4.4
-4.6
-1.6
-1.6
1.4
-1.6
-1.6
-7.6
3.4
9.4
6.25
-4.7
2.3
4.3
2.3
-0.7
-4.7
1.3
-0.7
1.3
-0.7
12.5
0.6
3.6
-2.4
-2.4
-2.4
0.6
0.6
6.6
2.6
-7.4
13.13.3.5.2  Calculate the denominator,  D,  of  the test  statistic:



                               D =  ฃ (Xฑ-X)2
                                   i=l



   Where:  Xf  =  the  ith centered observation



          X  =  the  overall mean of the centered  observations


          n  =  the  total  number of centered observations.


For this set  of data,

                           n = 50


                           X = -^ (0.0) = 0.0
                                50


                           D = 1433.4
                                  • •  •        '            i



13.13.3.5.3  Order  the centered observations from smallest to largest
    Where X(O is the ith ordered observation.   These  ordered observations are
listed in Table 10.
                                     179

-------
   TABLE 10.  ORDERED CENTERED OBSERVATIONS FOR SHAPIRO-MILK'S EXAMPLE.
                             ci>
(D
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
-13.3
-10.3
-8.4
-8.3
-7.6
-7.4
-7.4
-6.4
-5.4
-4.7
-4.7
-4.6
-4.4
-2.4
-2.4
-2.4
-1.6
-1.6
-1.6
-1.6
-0.7
-0.7
-0.7
-0.3
0.6
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
0.6
0.6
0.7
1.3
1.3
1.4
2.3
2.3
2.6
2.7
3.4
3.6
4.3
4.4
4.6
4.6
5.7
5.7
6.6
6.6
7.6
8.6
8.7
8.7
9.4
13.13.3.5.4  From Table 4, Appendix B, for the number of observations,  n,
obtain the coefficients a15 a2,  ...,  ak where  k is  n/2 if n  is  even  and
(n-l)/2 if n is odd.  For the data in this example, n = 50,  k  =  25.  The  a,-
values are listed in Table 11.

13.13.3.5.5  Compute the test statistic, W, as follows:

                         W= -i [ฃaAX(n-*+V-X(1))]2
                             D  i-i *

The differences xcn"i"'1> - X
-------
    TABLE 11.  COEFFICIENTS AND DIFFERENCES FOR SHAPIRO-MILK'S EXAMPLE
i
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
ai
0.3751
0.2574
0.2260
0.2032
0.1847
0.1691
0.1554
0.1430
0.1317
0.1212
0.1113
0.1020
0.0932
0.0846
0.0764
0.0685
0.0608
0.0532
0.0459
0.0386
0.0314
0.0244
0.0174
0.0104
0.0035
X<,M, . „ซ„•
22.7
19.0
17.1
16.9
15.2
14.0
14.0
12.1
11.1
9.3
9.3
9.0
8.7
6.0
5.8
5.1
4.2
3.9
3.9
3.0
2.0
2.0
1.4
0.9
0.0

X(50)
X(49)
x(48)
XC47>
Y<46>
A
x<45)
X(44)
X(43)
x<42)
X(41)
x<40)
X(39,
X(38)
X(37)
X(36)
x<35>
X(34)
Y<33)
A
X(32)
X(31)
X(30)
Y(29)
A
X(28)
X(27>
Y<26)
A

X
- XC6)
- x(7>
- x<8)
- XC9)
x(10)
- x<11>
- x<12)
- x(13)
- x<14>
- x(15)
- x(16)
- XC17)
- x(18>
- x<19)
YC20)
*" A
- x<21)
- XC22)
- x(23>
- x<24)
y(25)
"" A
13.13.3.5.6  The decision rule for this test is to compare W with the critical
value found in Table 6, Appendix B.  If the computed W is less than the
critical value, conclude that the data are not normally distributed.  For this
example, the critical value at a significance level of 0.01 and 50
observations (n) is 0.930.  Since W = 0.97 is greater than the critical value,
the conclusion of the test is that the data are normally distributed.

13.13.3.6  Test for Homogeneity of Variance

13.13.3.6.1  The test used to examine whether the variation in number of young
produced is the same across all effluent concentrations including the control,
is Bartlett's Test (Snedecor and Cochran, 1980).  The test statistic is as
follows:
                                      181

-------
                                 riins;]
          B--JS^	_J-	

   Where:  V,  =   degrees of freedom for each effluent  concentration  and
                  control,  V,- =  (n,- - 1)

          p  =   number of levels of effluent concentration  and  control

          nf  -   the number of  replicates for concentration  i

          In  -   loge

          i  -   1,  2, ...ป P where p is the number of concentrations
                 including the  control
          -32 _
13.13.3.6.2  For the data in this example (see Table 8),  all  effluent
concentrations including the control  have the same number of  replicates
(nf = 10  for all  i).   Thus,  V,- = 9 for all i.

13.13.3.6.3  Bartlett's statistic is  therefore:

          B- [(45)ln(31.8) -9 f) ln(.sf) ]/1. 04

             =  [45(3.5) - 9(16.!)]/!.04

             -  12.6/1.04

             -  12.1

13.13.3.6.4  B is approximately distributed  as chi-square with p -  1 degrees
of freedom, when the variances are in fact the same.  Therefore, the
appropriate critical value for this  test, at a significance level  of 0.01  with
four degrees of freedom, is 13.3.  Since B = 12.1 is less than the  critical
value of 13.3, conclude that the variances are not different.

13.13.3.7  Dunnett's Procedure

13.13.3.7.1  To obtain an estimate of the pooled variance for the Dunnett's
Procedure, construct an ANOVA table  as described in Table 12.
                                      182

-------
                          TABLE 12.  ANOVA TABLE
      Source
 df
Sum of Squares
     (SS)
Mean Square(MS)
    (SS/df)
      Between
      Within
P - 1
N - p
     SSB
     SSW
SB = SSB/(p-l)
Sy = SSW/(N-p)
      Total
N - 1
     SST
Where:     p  =  number effluent concentrations including
           N  =  total  number of observations n, +  n2 ... -
          n,-  =  number of observations in concentration i

         SSB = T,Tl/ni-Gz/N   Between Sum of Squares

         SST = f, ^YlJ-G2/N   Total  Sum of Squares
                                       the control
                                       •h n..
                1=1.7=1
         SSW = SST-SSB
              Within Sum of Squares
           G  =  the grand total  of all  sample observations,  G= ฃ r,
                                                          ;       i-i
          Tf  •=.  the total of the replicate measurements  for  concentration  i
          ,-j   =  the jth observatioh for concentration  i  (represents  the
                 number of young  produced by female  j  in  effluent
                 concentration i)
                                     183

-------
13.13.3.7.2  For the data in  this  example:
                     n, =  n2 = n3 = nA = n5 = 10
N
T,
T2
T'
T5
G
= 50
= Y -
-Jป-
" Y31 '
~ Y41 '
= Y51-
= T, -t

*• Y12 + '
f Y22 + .
f Y32 + .
f Y42 + .
f Y52 + .
- T, + T,

. . +
. . +,
. . +
. . +
. . +
+ T. +

YHO
Y
Y210
Y310
'410
YSIO
T, =

= 224
= 263
= 346
= 317
= 94
1244
                  SSB = &Tl/nฑ-G2/N
                   t  =
                    2
                              (22.4-26.3)
                         [5.64^(1/10)+(1/10)]
                         P ni
                  SST — if 2-i Yi-<—G
                      =  36,272-         =5321.28
                  SSW = SST-SSB   = 5321.28 - 3887.88 = 1433.40
                   SB  = SSB/(p-l)  = 3887.88/(5-l) = 971.97
                   Sj  = SSW/(N-p)  = 1433.40/(50-5) = 31.85
13.13.3.7.3  Summarize these calculations in an ANOVA table  (Table  13).
13.13.3.7.4  To perform the individual  comparisons, calculate  the t statistic
for each concentration and control  combination as follows:
Where: Y, ซ mean number of young produced  for  effluent concentration i
       Y, ซ= mean number of young produced  for  the  control
       Sw = square root of within mean square
       nt - number of replicates for the control
       n,- ป number of replicates for concentration i.
                                      184

-------
            TABLE 13.   ANOVA TABLE FOR DUNNETT'S PROCEDURE EXAMPLE
     Source            df           Sum of Squares        Mean Square(MS)
                                        (SS)                  (SS/df)
Between
Within
4
45
3887.88
1433.40
971.97
i 31.85
     Total             49              5321.28
Since we are looking for a decrease in reproduction from the control, the mean
for concentration i is subtracted from the control mean in the t statistic
above.  However, if we were looking for an increased response over the
control, the control mean would be subtracted from the mean at a
concentration.

13.13.3.7.5  Table 14 includes the calculated t values for each concentration
and control combination.  In this example, comparing the 1.56% concentration
with the control the calculation is as follows:

                    t2 = _ (22.4-26.3)        _ .C.
                          [5.64v/(l/10)



                        TABLE 14.  CALCULATED T VALUES
             Effluent Concentration (%)         i           t,-
1.56
3.12
6.25
12.5
2
3
4
5
-1.55
-4.84
-3.69
5.16
13.13.3.7.6  Since the purpose of this test is to detect a significant ,
reduction in mean reproduction, a one-sided test is appropriate.   The critical
value for this one-sided test is found in Table 5, Appendix C.   Since an entry
for 45 degrees of freedom for error is not provided in the table,  the entry
for 40 degrees of freedom for error,  an alpha level of 0.05 and four
concentrations (excluding the control) will be used,  2.23.  The mean
                                      185

-------
reproduction for concentration "i" is considered significantly less than the
mean reproduction for the control if t= is greater than the critical  value.
Since t5 is greater than 2.23, the 12.5% concentration has  significantly lower
reproduction than the control.  Hence the NOEC and the LOEC for reproduction
are 6.25% and 12.5%, respectively.

13.13.3.7.7  To quantify the sensitivity of the test, the minimum significant
difference (MSD) that can be statistically detected may be calculated:
                          MSD =  d


Where: d  - the critical value for the Dunnett's Procedure

       Sw = the square root of the within mean square

       n  - the common number of replicates at each concentration
            (this assumes equal replication at each concentration)

       n., - the number of replicates in the control.

13.13.3.7.8  In this example:
                     MSD = 2.23(5.64)^(1/10)+(1/10)

                          = 2.23 (5.64) (0.447)

                          = 5.62

13.13.3.7.9  Therefore, for this set of data, the minimum difference that can
be detected as statistically significant is 5.62.

13.13.3.7.10  This represents a 25% decrease in mean reproduction from the
control.

13.13.3.8  Calculation of the 1C

13.13.3.8.1  The reproduction data in Table 4 are utilized in this example.
As can  be seen from Figure 8, the observed means are not monotonically
non-increasing with respect to concentration.  Therefore, the means must be
smoothed prior to calculating the 1C.

13.13.3.8.2  Starting with the observed control mean,_Y.,= 22.4,  and the
observed mean_for the lowest effluent concentration, Y2 = 26.3,  we see that Y\
is less than Y2.

13.13.3.8.3  Calculate the smoothed means:

                M, = M2 = (7, + Y2)/2 = 24.35
                                      186

-------
                                     8
                                     u>
                                     (O
                                           &
                       o
                                                           tj
                                                           •ป~
                                                           I
                                                          •c:
                                                           i-
                                                           
                                                           S-
                                                           0)
4  '   '       o
                                                          4-> S-
                                                           o o.
                                                          i— 0)
                                                          0_ i.
                                                          00

                                                           0)
187

-------
13.13.3.8.4  Since Y3 = 34.6 is larger than M2,  average  Y3 with the previous
concentrations:

                M1 = M2 =  M3 =  (M, + M2 + Y3)/3 =  27.7.

13.13.3.8.5  Additionally, Y4 = 31.7 is larger than M3,  and  is  pooled with  the
first three means.  Thus:
(M, + M2
                 M3 + Y4)/4 = 28.7 = M1 = M2 =  M3 = M4
13.13.3.8.6  Since M4 > Y5 = 9.4,  set M5 = 9.4.   Likewise, M5 > Y6 = 0,  and M6
becomes 0.  Table 15 contains the smoothed means and Figure 8 gives a plot of
the smoothed means and the  interpolated response curve.
           TABLE 15.  DAPHNID, CERIODAPHNIA DUBIA, REPRODUCTION
                      MEAN RESPONSE AFTER SMOOTHING

Effluent
Cone. (%)
Control
1.56
3.12
6.25
12.5
25.0

i
1
2
3
4
5
6
Response
Means, Y,
(young/ female)
22.4
26.3
34.6
31.7
9.4
0.0
Smoothed
Means, M,-
(young/female)
28.75
28.75
28.75
28.75
9.40
0.00
13.13.3.8.7  Estimates of the  IC25 and  IC50 can be calculated  using the  Linear
Interpolation Method.  A 25% reduction  in reproduction, compared to the
controls, would result in a mean reproduction of  21.56 young per adult,  where
M,(l - p/100) = 28.75(1 - 25/100).  A 50% reduction in reproduction, compared
to the controls, would result  in a mean reproduction of 14.38  young per  adult,
where Mt(l - p/100) = 28.75(1 - 50/100).  Examining the smoothed means and
their associated concentrations  (Table  15), the two effluent concentrations
bracketing 21.56 young per adult are C, = 6.25% effluent and C5 = 12.5%
effluent.  The two effluent concentrations bracketing a response of 14.38
young per adult are also C4'= 6.25% and C5 = 12.5%.
                                      188

-------
 13.13.3.8.8  Using equation from Section  4.2  in  Appendix M,  the estimate of
 the IC25 is as follows:
         ICp = C..
              = 6. 25+ [28. 75 (1-25/100) -28.75]
                                                               _
                                                  (9.40-28.75)
              = 8.57% effluent

 13.13.3.8.9   The  estimate  of the  IC50  is as follows:

         ICp  = C+CM                       ~
        IC50 =6.25+128.75(1-50/100) -28.75]   (12-5-6.25)
                                                  (9.40-28.75)

              =  10.89% effluent

 13.13.3.8.10    When the  ICPIN program was used to analyze this data set for
 the  IC25, requesting 80  resamples, the estimate of the IC25 was 8.5715%
 effluent.  The  empirical 95% confidence interval for the true mean was 8.3112%
 and  9.0418% effluent.  The computer output for this data set is provided in
 Figure 9.

 13.13.3.8.11    When the  ICPIN program was used to analyze this data set for
 the  IC50, requesting 80  resamples, the estimate of the IC50 was 10.8931%
 effluent.  The  empirical 95% confidence interval for the true mean was
 10.4373% and  11.6269% effluent.  The computer output for this data set is
 provided in Figure 10.

 13.14  PRECISION AND ACCURACY

 13.14.1  PRECISION

 13.14.1.1  Single- Laboratory Precision

 13.14.1.1.1   Information on the single-laboratory precision of the daphnid,
 Cenodaphma dubia, Survival and reproduction test based on the NOEC  and  LOEC
 yM^n>from nine tests with the reference toxicant sodium pentachlorophenate
 (NaPCP) 1S provided in Table 16.  The NOECs  and LOECs  of all  tests fell  in  the
 same concentration .range, indicating maximum possible  precision   Table  17
gives precision data for the IC25 and IC50 values  for  seven tests  with the
reference toxicant NaPCP.  Coefficient of variation  was  41% for the IC25  and
2o/o for the IC50.

 13.14.1.1.2  Ten sets  of data from six laboratories  met  the acceptability
criteria,  and were statistically analyzed  using nonparametric  procedures to
determine NOECs and LOECs.
                                     189

-------
Cone. ID
Cone. Tested
Response 1
Response 2
Response 3
Response 4
Response 5
Response 6
Response 7
Response 8
Response 9
Response 10
1
0
27
30
29
31
16
15
18
17
14
27
2
1.56
32
35
32
26
18
29
27
16
35
13
3
3.12
39
30
33
33
36
33
33
27
38
44
4
6.25
27
34
36
34
31
27
33
31
33
31
5
12.5
10
13
7
7
7
10
10
16
12
2
6
25.0
0
0
0
0
0
0
0
0
0
0
*** Inhibition Concentration Percentage Estimate ***
Toxicant/Effluent: Effluent
Test Start Date: Example   Test Ending Date:
Test Species: Ceriodaphnia dubia
Test Duration:             7-d
DATA FILE: cdmanual.icp
OUTPUT FILE: cdmanual.i25
Cone.
ID
1
2
3
4
5
6
Number
Replicates
10
10
10
10
10
10
Concentration
%
0.000
1.560
3.120
6.250
12.500
25.000
Response
Means
22.400
26.300
34.600
31.700
9.400
0.000
Std.
Dev. F
6.931
8.001
4.835
2.946
3.893
0.000
Pooled
Response Means
28.750
28.750
28.750
28.750
9.400
0.000
 The  Linear Interpolation  Estimate:      8.5715   Entered P Value:  25

 Number of Resamplings:    80                   .,„.,.•        n  1001
 The  Bootstrap Estimates Mean:    8.5891 Standard Deviation:     ฐ;1831
 Original  Confidence Limits:   Lower:      8.3112 Upper:      9.0418
 Resampling time in Seconds:     2.53   Random Seed:  -641671986
       Figure  9.   Example  of ICPIN  program output  for  the  IC25.
                                     190

-------
Cone. ID
Cone. Tested
Response 1
Response 2
Response 3
Response 4
Response 5
Response 6
Response 7
Response 8
Response 9
Response 10
1
0
27
30
29
31
16
15
18
17
14
27
2
1.56
32
35
32
26
18
29
27
16
35
13
3
3.12
39
30
33
33
36
33
33
27
38
44
4
6.25
27
34
36
34
31
27
33
31
33
31
5
12.5
10
13
7
7
7
10
10
16
12
2
6
25.0
0
0
0
0
0
0
0
0
0
0
*** Inhibition Concentration Percentage Estimate ***
Toxicant/Effluent: Effluent
Test Start Date: Example   Test Ending Date:
Test Species: Ceriodaphnia dubia
Test Duration:             7-d
DATA FILE: cdmanual.icp
OUTPUT FILE: cdmanual.i50
Cone.
ID
1
2
3
4 .
5
6
Number
Replicates
10
10
10
10
10
10
Concentration
%
0.000
1.560
3.120
6.250
12.500
25.000
Response
Means
22.400
26.300
34.600
31.700
9.400
0.000
Std.
Dev . F
6.931
8.001
4.835
2.946
3.893
0.000
Pool ed
Response Means
28.750
28.750
28.750
28.750
9.400
0.000
The Linear Interpolation Estimate:    10.8931   Entered P Value: 50

Number of Resamplings:   80
The Bootstrap Estimates Mean:  10.9316 Standard Deviation:     0.3357
Original Confidence Limits:   Lower:    10.4373 Upper:    11.6269
Resampling time in Seconds:     2.58  Random Seed: 172869646
         Figure 10.   Example of ICPIN program output for the IC50.
                                    191

-------
  TABLE 16.  SINGLE LABORATORY PRECISION OF THE DAPHNID, CERIODAPHNIA
             DUBIA, SURVIVAL AND REPRODUCTION TEST, USING NAPCP AS A
             REFERENCE TOXICANT1'2


Test
I3
24
3
45
5
6
7
8
9
1 For a discussion of
tests see Section 4,
" Pi ซs 4* •ป ฃ\j* n •%* 4* *•ป ** 4* <•ป **i i*i\stฃ

NOEC
(mg/L)
0.25
0.20
0.20
0.30
0.30
0.30
0.30
0.30
0.30
the precision

LOEC
(mg/L)
0.50
0.60
0.60
0.60
0.60
0.60
0.60
0.60
0.60
of data from
Chronic
Value
(mg/L)
6.35
0.35
0.35
0.42
0.42
0.42
0.42
0.42
0.42
chronic toxicity
Quality Assurance.
'*•ป i/ttM/ssJ ts\i fil^ T T T t^ 1 xM.t*irซ Ami<^4ป4^ป Q A r\l r\i~i\i D Vi *\ i^ /^ I
      EMSL-Cincinnati,  OH.   Tests  were conducted in  reconstituted  hard  water
      (hardness  - 180 mg CaCO,/L;'pH = 8.1).
      Concentrations used in Test  1  were:  0.03,  0.06,  0.12,  0.25,  0.50,
      1.0 mg NaPCP/L.
                                   2 and 3 were: 0.007,  0.022,  0.067, 0.20,
Concentrations used in Tests
0.60 mg NaPCP/L.
Concentrations used in Tests 4
0.30, 0.60 mg NaPCP/L.
                                     through  9 were:  0.0375,  0.075,  0.150,
13.14.1.2  Multilaboratory Precision

13.14.1.2.1  A multilaboratory study was performed by the Aquatic Biology
Branch, EMSL-Cincinnati in 1985e, involving a total of 11 analysts in 10
different laboratories (Neiheisel et. a!., 1988; USEPA, 1988e).  Each analyst
performed one-to-three seven-day tests using aliquots of a copper-spiked
effluent sample, for a total of 25 tests.  The tests were performed on the
same day in all participating laboratories, using a pre-publication draft of
Method 1002.0.  The NOECs and LOECs for these tests were within one
concentration interval which, with a dilution factor of 0.5, is equivalent to
a two-fold range in concentration (Table 18).
                                      192

-------
   TABLE  17.   THE  DAPHNID,  CERIODAPHNIA  DUBIA,  SEVEN-DAY  SURVIVAL AND
              REPRODUCTION  TEST  PRECISION  FOR A SINGLE LABORATORY
              USING  NAPCP AS  THE REFERENCE TOXICANT (USEPA,  1991a)
  Test  Number         NOEC (mg/L)          IC25 (mg/L)          IC50 (mg/L)
19
46A
46B
49
55
56
n
Mean
CV(%)
0.30
0.20
0.20
0.20
0.20
0.10
7
NA
NA
0.3754
0.0938
0.2213
0.2303
0.2306
0.2241
7
0.2157
41.1
0.4508
0.2608
0.2879
0.2912
0.3177
0.2827
7
0.2953
27.9
13.14.1.2.2  A second multilaboratory study of Method 1002.0 (using the first
edition of this manual; USEPA, 1985c), was coordinated by Battelle, Columbus
Division, and involved 11 participating laboratories (Table 19) (DeGraeve et
a]., 1989).  All participants used 10% DMW (10% PERRIERฎ'Water) as the culture
and dilution water, and used their own formulation of food for culturing and
testing the Ceriodaphnia dubia.  Each laboratory was to conduct at least one
test with each of eight blind samples.  Each test consisted of 10 replicates
of one organism each for five toxicant concentrations and a control.  Of the
116 tests planned, 91 were successfully initiated, and 70 (77%) met the
survival and reproduction criteria for acceptability of the results (80%
survival and nine young per initial female).  If the reproduction criteria of
15 young/female, used in this edition of the method, had been applied to the
results of the interlaboratory study, 22 additional tests would have been
unacceptable.  The overall precision  (CV) of the test was 27% for the survival
data (7-day LC50s) and 37.5% and 39.0% for the reproduction data (IC50s and
IC25s, respectively).
                                                         i
13.14.2  ACCURACY

13.14.2.1  The accuracy of toxicity tests cannot be determined.
                                      193

-------
TABLE 18.  INTERLABORATORY PRECISION FOR THE DAPHNID, CERIODAPHNIA
           DUBIA, SURVIVAL AND REPRODUCTION TEST WITH COPPER SPIKED
           EFFLUENT (USEPA, 1988e)
Endooints (%
Analyst
3
4
4
5
5
6
6
10
10
11
Test
1
1
2
1
2
1
2
1
2
1
Reproduction
NOEC LOEC
12
6
6
6
12
12
6
6
6
12
25
12
12
12
25
25
12
12
12
25
Effluent)
••
Survival
NOEC LOEC
25
12
25
12
12
25
25
12
12
25
50
25
50
25
25
50
50
25
25
50
                                194

-------
TABLE 19.  INTERLABORATORY PRECISION DATA FOR THE DAPHNID, CERIODAPHNIA
           DUBIA, SUMMARIZED FOR EIGHT REFERENCE TOXICANTS AND EFFLUENTS
           (USEPA, 1991a)
Test Material
Sodium chloride
Industrial
Sodium chloride
Pulp and Paper
Potassium dichromate
Pulp and Paper
Potassium dichromate
Industrial
n
Mean
Standard Deviation
Mean IC50
1.34
3.6
0.96
60.0
35.8
70.2
53.2
69.8

CV%
29.9
83.3
57.4
28.3
30.8
7.5
25.9
37.0
8
37.5
23.0
Mean IC25
1.00
3.2
0.09
47.3
23.4
55.7
29.3
67.3
i
i
CV%
34.3
78.1
44.4
27.0
32.7
12.2
46.8
36.7
8
39.0
19.1
                                    195

-------
                                  SECTION  14

                                  TEST METHOD

              GREEN ALGA, SELENASTRUM CAPRICORNUTUH, GROWTH TEST
                                 METHOD 1003.0

14.1  SCOPE AND APPLICATION

14.1.1  This method measures the chronic toxicity of effluents and receiving
water to the freshwater green alga, Selenastrum capn'cornutum, in a four-day
static test.  The effects include the synergistic, antagonistic, and additive
effects of all the chemical, physical, and biological components which
adversely affect the physiological and biochemical functions of the test
organisms.
                                                                      *
14.1.2  Detection limits of the toxicity of an effluent or pure substance are
organism dependent.

14.1.3  Brief excursions in toxicity may not be detected using 24-h composite
samples.  Also, because of the long sample collection period involved in
composite sampling, and because the test chambers are not sealed, highly
degradable or highly volatile toxicants present in the source may not be
detected in the test.

14.1.4  This test method is commonly used  in one of two forms:  (1) a
definitive test, consisting of a minimum of five effluent concentrations and a
control, and (2) a receiving water test(s), consisting of one or more
receiving water concentrations and a control.

14.1.5  This test is very versatile because it can also be used to identify
wastewaters which are biostimulatory and may cause nuisance growths of algae,
aquatic weeds, and other organisms at higher trophic levels.

14.2  SUMMARY OF METHOD

14.2.1  A green alga, Selenastrum capricornutum, population is exposed in a
static system to a series of concentrations of effluent, or to receiving
water, for 96 h.  The response of the population is measured in terms of
changes in cell density (cell counts per ml), biomass, chlorophyll  content, or
absorbance.

14.3  INTERFERENCES

14.3.1  Toxic substances may be introduced by contaminants in dilution water,
glassware, sample hardware,  and testing equipment (see Section 5, Facilities,
Equipment, and Supplies).

14.3.2  Adverse effects of high concentrations of suspended and/or dissolved
solids, color, and extremes of pH may mask the presence of toxic substances.
                                      196

-------
14.3.3  Improper effluent sampling and handling may adversely affect test
results (see Section 8, Effluent and Receiving Water Sampling, Sample
Handling,  and Sample Preparation for Toxicity Tests).
14.3.4  Pathogenic organisms and/or planktivores in the dilution water and
effluent may affect test organism survival and growth, and confound test
results.
14.3.5  Nutrients in the effluent or dilution water may confound test results.
14.4  SAFETY
14.4.1  See Section 3, Safety and Health.
14.5  APPARATUS AND EQUIPMENT
14.5.1  Laboratory Selenastrum capricornutum culture unit.—  see culturing
methods below and USEPA, 1993b.  To test effluent toxicity, sufficient numbers
of log-phase-growth organisms must be available.
14.5.2  Samplers -- automatic sampler, preferably with sample cooling
capability, that can collect a 24-h composite sample of 5 L or more.
14.5.3  Sample containers -- for sample shipment and storage  (see Section 8,
Effluent and Receiving Water Sampling, Sample Handling, and Sample Preparation
for Toxicity Tests).
14.5.4  Environmental chamber, incubator, or equivalent facility -- with
"cool-white" fluorescent illumination (86 ฑ 8.6 fiE/m /s, 400 ฑ 40 ft-c, or
4306  lux) and temperature control  (25 ฑ 1ฐC).
14.5.5  Mechanical shaker -- capable of providing orbital motion at the rate
of 100  cycles per minute (cpm).
14.5.6  Light meter -- with a range of 0-200 /*E/m2/s (0-1000 ft-c).
14.5.7  Water purification system  -- MILLIPORE MILLI-Qฎ,  deionized water or
equivalent  (see Section  5, Facilities, Equipment, and Supplies).
14.5.8  Balance -- analytical, capable of accurately weighing 0.00001  g.
14.5.9  Reference weights, class S  -- for checking  performance  of balance.
14.5.10 Volumetric flasks and graduated  cylinders  -- class A,  10-1000 mL,
borosilicate glass, for  culture work and  preparation  of test  solutions.
14.5.11 Volumetric pipets --  class A, 1-100 mL.
14.5.12 Serological  pipets  --  1-10 mL, graduated.
14.5.13 Pipet  bulbs  and fillers  -- PROPIPETฎ,  or equivalent.

                                      197

-------
 14.5.14  Wash bottles -- for rinsing small  glassware,  instrument  electrodes,
 and probes.
 14.5.15  Test chambers -- four (minimum of  three)  125  or  250  mL borosilicate,
 Erlenmeyer flasks,  with foam plugs  or stainless  steel  or  Shumadzu closures.
 For special  glassware cleaning requirements (see Section  5, Facilities,
 Equipment,  and Supplies).
 14.5.16  Culture chambers -- 1-4  L  borosilicate, Erlenmeyer flasks.
 14.5.17  Thermometers,  glass or electronic,  laboratory grade  -- for measuring
 water  temperatures.
 14.5.18  Bulb-thermograph or electronic-chart  type  thermometers -- for
 continuously recording temperature.
 14.5.19  Thermometer,  National  Bureau of Standards  Certified,  (see USEPA
 Method 170.1,  USEPA,  1979b)  --  to calibrate  laboratory thermometers.
 14.5.20  Meters,  pH and specific conductivity  -- for routine  physical and
 chemical  measurements.
 14.5.21  Tissue  grinder -- for  chlorophyll  extraction.
 14.5.22  Fluorometer  (Optional) --  equipped  with chlorophyll  detection light
 source, filters,  and  photomultiplier  tube (Turner Model 110 or equivalent).
 14.5.23  UV-VIS  spectrophotometer --  capable of  accommodating  1-5 cm cuvettes.
 14.5.24  Cuvettes for  spectrophotometer  --  1-5 cm light path.
 14.5.25  Electronic particle counter  (Optional)  --  Coulter Counter, Model ZBI,
 or  equivalent, with mean  cell  (particle) volume determination.
 14.5.26  Microscope --  with  10X, 45X,  and 100X objective  lenses,  10X ocular
 lenses, mechanical stage, substage  condenser, and light source (inverted or
 conventional microscope).
 14.5.27   Counting chamber --  Sedgwick-Rafter, Palmer-Maioney,  or
 hemocytometer.
 14.5.28   Centrifuge --  with  swing-out  buckets having a capacity of 15-100 ml.
 14.5.29   Centrifuge tubes --  15-100 ml, screw-cap.
 14.5.30   Filtering apparatus  -- for membrane and/or glass fiber filters.
 14.6  REAGENTS AND CONSUMABLE MATERIALS
14.6.1  Sample containers --  for sample shipment and storage (see Section 8,
Effluent and Receiving Water Sampling, Sample Handling, and Sample Preparation
for Toxicity Tests).
                                      198

-------
14.6.2  Data sheets (one set per test) -- for recording data.

14.6.3  Tape, colored -- for labeling test chambers.

14.6.4  Markers, waterproof -- for marking containers, etc.

14.6.5  Reagents for hardness and alkalinity tests -- see USIEPA Methods 130.2
and 310.1, USEPA, 1979b.

14.6.6  Buffers pH 4, pH 7, and pH 10 (or as per instructions of instrument
manufacturer) for instrument calibration (see USEPA Method 150.1, USEPA,
1979b).

14.6.7  Specific conductivity standards (see USEPA Method 120.1, USEPA,
1979b).                                                  :

14.6.8  Standard particles -- such as chicken or turkey fibroblasts or polymer
microspheres, 5.0 ฑ 0.03 urn diameter, 65.4 /*m  volume, for calibration of
electronic particle counters (available from Duke Scientific Co., 1135D, San
Antonio Road, Palo Alto, CA 94303).

14.6.9  Membranes and filling solutions for DO probe  (see USEPA Method 360.1,
USEPA, 1979b), or reagents -- for modified Winkler analysis.

14.6.10  Laboratory quality control samples and standards -- for calibration
of the above methods.

14.6.11  Reference toxicant solutions -- see Section  4, Quality Assurance.

14.6.12  Reagent water  -- defined as distilled or deionized water that does
not contain substances which are toxic to the test organisms (see Section 5,
Facilities, Equipment,  and Supplies).

14.6.13  Effluent or receiving water and dilution water -- see Section 7,
Dilution Water; and Section 8, Effluent and Receiving Water Sampling, Sample
Handling, and Sample Preparation for Toxicity Testing.

14.6.14  Acetone -- pesticide-grade or equivalent.

14.6.15  Dilute  (10%) hydrochloric acid -- carefully  add 10 ml of concentrated
HC1 to 90 ml of MILLI-Qฎ water.                          :

14.6.16  TEST ORGANISMS, GREEN ALGA, SELENASTRUM CAPRICORNUTUM

14.6.16.1  Selenastrum  capn'cornutum, a unicellular coccoid green alga is the
test organism.

14.6.16.2  Algal Culture Medium  is prepared as follows:

   1.  Prepare  (five) stock nutrient solutions using reagent grade chemicals as
       described  in Table 1.  Cautionary note:  EDTA may affect metal  toxicity.
       It  is recommended that tests be conducted with  and without EDTA in the

                                      199

-------
 TABLE 1.   NUTRIENT STOCK  SOLUTIONS FOR MAINTAINING ALGAL  STOCK CULTURES
            AND TEST CONTROL CULTURES
STOCK            COMPOUND                       AMOUNT DISSOLVED IN
SOLUTION                                       500  mL  MILLI-Qฎ WATER
1. MACRONUTRIENTS

   A.             MgCl2.6H20                          6.08  g
                  CaCl,-2H20                          2.20  g
                  NaN03                              12.75  g

   B.             MgS04-7H20                          7.35  g

   C.             K2HP04                              0.522  g

   D.             NaHC03            .                  7.50  g

2. MICRONUTRIENTS

                  H,B03                              92.8   mg
                  MnCl2.4H20                        208.0   mg
                  ZnC12                               1.64  mg1
                  FeCl3.6H20                         79.9   nra
                  CoCl2-6H20                          0.714  mg,
                  Na2MoO,-2H20                         3.63  mg,
                  CuCl2-2H20                          0.006  mg4
                  Na2EDTA.2H20                       150.0   mg
                  Na2Se04                             1.196  mg5
   ZnCl? - Weigh out 164 mg and dilute to 100 mL.  Add 1 mL of this
   solution to Stock 2, micronutrients.
   CoCl2*6H20  -  Weigh  out  71.4 mg and dilute to 100 mL.  Add 1 mL of
   this solution to Stock 2, micronutrients.
   NajjMoO^HjjO  - Weigh  out  36.6  mg  and dilute  to 10  mL.   Add  1  mL
   of this solution to  Stock 2,  micronutrients.
   CuCl2-2H20  -  Weigh  out  60.0 mg and dilute to 1000 mL.   Take 1 mL of
   this solution and dilute to 10 mL.  Take  1 mL of  the  second  dilution
   and add to Stock 2,  micronutrients.
   Na?Se04 - Weigh  out  119.6 mg and  dilute to 100 mL.  Add 1 mL of this
   solution to Stock 2, micronutrients.
                                    200

-------
       culture  media  if metals  are  suspected  in  the  effluent  or receiving
       water.
    2.  Add  1 ml of  each stock solution,  in  the order listed  in  Table  1,  to
       approximately  900 ml  of  MILLI-Qฎ  water.   Mix  well  after  the  addition  of
       each solution.   Dilute to  1  L,  mix well,  and  adjust the  pH to  7.5 ฑ 0.1,
       using 0.1N NaOH  or HC1,  as appropriate.   The  final concentration  of
       macronutrients and micronutrients in the  culture medium  is given  in
       Table 2.
    3.  Immediately  filter the pH-adjusted medium through  a 0.45 /zm  pore
       diameter membrane at  a vacuum of  not more than 380 imm  (15 in.)  mercury,
       or at a  pressure of not  more than one-half atmosphere  (8 psi).  Wash  the
       filter with  500  mL deionized water prior  to use.
    4.  If the filtration is  carried out  with  sterile apparatus,  filtered medium
       can be used  immediately, and no further sterilization  steps  are required
       before the inoculation of  the medium.  The medium can  also be  sterilized
       by autoclaving after  it  is placed in the  culture vessels.  If  a 0.22  ^g
       filter is used no sterilization is needed.
    5.  Unused sterile medium should not  be  stored more than one week  prior to
       use, because there may be  substantial  loss of water by evaporation.

14.6.16.3  Stock Algal  Cultures

14.6.16.3.1  See Section  6, Test Organisms,  for information on  sources  of
"starter" cultures of  the green  alga, Selenastrum capricornutum.

14.6.16.3.2  Upon receipt of the "starter" culture  (usually about  10  ml), a
stock  culture  is initiated  by  aseptically  transferring 1 rnL to  a culture flask
containing control algal  culture medium (prepared as described  above).  The
volume of stock culture  medium initially prepared will depend  upon the  number
of  test flasks to be inoculated  later from the  stock, or other  planned  uses,
and may range  from 25  ml  in a  125  ml  flask to 2  L in a 4-1. flask.   The
remainder of the starter  culture can  be held in  reserve for up  to six months
in  a refrigerator (in  the dark)  at 4ฐC.

14.6.16.3.3  Maintain  the stock  cultures at 25 ฑ  1ฐC, under continuous
"Cool-White" fluorescent  lighting  of  86 ฑ  8.6 /*E/m2/s (400  ฑ 40 ft-c). Shake
continuously at 100 cpm  or  twice daily  by  hand.

14.6.16.3.4  Transfer  1 to  2 mL  of stock culture weekly to 50  - 100 ml of new
culture medium to maintain  a continuous supply of "healthy" cells  for tests.
Aseptic techniques should be used  in maintaining the algal  cultures,  and
extreme care should be exercised to avoid  contamination.   Examine  the stock
cultures with a microscope  for contaminating microorganisms at each transfer.

14.6.16.3.5  Viable unialgal culture material may be maintained for long
periods of time if placed in a refrigerator at 4ฐC.

14.6.16.4  It is recommended that chronic toxicity tests  be performed monthly
with a reference toxicant.  Algal cells four to  seven days;  old are  used to
monitor the chronic toxicity (growth)  of the reference toxicant to  the algal
stock produced by the culture unit (see Section  4, Quality  Assurance,
Subsection 4.17).

                                      201                 i

-------
TABLE 2.  FINAL CONCENTRATION OF MACRONUTRIENTS AND MICRONUTRIENTS
          IN THE CULTURE MEDIUM
MACRONUTRIENT
NaN03
MgCl2-6H20
CaCl2.2H20
MgS04-7H20
K2HPOA
NaHC03


MICRONUTRIENT
H3B03
MnCl2-4H20
ZnCl2
CoCl2-6H20
CuCl2-2H20
Na2Mo04-2H20
FeCT3.6H20
Na2EDTA-2H20
Na2Se04
CONCENTRATION
Ima/D
25.5
12.2
4.41
14.7
1.04
15.0


CONCENTRATION
(ua/Ll
185.0
416.0
3.27
1.43
0.012
7.26
160.0
300.0
2.39
ELEMENT
N
Mg
Ca
S
P
Na
K
C
ELEMENT
B
Mn
Zn
Co
Cu
Mo
Fe
--
Se
CONCENTRATION
lm/L}
4.20
2.90
1.20
1.91
0.186
11.0
0.469
2.14
CONCENTRATION
fua/L)
32.5
115.0
1.57
0.354
0.004
2.88
33.1
	
0.91
                                 202

-------
14.6.16.5  Record Keeping                                ',.

14.6.16.5.1  Records, kept in a bound notebook, include (1) dates culture
media was prepared, (2) source of "starter" cultures, (3) date stock cultures
were started, (4) cell density in stock cultures, and (5) dates and results of
reference toxicant tests performed (see Section 4, Quality Assurance).

14.7  EFFLUENT AND RECEIVING MATER COLLECTION, PRESERVATION, AND STORAGE

14.7.1  See Section 8, Effluent and Receiving Water Sampling, Sample Handling,
and Sample Preparation for Toxicity Tests.

14.8  CALIBRATION AND STANDARDIZATION

14.8.1  See Section 4, Quality Assurance.

14.9  QUALITY CONTROL

14.9.1  See Section 4, Quality Assurance.

14.10  TEST PROCEDURES

14.10.1  TEST SOLUTIONS

14.10.1.1  Receiving  Waters

14.10.1.1.1  The sampling  point  is determined  by  the  objectives  of  the  test.
Receiving  water  toxicity  is  determined with  samples  used  directly as  collected
or  after samples are  passed  through  a 60  urn  NITEXฎ filter and compared  without
dilution against a control.   Using four replicate chambers per test,  each
containing 100 ml  and 400  ml  for chemical  analyses,  would require
approximately  1  L or  more  of sample  for the  test.

14.10.1.2  Effluents

14.10.1.2.1  The selection of the effluent test  concentrations should be based
on  the  objectives of the  study.   A dilution  factor of 0.5 is commonly used.   A
dilution factor  of 0.5 provides  precision of ฑ 100%,  and  testing of
concentrations  between 6.25% and 100% effluent using five effluent
concentrations  (6.25%,  12.5%, 25%,  50%,  and  100%).   Improvements in precision
decline rapidly  if the dilution  factor  is increased  beyond 0.5 and  precision
declines rapidly if  a smaller dilution, factor is used,   therefore,  USEPA
recommends using a > 0.5 dilution factor.

 14.10.1.2.2   If the  effluent is  known or suspected  to be highly  toxic,  a lower
 range of effluent concentrations should be used  (such as 25%,  12.5%,  6.25%,
 3.12%,  and 1.56%).  If a high rate of mortality is  observed during  the first
 1 to 2 h of  the test, additional dilutions should be added at the lower range
 of the effluent concentrations.
                                       203

-------
 14.10.1.2.3   The  volume  of effluent  required  for  the test  is  1 to 2  L.
 Sufficient test solution (approximately  900 or  1500 ml)  is  prepared  at each
 effluent  concentration to provide  400  ml additional volume  for chemical
 analyses  at  the high, medium,  and  low  test concentrations.  There is no daily
 renewal of test solution.

 14.10.1.2.4   Tests  should begin  as soon  as possible, preferably within 24 h of
 sample collection.  The  maximum  holding  time  following retrieval of  the sample
 from the  sampling device should  not  exceed 36 h for off-site  toxicity tests
 unless permission is granted by  the  permitting  authority.   In no case should
 the sample be used  in a  test more  than 72 h after sample collection  (see
 Section 8, Effluent and  Receiving  Water  Sampling,  Sample Handling, and Sample
 Preparation  for Toxicity Tests).                                      :

 14.10.1.2.5   Just prior  to  test  initiation (approximately 1 h) the temperature
 of sufficient quantity of the  sample to  make  test solutions should be adjusted
 to the test  temperature  and maintained at that  temperature  during the addition
 of dilution  water.

 14.10.1.2.6   The  DO of the  test  solutions should  be checked prior to test
 initiation.   If any of the  solutions are supersaturated with  oxygen or any
 solution  has  a DO concentration  below  4.0 mg/L, all of the  solutions and the
 control must  be gently aerated.

 14.10.1.2.7   Effluents may  be  toxic  and/or nutrient poor.   "Poor" growth in an
 algal toxicity test, therefore,  may  be due to toxicity or nutrient limitation,
 or both.  To  eliminate false negative  results due to low nutrient
 concentrations, 1 ml of  each stock nutrient solution is added per liter of
 effluent  prior to use in  preparing the test dilutions.  Thus, all test
 treatments and controls will contain at  a minimum the concentration of
 nutrients in  the  stock culture medium.

 14.10.1.2.8   If samples  contain  volatile substances, the test sample should be
 added below the surface of  the dilution water towards the bottom of the test
 container through an appropriate delivery tube.

 14.10.1.3  Dilution Water

 14.10.1.3.1   Dilution water may  be stock culture  medium,  any uncontaminated
 receiving water,  a standard synthetic  (reconstituted) water, or some other
 natural water  (see Section  7, Dilution Water).  However,  if water other than
 the stock culture medium  is used for dilution water, 1 ml of each stock
 nutrient solution should be added  per  liter of dilution water.  Natural  waters
 used as dilution water must be filtered through a prewashed filter,  such as a
 6F/A,  GF/C,  or equivalent filter, that provides 0.45 ^m particle size
 retention.

 14.10.1.3.2   If the growth of the  algae in the test solutions is to be
measured with an electronic particle counter,  the effluent and dilution  water
must be filtered through a GF/A or GF/C filter,  or other  filter providing
 0.45 urn particle size retention,  and checked for  "background" particle count


                                     204

-------
before it is used in the test.   Glass-fiber filters generally provide more
rapid filtering rates and greater filtrate volume before plugging.

14.10.1.4  Preparation of Inoculum

14.10.1.4.1  The inoculum is prepared no more than 2 to 3 h prior to the
beginning of the test, using SeTenastrum capricornutum harvested from a four-
to-seven-day stock culture.  Each milliliter of inoculum must contain enough
cells to provide an initial cell density of approximately 10,000 cells/ml
(ฑ 10%) in the test flasks.  Assuming the use of 250 ml flasks, each
containing 100 ml of test solution, the inoculum must contain 1,000,000
cells/ml.

14.10.1.4.2  Estimate the volume of stock culture required to prepare the
inoculum.  As an example, if the four-to-seven-day-old stock culture used as
the source of the inoculum has a cell density of 2,000,000 cells/ml, a test
employing 24 flasks, each containing 100 ml of test medium and inoculated with
a total of 1,000,000 cells, would require 24,000,000 cells or 15 ml of stock
solution (24,000,000/2,000,000) to provide sufficient inoculum.  It is
advisable to prepare a volume 20% to 50% in excess of the minimum volume
required, to cover accidental loss in transfer and handling.

14.10.1.4.3  Prepare the inoculum as follows:

   1. Centrifuge 15 ml of stock culture at 1000 x g for 5 min.  This volume
      will provide a 50% excess in the number of cells.
   2. Decant the supernatant and resuspend the cells in 10 ml of control
      medium.
   3. Repeat the centrifugation and decantation step, and resuspend the cells
      in 10 ml control medium.
   4. Mix well and determine the cell density in the algal concentrate.  Some
      cells will be lost in the concentration process.
   5. Determine the density of cells (cells/ml) in the stock culture  (for this
      example, assume 2,000,000 per ml).
   6. Calculate the required volume of stock culture as follows:
                                                         i  •

Volume  (ml) of      Number test flasks      Volume of test             10,000
Stock Culture   =       to be used.     x    Solutions/flask     x    cells/ml
  Required                Cell density  (cells/ml) in the stock culture

                =     24 flasks x  100 ml/flask x 10.000 cells/ml
                                 2,000,000 cells/ml      ;

                =    12.0 ml Stock  Culture                ,

   7.  Dilute  the cell concentrate  as needed to obtain a cell density  of
       1,000,000 cells/ml,  and check  the cell density in the final inoculum.
   8.  The  volume of the  algal  inoculum  should be considered in calculating  the
       dilution  of  toxicant  in the  test  flasks.           ;
                                      205

-------
 14.10.2  START OF THE TEST

 14.10.2.1  Label  the test chambers with a marking pen  and  use the color-coded
 tape to identify  each treatment and replicate.   A minimum  of five effluent
 concentrations and a control  are used for each  effluent  test.   Each  treatment
 (including the control)  should have four (minimum of three)  replicates.

 14.10.2.2  Randomize the position of the test flasks at  the  beginning  of  the
 test (see Appendix A).   Preparation of a position chart  may  be helpful.

 14.10.2.3  The test begins when the algae are added  to the test flasks.   Mix
 the  inoculum well,  and add 1  ml to the test  solution in  each randomly  arranged
 flask.   Make a final  check of the cell  density  in three  of the test  solutions
 at time "zero" (within 2 h of the inoculation).

 14.10.2.3.1  Alkalinity,  hardness,  and conductivity  are  measured  at  the
 beginning of the  test in the  high,  medium, and  low effluent  concentrations  and
 control  before they are  dispensed to the test chambers and the data  recorded
 on the  data sheet (Figure 1).
    Discharger:
    Location:
Test Dates:
  Analyst:
                               Effluent Concentration
Parameter
Temperature
DH
Alkalinitv
Hardness
Conductivity
Chlorine

Control























































Remarks







    Figure 1.   Data  form  for the green  alga,  Selenastrum capn'cornutum,
                growth test.  Routine chemical and physical determinations.
14.10.3  LIGHT, PHOTOPERIOD, AND TEMPERATURE

14.10.3.1  Test flasks are incubated under continuous illumination at 86 ฑ
8.6 nE/m/s (400 ฑ 40 ft-c),  at 25 ฑ 1ฐC,  and should be shaken continuously at
100 cpm on a mechanical shaker or twice daily by hand.  Flask positions in the
incubator should be randomly rotated each day to minimize possible spatial
differences in illumination and temperature on growth rate.  If it can be
verified that test specifications are met at all positions, this need not be
done.

                                      206

-------
14.10.4  DISSOLVED OXYGEN (DO) CONCENTRATION

14.10.4.1  Because of the continuous illumination of the test flasks, DO
concentration should never be a problem during the test and no aeration will
be required.

14.10.5  OBSERVATIONS DURING THE TEST

14.10.5.1  Routine Chemical and Physical Determinations

14.10.5.1.1  Temperature should be monitored continuously or observed and
recorded daily for at least two locations in the environmental control system
or the samples.  Temperature should be checked in a sufficient number of test
vessels at least at the end of the test to determine variability in the
environmental chamber.

14.10.5.1.2  Temperature and pH are measured at the end of each 24-h exposure
period in at least one test flask at each concentration and in the control.

14.10.5.1.3  Record all the measurements on the data sheet (Figure 1).

14.10.5.2  Biological Observations

14.10.5.2.1  Toxic substances in the test solutions may degrade or volatilize
rapidly, and the inhibition in algal growth may be detectable only during the
first one or two days in the test.  It may be desirable, therefore, to
determine the algal growth response daily.  Otherwise, biological observations
are not required until the test is terminated and the test solutions are not
renewed during the test period.

14.10.6  TERMINATION OF THE TEST
                                                         I
14.10.6.1  The test is terminated 96 h after initiation.  The algal growth  in
each flask is measured by one of the following methods:  (a) cell counts, (b)
chlorophyll content, or (c) turbidity (light absorbance).

14.10.6.2  Cell counts

14.10.6.2.1  Automatic Particle Counters

14.10.6.2.1.1  Several types of automatic electronic and optical particle
counters are available for use in the rapid determination of cell density
(cells/ml) and mean cell volume (MCV) in MN /cell.  The Coulter Counter is
widely used and is discussed in detail  in USEPA  (1978b).

14.10.6.2.1.2  If biomass data are desired for algal growth potential
measurements, a Model ZM Coulter Counter is used.  However, the instrument
must be calibrated with a reference sample of particles of known volume.

14.10.6.2.1.3  When the Coulter Counter is used, an aliquot (usually 1 mL)  of
the test culture is diluted 10X to 20X with a 1% sodium chloride electrolyte
solution,  such as ISOTONฎ, to facilitate counting.  The resulting dilution  is

                                      207

-------
counted using an aperture tube with a lOO-y^m diameter aperture.  Each cell
(particle) passing through the aperture causes a voltage drop proportional to
its volume.  Depending on the model, the instrument stores the information on
the number of particles and the volume of each, and calculates the mean cell
volume.  The following procedure is used:

   1. Mix the algal culture in the flask thoroughly by swirling the contents
      of the flask approximately six times in a clockwise direction, and then
      six times in the reverse direction; repeat the two-step process at least
      once.
   2. At the end of the mixing process, stop the motion of the liquid in the
      flask with a strong brief reverse mixing action, and quickly remove 1 ml
      of cell culture from the flask with a sterile pi pet.
   3. Place the aliquot in a counting beaker, and add 9 ml (or 19 ml) of
      electrolyte solution (such as Coulter ISOTONฎ).
   4. Determine the cell density (and MCV, if desired).

14.10.6.2.2  Manual microscope counting method

14.10.6.2.2.1  Cell counts may be determined using a Sedgwick-Rafter,
Palmer-Maioney, hemocytometer, inverted microscope, or similar methods.  For
details on microscope counting methods, see APHA (1992) and USEPA (1973).
Whenever feasible, 400 cells per replicate are counted to obtain ฑ 10%
precision at the 95% confidence level.  This method has the advantage of
allowing for the direct examination of the condition of the cells.

14.10.6.3  Chlorophyll Content

14.10.6.3.1  Chlorophyll may be estimated in-vivo fluorometrically, or
in-vitro either fluorometrically or spectrophotometrically.  In-vivo
fluorometric measurements are recommended because of the simplicity and
sensitivity of the technique and rapidity with which the measurements can be
made (Rehnberg et al., 1982).

14.10.6.3.2  The in-vivo chlorophyll measurements are made as follows:

   1. Adjust the "blank" reading of the fluorometer using the filtrate from an
      equivalent dilution of effluent filtered through a 0.45 /j,m particle
      retention filter.
   2. Mix the contents of the test culture flask by swirling successively in
      opposite directions (at least three times), and remove 1 ml of culture
      from the flask with a sterile pi pet.
   3. Place the aliquot in a small  disposable vial  and record the fluorescence
      as soon as the reading stabilizes.  (Do not allow the sample to stand in
      the instrument more than 1 min).
   4. Discard the sample.

14.10.6.3.3  For additional  information on chlorophyll measurement methods,
(see APHA, 1992).
                                      208

-------
14.10.6.4  Turbidity (Absorbance)

14.10.6.4.1  A second rapid technique for growth measurement involves the use
of a spectrophotometer to determine the turbidity, or absorbance, of the
cultures at a wavelength of 750 nm.  Because absorbance is a complex function
of the volume, size, and pigmentation of the algae, it would be useful to
construct a calibration curve to establish the relationship between absorbance
and cell density.

14.10.6.4.2  The algal growth measurements are made as follows:

   1. A blank is prepared as described for the fluorometric analysis.
   2. The culture is thoroughly mixed as described above.
   3. Sufficient sample is withdrawn from the test flask with a sterile pipet
      and transferred to a 1- to 5-cm cuvette.
   4. The absorbance is read at 750 nm and divided by the light path length of
      the cuvette, to obtain an "absorbance-per-centimeter" value.
   5. The 1-cm absorbance values are used in the same manner as the cell
      counts.
                                                         i
14.10.6.5  Record the data as indicated in Figure 2.
  Discharger:
  Location:
                           Test Dates:
                            Analyst: _
Concentration
Control
Cone:
Cone:
Cone:
Cone:
Cone:
Cell
1






Density Measurement
Replicate
2






3






4






Treatment
Mean






Comments






  Comments:

     Figure 2.
Data form for the
growth test, cell
green alga,  Selenastrunt capricornutum,
density determinations.
14.11  SUMMARY OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA

14.11.1  A summary of test conditions and test acceptability criteria is
presented in Table 3.

14.12  ACCEPTABILITY OF TEST RESULTS

14.12.1  For the test results to be acceptable, the algal cell density in the
control flasks must exceed 1 X 106 cells/ml with EDTA or 2 X 105  cells/ml
without EDTA at the end of the test, and not vary more than 20% among
replicates.

                                      209

-------
TABLE 3.  SUMMARY OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA FOR
          GREEN ALGA, SELENASTRUM CAPRICORNUTUM, GROWTH TOXICITY TESTS
          WITH EFFLUENTS AND RECEIVING WATERS
  1. Test type:
  2. Temperature:
  3. Light quality:

  4. Light intensity:
  5. Photoperiod:
  6. Test chamber size:
  7. Test solution volume:
  8. Renewal of test solutions:
  9. Age of test organisms:
  9. Initial cell density in
      test chambers:
 10. No. replicate chambers
      per concentration:
 11. Shaking rate:

 12. Aeration:
 13. Dilution water:
Static non-renewal
25 ฑ 1ฐC
"Cool white" fluorescent
lighting
86 ฑ 8.6 /.E/m2/s (400 ฑ 40 ft-c
or 4306 lux)
Continuous i11umination
125 mL or 250 mL
50 mL or 100 mL1
None
4 to 7 days

10,000 cells/mL

4 (minimum of 3)
100 cpm continuous, or twice
daily by hand
None
Algal stock culture medium,
enriched uncontaminated source
of receiving or other natural
water, synthetic water prepared
using MILLIPORE MILLI-Qฎ or
equivalent deionized water and
reagent grade chemicals, or DMW
(see Section 7, Dilution Water)
  For tests not continuously shaken use 25 mL in 125 mL flasks and 50 mL
  in 250 mL flasks.
                                   210

-------
 TABLE 3.   SUMMARY OF TEST CONDITIONS AND TEST ACCEPTABILITY CRITERIA FOR
           GREEN ALGA, SELENASTRUM CAPRICORNUTUM, GROWTH TOXICITY TESTS
           WITH EFFLUENTS AND RECEIVING WATERS (CONTINUED)
14.  Test concentrations:



15.  Test dilution factor:


16.  Test duration:

17.  Endpoint:
18. Test acceptability
      criteria:
19. Sampling requirements:
20. Sample volume required:
Effluents: Minimum of 5 and a control
Receiving Water: 100% receiving water
or minimum of 5 and a control

Effluents:  > 0.5
Receiving Waters:  None or > 0.5

96 h

Growth (cell counts, chlorophyll
fluorescence, absorbance, biomass)
1 X 106 cells/mL with EDTA or 2 X 10s
cells/mL without EDTA in the controls:
Variability of controls should not
exceed 20%

For on-site tests, one sample collected
at test initiation, and used within 24
h of the time it is removed from the
sampling device.  For off-site tests,
holding time must not exceed 36 h (see
Section 8, Effluent and Receiving Water
Sampling, Sample Handling, and Sample
Preparation for Toxicity Tests,
Subsection 8.5.4)

1 or 2 L depending on test volume
                                    211

-------
14.13  DATA ANALYSIS

14.13.1  GENERAL

14.13.1.1  Tabulate and summarize the data.  A sample set of algal growth
response data is shown in Table 4.

   TABLE 4.  GREEN ALGA, SELENASTRUM CAPRICORNUTUM, GROWTH RESPONSE DATA
Toxicant Concentration
                                                              Cd/L)
       Replicate  Control
     10
20
40
80
A
B
C
1209
1180
1340
1212
1186
1204
826
628
816
493
416
413
127
147
147
49.3
40.0
44.0
Log10
"• 1 U
Trans-
formed
A
B
C
3.082
3.072
3.127
3.084
3.074
3.081
2.917
2.798
2.912
2.693
2.619
2.616
2.104
2.167
2.167
1.693
1.602
1.643
       Mean(Y,.)     3.094     3.080    2.876    2.643    2.146
                               1.646
14.13.1.2  The endpoints of toxicity tests using the green alga, Selenastrum
capricornutum, are based on the adverse effects on cell growth  (see Section 9,
Chronic Toxicity Test Endpoints and Data Analysis).  The EC50, the IC25, and
the IC50 are calculated using the point estimation techniques, and LOEC and
NOEC values for growth are obtained using a hypothesis testing approach such
as Dunnett's Procedure (Dunnett, 1955) or Steel's Many-one Rank Test (Steel,
1959; Miller, 1981).  Separate analyses are performed for the estimation of
the LOEC and NOEC endpoints and for the estimation of the EC50, IC25, and
IC50.  See the Appendices for examples of the manual computations, and
examples of data input and program output.

14,13.1.3  The statistical tests described here must be used with a knowledge
of the assumptions upon which the tests are contingent.  Tests for normality
and homogeneity of variance are included in Appendix B.  The assistance of a
statistician is recommended for analysts who are not proficient in statistics.

14.13.2  EXAMPLE OF ANALYSIS OF ALGAL GROWTH DATA

14.13.2.1  Formal statistical analysis of the growth data is outlined on the
flowchart in Figure 3.  The response used in the statistical analysis is the
number of cells per millilHer per replicate.  Separate analyses are performed
for the estimation of the NOEC and LOEC endpoints and for the estimation* of'
the IC25 and IC50 endpoints.

                                      212

-------
                STATISTICAL ANALYSIS OF ALGAL GROWTH TEST
                             GROWTH RESPONSE DATA
                                   CELLS/ML
    POINT ESTIMATION
    ENDPOINT ESTIMATE
        IC25, IC50
SHAPIRO-WILK-STEST
                    NON-NORMAL DISTRIBUTION
                   NORMAL DISTRIBUTION
         HOMOGENEOUS
           VARIANCE
                                BARTLETTSTEST
                          HETEROGENEOUS
                             VARIANCE
1

NO
r1 .
EQUAL NUMI
REPLJCAT

T-TESTWITH
BONFERRONI
ADJUSTMENT



JEROF
ES?
i YES
DUNNETTS
TEST





YES
EQUAL NUMBER OF
REPLICATES?

STEEL'S MANY-ONE
RANK TEST



i

NO
WII.COXON RANK SUM
TEST WITH
BONFERRONI ADJUSTMENT


                              ENDPOINT ESTIMATES
                                  NOEC, LOEC
Figure 3.    Flowchart  for  statistical  analysis of the green  alga, Selenastrum
            capricornutum, growth response data.

                                     213

-------
14.13.2.2  The statistical analysis using hypothesis tests consists of a
parametric test, Dimnett's Procedure, and a nonparametric test, Steel's
Many-one Rank Test.  The underlying assumptions of the Dunnett's Procedure,
normality and homogeneity of variance, are formally tested.  The test for
normality is the Shapiro-Wilk's Test, and Bartlett's Test is used to test for
homogeneity of variance.  If either of these tests fails, the nonparametric
test, Steel's Many-one Rank Tests, is used to determine the NOEC and LOEC
endpoints.  If the assumptions of Dunnett's Procedure are met, the endpoints
are determined by the parametric test.

14.13.2.3  Additionally, if unequal numbers of replicates occur among the
concentration levels tested there are parametric and nonparametric alternative
analyses.  The parametric analysis is a t test with the Bonferroni adjustment
(see Appendix D).  The Wilcoxon Rank Sum Test with the Bonferroni adjustment
is the nonparametric alternative (see Appendix F).

14.13.2.4  Data from an algal growth test with cadmium chloride will be used
to illustrate the statistical analysis.  The cell counts were Iog10
transformed in an effort to stabilize the variance for the ANOVA analysis.
The raw data, Iog10 transformed  data,  mean  and  standard  deviation  of the
observations at each concentration including the control are listed in
Table 4.  A plot of the Iog10 transformed cell  counts  for each  treatment  is
provided in Figure 4.

14.13.2.5  Test for Normality

14.13.2.5.1  The first step of the test for normality is to center the
observations by subtracting the mean of all the observations within a
concentration from each observation in that concentration.  The centered
observations are summarized in Table 5.

            TABLE 5.  CENTERED OBSERVATIONS FOR SHAPIRO-WILK'S EXAMPLE
Toxicant Concentration
                                                                Cd/L)
         Replicate    Control
     10
20
40
80
A
B '
C
-0.012
-0.022
0.033
0.004
-0.006
0.001
0.041
-0.078
0.036
0.050
-0.024
-0.027
-0.042
0.021
0.021
0.047
-0.044
-0.003
                                      214

-------
*  ESS
5  oc-piij
2  oEjj-
=  ^5
UJ
p  2$
<ป
8
   i:oc>;
M  SSSS
uj  uJ2iE
z  E>z
^^  ^~ -^ /n
   S
                                                                 ^^   to

                                                                 i   -i
                                                                 P   Is
                                                             -O
                                                                     O) ^j

                                                                       o>

                                                                     Q)
                                                                     P~ ^^
                                                                     ฃ ^

                                                                     **" o.
                                                                     c ">
                                                                     22
                                                                      03
                                                                       01
                                                                     0 S-
                                                                     i— m
                                                                     CL. (J
                                :3SNOdS3d H1MOU0
                                                                     
-------
14.13.2.5.2  Calculate the denominator,  D,  of the test statistic:
                                D =
Where:  X,-
                 the ith centered observation

            X  - the overall mean of the centered observations

            n  = the total number of centered observations.

For this set of data,      n = 18

                           x = — (o.ooo)  =  o.ooo
                               18


                           D = 0.0214

14.13.2.5.3  Order the centered observations from smallest to largest:

                  X<1) is the ith ordered observation.  These ordered observations are
listed in Table 6.

     TABLE 6.  ORDERED CENTERED OBSERVATIONS FOR SHAPIRO-WILK'S EXAMPLE
i
1
2
3
4
5
6
7
8
9
X(i)
-0.078
-0.044
-0.042
-0.027
-0.024
-0.022
-0.012
-0.006
-0.003
i
10
11
12
13
14
15
16
17
18
X(i>
0.001
0.004
0.021
0.021
0.033
0.036
0.041
0.047
0.050
14.13.2.5.4  From Table 4, Appendix B, for the number of observations,  n,
obtain the coefficients a,,  a2,  ..., ak where k is n/2 if n is even and
(n-l)/2 if n is odd.  For the data in this example,  n = 18, k = 9.  The a,-
values are listed in Table 7.
                                      216

-------
14.13.2.5.5  Compute the test statistic, W, as follows:
                                    ,
                             D 1=1
The differences x(n"i+1) - X(i) are listed in Table 7.

     TABLE 7.  COEFFICIENTS AND DIFFERENCES FOR SHAPIRO-MILK'S EXAMPLE


                                         XC i)               '
                                           -
1
2
3
4
5
6
7
8
9
0.4886
0.3253
0.2553
0.2027
0.1587
0.1197
0.0837
0.0496
0.0163
0.128
0.091
0.083
0.063
0.057
0.043
0.033
0.010
0.004
X(18)
XC17)
x(16)
X(15)
x<14)
x(13>
X(12)
x(
X(10)
- X(1>
T A
- X(3)
- X(4)
- x(5)
- xฃ

- x(8)
- x(9)
For this set of data:
                        W =
                            0.0214
(0.1436)2 = 0.964
14.13.2.5.6  The decision rule for this test is to compare W with the critical
value found in Table 6, Appendix B.  If the computed W is less than the
critical value, conclude that the data are not normally distributed.  For this
example, the critical value at a significance level of 0.01 and 18
observations (n) is 0.858.  Since W = 0.964 is greater than the critical
value, the conclusion of the test is that the data are normally distributed.

14.13.2.6  Test for Homogeneity of Variance

14.13.2.6.1  The test used to examine whether the variation in mean cell count
is the same across all toxicant concentrations including the control, is
Bartlett's Test (Snedecor and Cochran, 1980).  The test statistic is as
follows:
            B =
                                      217

-------
   Where:  V,-   ป   degrees  of  freedom  for  each toxicant concentration and
                 control,  V,- = (n,- -  1)

           p   =   number of levels  of toxicant  concentration  including the
                 control

          nf   -   the number  of replicates  for  concentration  i

          In   =   loge

           i   =   1,  2,  ...,  p, where p is  the  number of  concentrations
                 including the control
              -   i=i
14.13.2.6.2  For the data in this example,  (see Table 4)  all  toxicant
concentrations including the control  have the same number of replicates
(n{ - 3  for all  i).   Thus,  V, = 2 for all i.


14.13.2.6.3  Bartlett's statistic is therefore:
       B =  [(12) in (0.0018) -2   112(5-1)] /1. 194
                                i-l
         =  [12(-6.3200) - 2(-41.9082)]/1.194

         =  7.9764/1.194

         =  6.6804

14.13.2.6.4  B is approximately distributed as chi-square with p - 1 degrees
of freedom, when the variances are in fact the same.  Therefore, the
appropriate critical value for this test, at a significance level of 0.01 with
five degrees of freedom, is 15.09.  Since B = 6.6804 is less than the critical
value of 15.09, conclude that the variances are not different.
                                      218

-------
14.13.2.7  Dunnett's Procedure
14.13.2.7.1  To obtain an estimate of the  pooled  variance for Dunnett's
Procedure, construct an ANOVA table as described  in Table 8.
                          TABLE 8.  ANOVA  TABLE
Source df
Between p - 1
Within N - p
Total N - 1
Sum of Squares
(SS)
SSB
SSW
SST
Mean Square (MS)
(SS/df)
S* = SSE!/(p-l)
Sj = SSW/(N-p)
i
Where:      p  = number of toxicant  concentrations  including the control
            N  = total  number of observations n1 + n2 ...  4- n
            n,-  = number of observations in concentration I
                                                         i
          SSB  = 'L,Tl/ni-G2/N     Between Sum of  Squares;
          SST = E TYlj-Gz/N      Total Sum of Squares

          SSW = SST-SSB           Within Sum of Squares

            6  = the  grand  total  of  all  sample observations,  G = ฃ TV
            TJ = the total of the replicate measurements for
                 concentration  i
            ,-J = the jth observation for concentration  i  (represents
                 the cell  count for toxicant concentration i in test
                 chamber j)
                                     219

-------
14.13.2.7.2  For the data in this example:
                         = n2 = n3 =  n4 = n5 = n6 = 3
                           18
                      TI = Y,, +  Y12 + Y13  =  9.281
                      _i   ..n    ..12   yซ  =  g>23g
    =  2     22    23
    = Y3  + Y32 + Y33
       '  + Y* + Y   = 7.928
        S
    - Y61 + Y62 +  Y63
               +  Y 3 = 6.438
                                           8.627
                                           7.928
                                           6.438
                                           4.938
  6  =

SSB =
                           T, + T2 + T3 + T4 + T5 + T6 = 46.451
                         = _i_(374.606)  -  (46.451)2  = 4.997
                            3                  18
                     SST =
                           i=1:7=1
                         =  124.890   -  (46.451T  =  5.018
                                           18
                     SSW = SST-SSB   = 5.018 - 4.997 = 0.0210
                      S2B  = SSB/(p-l) = 4.996/(6-l)  -  0.9990
                           SSW/(N-p) =  0.021/(18-6)  =  0.0018
14.13.2.7.3  Summarize  these calculations in the ANOVA table (Table 9)
            TABLE  9.   ANOVA TABLE FOR DUNNETT'S PROCEDURE EXAMPLE
      Source
       Total
 df
17
Sum of Squares
     (SS)
   5.017
Mean Square(MS)
    (SS/df)
Between
Within
5
12
4.997
0.021
0.999
0.0018
                                       220

-------
                                                                                         1
 14.13.2.7.4   To  perform the  individual  comparisons,  calculate the t statistic
 for  each  concentration,  and  control  combination  as  follows:
                            t  -
                            L  ~
                                            + (1/nฑ)
 Where:   Y,-  = mean cell count for toxicant concentration i

         Y1  = mean cell count for the control

         SH  = square root of the within mean square

         n.,  = number of replicates for the control

         n,-  = number of replicates for concentration i.

 14.13.2.7.5  Table 10  includes the calculated t values  for each concentration
 and control combination.  In this example, comparing the 5 /*g/L concentration
 with the control the calculation is as follows:
                    ta -
                              (3.094-3.080)
                          [0.0424^(1/3) +(1/3) ]
                      TABLE 10.  CALCULATED T VALUES
Toxicant Concentration
(M9 Cd/L)
5
10
20
40
80
i

2
3
4
5
6
tf'

0.405
6.300
13.035
27.399
41.850
14.13.2.7.6  Since the purpose of this test is to detect a significant
reduction in mean cell count, a one-sided test is appropriate.  The critical
value for this one-sided test is found in Table 5, Appendix C.  For an overall
alpha level of 0.05, 12 degrees of freedom for error and five concentrations
(excluding the control) the critical value is 2.50.  The mean count for
concentration i is considered significantly less than the mean count for the
control  if t, is  greater than the  critical  value.   Since t,,  t,, t, and t,  are
greater than 2.50, the 10, 20, 40 and 80 Mg/L concentrations have
significantly lower mean cell counts than the control.   Hence the NOEC and the
LOEC for the test are 5 /j.g/1 and 10 yttg/L, respectively.

                                      221                  '

-------
14.13.2.7.7  To quantify the sensitivity of the test, the minimum significant
difference (MSD) that can be statistically detected may be calculated.
                          MSD = d S^CL/n^ + (1/12)
Where:  d  = the critical value for Dunnett's Procedure

        Sw - the square root of the within mean square

        n  - the common number of replicates at each concentration
             (this assumes equal replication at each concentration)

        n! - the number of replicates in the control.

14.13.2.7.8  In this example:
                     MSD = 2. 50 (0.0424) Vd/3)+ (1/3)

                          = 2.50 (0.0424)(0.8165)

                          = 0.086

14.13.2.7.9  The MSD (0.086) is  in transformed units.  An approximate MSD in
terms of cell count per 100 ml may be calculated via the following
conversion.

    1. Subtract the MSD from the transformed control mean.

          3.094 - 0.086 = 3.008

   2. Obtain the untransformed values for the control mean and the difference
      calculated in 1.

          10(3.094>

          10C3.008)

   3. The untransformed MSD  (MSD ) is determined by subtracting the
      untransformed values from  2.

        MSUU = 1241.6 - 1018.6 = 223

14.13.2.7.10  Therefore, for this set of data, the minimum difference in mean
cell count between the control and any  toxicant concentration that can be
detected as statistically significant is 223.

14.13.2.7.11  This represents  a  decrease in growth of 18% from the control.


14.13.2.8  Calculation of the  ICp


                                      222

-------
                                                                                         1
 14.13.2.8.1  The growth data in Table 4 are utilized in this example.
 Table  11 contains the means for each toxicant concentration.  As can be seen,
 the observed means are monotonically non-increasing with respect to
 concentration.  Therefore, it is not necessary to smooth the means prior to
 calculating the ICp.  See Figure 5 for a plot of the response curve.

           TABLE 11.  ALGAL MEAN GROWTH RESPONSE AFTER SMOOTHING
Toxicant
Cone.
(/*ง Cd/L)
Control
5
10
20
40
80


i
1
2
3
4
5
6
Response
means,Y-
(cells/mL)
1243
1201
757
441
140
44
Smoothed
mean, M-
( cells/ml)
1243
1201
757
441
140
44
14.13.2.8.2  An IC25 and IC50 can be estimated  using  the  Linear  Interpolation
Method (Appendix M).  A 25% reduction in  cell count,  compared to the controls
would result in a mean count of 932 cells,  where  M^l-p/100) = 1243(1-25/100)
A 50% reduction in cell count,  compared to  the  controls,  would result in a
mean count of 622 cells.  Examining the means and their associated
concentrations (Table 11),  the  response,  932 cells, is bracketed by C, = 5 Mg
Cd/L and C3 = 10  /*g  Cd/L.   The  response,  622 cells, is bracketed by C, = 10  aq
Cd/L and C4 = 20  //g  Cd/L.                                            3

14.13.2.8.3  Using the equation from section 4.2  of Appendix M, the estimate
of the IC25 is calculated  as follows:
     ICp =
    IC25 = 5+[1243 (1-25/100)-1201]	(10-5)
                                        (757-1201)

          = 8 /*g Cd/L.

14.13.2.8.4  The IC50 estimate  is  14 /^g Cd/L:

     IC25 = 6.25+[28.75(1-25/100)-28.75]
                                               (9.40-28.75)
      IC50 = 10+ [1243 (1-50/100) -757]
                                                 __
                                          (441-757)
              14 Mg Cd/L.
                                     223

-------
   LU


   I
LU

0
a. LU
LU X
CE f=
Q LU


II
                                                                               •t-J

                                                                               o
                                                                               S-
                                                                               cn

                                                                               o
                                                                               o
                                                                        Q.
                                                                        fO
                                                                        O

                                                                        s
                                                               TO
                                                               rt



                                                               O
                                                                      LU
                                                                      O

                                                                      O
                                                                      o
                                                                       o
                                                                       ง
                                                                 -o
88888888ฐ^
        CM
                      o>  oo
                                    
                                                                        0)
                                                                        s-
                                                                        cn

                                                                        OJ
                                                                        O)


                                                                        •a
                                                                        
                                                                        5-
                                                                        0)
                                                                        00
                                                                               -a
                      lNnO0113ONV3W
S- •

<4-
O


O -f->

ol -a




IT)

CD
s_
                                  224

-------
                                                                                          1
                                                          I
 14.13.2.8.5  When the  ICPIN program was  used  to  analyze  this  set  of  data,
 requesting 80 resamples, the estimate of the  IC25 was  8.0227  /*g Cd/L.  The
 empirical 95% confidence interval for the true mean was  6.4087 ปg Cd/L and
 10.0313 /zg Cd/L.  The  ICPIN computer program  output for  the IC25  for this data
 set is shown in Figure 6.
             .
 14.13.2.8.6  When the  ICPIN program was  used  to  analyze  this  set  of  data,
 requesting 80 resamples, the estimate of the  IC50 was  14.2774 /*g  Cd/L.  The
 empirical 95% confidence interval for the true mean was  9.7456 /ig Cd/L and
 18.5413 ng Cd/L.  The computer program output for the  IC50 for this  data set
 is shown in Figure 7.

 14.13.3  BIOSTIMULATION

 14.13.3.1  Where the growth response in  effluent (or surface water)  exceeds
growth in the control flasks, the percent stimulation, S(%),  is calculated as
 shown below.  Values which are significantly greater than the control  indicate
 a possible degrading enrichment effect on the receiving  water (Walsh et al.,
 1980)i
                                      T-C
                                           x 100
   Where:   T  =  Mean effluent or surface water response

            C  =  Mean control response

14.14  PRECISION AND ACCURACY

14.14.1  PRECISION

14.14.1.1  Single-Laboratory Precison

14.14.1.1.1  Data from repetitive 96-h toxicity tests conducted with cadmium
chloride as the reference toxicant, using medium containing EDTA, are shown in
Table 12.  The precision (CV) of the 10 EC50s was 10.2%.
                                                          i
14.14.1.2  MultiLaboratory Precision

14.14.1.2.1  Data on the multilaboratory precision of this test are not yet
available.

14.14.2  ACCURACY

14.14.2.1  The accuracy of toxicity tests cannot be determined.
                                     225

-------
Cone. ID

Cone. Tested
Response
Response
Response
1
2
3
1
0
1209
1180
1340
2
5
1212
1186
1204
3
10
826
628
816
4
20
493
416
413
5
40
127
147
147
6
80
49.3
40.0
44.0
*** Inhibition Concentration Percentage Estimate ***
Toxicant/Effluent: Cadmium
Test Start Date: Example   Test Ending Date:
Test Species: Selenastrum capricornutum
Test Duration:            96 h
DATA FILE: scmanual.icp
OUTPUT FILE: scmanual.i25
Cone.
ID
i
2
3
4
5
6
Number
Replicates
3
3
3
3
3
3
Concentration
ug/i
0.000
5.000
10.000
20.000
40.000
80.000
Response
Means
1243.000
1200.667
756.667
440.667
140.333
44.433
Std.
Dev.
85.247
13.317
111.541
45.347
11.547
4.665
Pooled
Response Means
1243.000
1200.667
756.667
440.667
140.333
44.433
The Linear Interpolation Estimate:     8.0227   Entered P Value: 25

Number of Resamplings:   80
The Bootstrap Estimates Mean:
Original Confidence Limits:
Expanded Confidence Limits:
Resampling time in Seconds:
  8.1627 Standard Deviation:
Lower:     7.2541 Upper:
Lower:     6.4087 Upper:    I?:0,313
  1.65  Random. Seed: -1575623987
                                                                ฐ-4733
                Figure  6.   ICPIN  program  output  for  the IC25.
                                     226

-------
                                                                                        1
Cone. ID

Cone. Tested
Response
Response
Response
1
2
3
1
0
1209
1180
1340
2
5
1212
1186
1204
3
10
826
628
816
4
20
493
416
413
5
40
127
147
147
6
80
49.3
40.0
44.0
*** Inhibition Concentration Percentage Estimate ***
Toxicant/Effluent: Cadmium
Test Start Date: Example   Test Ending Date:
Test Species: Selenastrum capricornutum
Test Duration:            96 h
DATA FILE: scmanual.icp
OUTPUT FILE: scmanual.iBO
Cone.
ID
1
2
3
4
5
6
Number
Replicates
3
3
3
3
3
3
Concentration
ug/1
0.000
5.000
10.000
20.000
40.000
80.000
Response
Means
1243.000
1200.667
756.667
440.667
140.333
44.433
Std.
Dev.
85.247
13.317
111.541
45.347
11.547
4.665
Pooled
Response Means
1243.000
1200.667
756.667
440.667
140.333
44.433
The Linear Interpolation Estimate:    14.2774   Entered P Value: 50

Number of Resamplings:   80
The Bootstrap Estimates Mean:  14.2057 Standard Deviation:     1.1926
Original Confidence Limits:   Lower:    12.1194 Upper:    16.3078
Expanded Confidence Limits:   Lower:     9.7456 Upper:  I  18.5413
Resampling time in Seconds:     1.65  Random Seed: -1751550803
               Figure  7.   ICPIN  program  output  for  the  IC50.
                                    227

-------
TABLE 12.  SINGLE LABORATORY PRECISION OF THE GREEN ALGA, SELENASTRUM
           CAPRICORNUTUM, 96-H TOXICITY TESTS, USING THE REFERENCE
           TOXICANT CADMIUM CHLORIDE (USEPA, 1991a)
Test Number
1
2
3
4
5
6
7
8
9
10
n
Mean
CV (%)
EC50 (mg/L)
2.3
2.4
2.3
2.8
2.6
2.1
2.1
2.1
2.6
2.4
10.0
2.37
10.2
                                   228

-------
                                                                                            1
                                 CITED REFERENCES


AOAC.  1990.  Agricultural chemicals; contaminants; drugs.  Vol. 1, Official
     methods of analysis.  15th edition.  Association of Official Analytical
     Chemists, Arlington, VA.

APHA.  1992.  In: Standard methods for the examination of water and
     wastewater. 18th edition.  American Public Health Association,
     Washington, DC.

ASTM.  1993.  Practice for using brine shrimp nauplii as food for test
     animals in aquatic toxicology.  Annual Book of ASTM Standards. Vol.
     11.04.  ASTM E1203-92. .American Society for Testing and Materials,
     Philadelphia, PA.

Beckett, D.C. and P.A. Lewis.  1982.  An efficient procedure for slide
     mounting of larval chironomids.  Trans. Amer. Fish. Soc. 101(1):96-99.

Benoit, D.A., F.A. Puglisi, and D.L. Olson.  1982.  A fathead minnow,
     Pimephales promelas, early life stage toxicity test method evaluation
     and exposure to four organic chemicals.  Environ. Pollut. (Series A)
     28:189-197.

Berner, D.D.  1986.  The taxonomy of Cen'odaphnia (Crustacea:Cladocera) in
     U. S. Environmental Protection Agency Cultures.  Environmental
     Monitoring and Support Laboratory, U. S. Environmental Protection
     Agency, Cincinnati, OH 45268. EPA/600/4-86/032.

Birge, W.J., J.A. Black, and B.A. Ramey.  1981.  The reproductive toxicology
     of aquatic contaminants.  In: Hazard Assessments of Chemicals, Current
     Developments, Vol. 1.  Academic Press, Inc.  pp. 59-114.

Birge, W.J. and R.A. Cassidy.  1983.  Importance of structure-activity
     relationships in aquatic toxicology.  Fundam. Appl. Toxicol. 3:359-368.

Birge, W.J., J.A. Black, and A.G. Westerman.  1985.  Short-term fish and
     amphibian embryo-larval tests for determining the effects of toxicant
     stress on early life stages and estimating chronic values for single
     compounds and complex effluents.  Environ. Toxicol. Chem. 4:807-821.

Birge, W.J., J.A. Black, T.M. Short, and A.G. Westerman.  1989.   A
     comparative ecological and toxicological investigation of an STP
     effluent  and its receiving stream.  Environ. Toxicol. Chem.
     8(5):437-450.

Brungs, W.A.  1971.  Chronic effects of low dissolved oxygen concentrations
     on fathead minnows (Pimephales promelas).   J. Fish. Res. Board Can.
     28:1119-1123.
                                        229

-------
Collins, M.T., J.B. Gratzer, D.L. Dawe, and T.G. Nemetz.  1976.  Effects of
     antibacterial agents on nitrification in aquatic recirculating systems.
     J. Fish. Res. Board Can.  33.215-218.

DeGraeve, G.M., J.D. Cooney, T.L. Pollock, N.G. Reichenbach, J.H. Dean, M.D.
     Marcus, and D.O. Mclntyre.  1988.  Fathead minnow 7-day test: round
     robin study.  Intra- and inter!aboratory study to determine the
     reproducibility of the seven-day fathead minnow larval survival and
     growth test.  Columbus, Battelle, Columbus, OH.  Available from:
     American Petroleum Institute, 1220 L Street, NW, Washington, DC  20005,
     Report No. 4486.

DeGraeve, G.M., J.D. Cooney, B.H. Marsh, T.L. Pollock, and N.G. Reichenbach.
     1989.  Precision of the EPA seven-day Ceriodaphnia dubia test: Intra-
     and Interlaboratory study to determine the reproducibility of the
     seven-day Ceriodaphnia dubia survival and reproduction tests.  EPRI EN-
     6469. Electric Power Institute, Palo Alto, CA 94369.

DeGraeve, G.M., J.D. Cooney, D.O. Mclntyre, T.L. Pollock, N.G. Reichenbach,
     J.H. Dean, and M.D. Marcus.  1991.  Variability in the performance of
     the 7-day fathead minnow (Pimephales promelas) larval survival and
     growth test:  An intra- and interlaboratory study.  Environ. Toxicol.
     Chem. 10:1189-1201.

DeWoskin, R.S.  1984.  Good laboratory practice regulations: A comparison.
     Research Triangle Institute, Research Triangle Park, NC. 63 pp.

Dunnett, C.W.  1955.  Multiple comparison procedure for comparing several
     treatments with a control.  J. Amer. Statist. Assoc. 50:1096-1121.

Eagleson, K.W., D.L. Lenat, L. Ausley, and F. Winborne.  1990.  Comparison of
     measured in-stream biological responses with responses predicted by
     Ceriodaphnia chronic toxicity test.  Environ. Toxicol. Chem. 9:1019-
     1028.

Emerson, K., R.C. Russo, R.E. Lund, and R.V. Thurston.  1975.  Aqueous
     ammonia equilibrium calculations; effect of pH and temperature.  J. Fish
     Res. Bd. Can. 32(12):2380-2383.

FDA.  1978.  Good laboratory practices for nonclinical laboratory studies.
     Part 58, Fed. Reg. 43(247):60013-60020, December 22, 1978.

Finney, D.J.  1948.  The Fisher-Yates test of significance in 2X2 contingency
     tables.  Biometrika 35:145-156.

Finney, D.J.  1971.  Probit Analysis. Third Edition.  Cambridge Press, New
     York, NY.  668 pp.

Flickinger, S.A.  1966.  Determination of sexes in fathead minnow.  Trans.
     Am. Fish. Soc. 98(3):526-527.
                                        230

-------
Gast, M.H. and W.A. Brungs.  1973.  A procedure for separating eggs of the
     fathead minnow.  Prog. Fish-Cult. 35:54.

Hodges, J.L., Jr. and E.L. Lehmann.  1956.  The efficiency of some
     nonparametric competitors of the t-test.  Ann. Math. Statist.
     31:625-642.

Hoffman, G.L. and A.J. Mitchell.  1980.  Some chemicals that have been used
     for fish diseases and pests. Fish Farming  Exp. Station, Stuttgart, AR
     72160.  8 pp.                                         .    .

Jensen, A.L.  1972.  Standard error of LC50 and sample size in fish bioassay.
     Water Res. 6:85-89.

Leger, P., Sorgeloos, P., Millamena, O.M., and Simpson, K.L.  1985.
     "International Study on Artemia.  XXV.  Factors determining the
     nutritional effectiveness of Artemia:  The relative impact of
     chlorinated hydrocarbons and essential fatty acids in San Francisco Bay
     and San Pablo Bay Artemia."  J. Exp. Mar. Biol. Ecol., Vol.  93, 1985.
     pp. 71-82.

Leger, P., Bengtson, D.A., Simpson, K.L. and Sorgeloos, P.  1986.  "The use
     and nutritional value of Artemia as a food source," M. Barnes (ed.),
     Oceanography and Marine Biology Annual Review, Vol. 24, Aberdeen
     University Press, Aberdeen, Scotland, 1986, pp. 521-623.

Macek, K.J. and B.H. Sleight.  1977.  Utility of toxicity tests with embryos
     and fry of fish in evaluating hazards associated with the chronic
     toxicity of chemicals to fishes.  In: Mayer, F. L. and J. L. Hamelink
     (eds.).  Aquatic Toxicology and Hazard Evaluation, ASTM STP 634,
     American Society for Testing and Materials, Philadelphia, PA.   pp.
     137-146.

Marking, L.L. and V.K. Dawson.  1973. . Toxicity of quinaldine sulfate to
     fish.  Invest. Fish Contr. No. 48, U. S. Fish and Wildlife Service,
     Washington, DC.  8 pp.

McKim, J.M.  1977.  Evaluation of tests with the early life stages of fish
     for predicting long-term toxicity.  J. Fish. Res. Board Can.
     34:1148-1154.

Miller, R.G.  1981.  Simultaneous statistical inference.  Springer-Verlag,
     New York, NY.  299 pp.

Mount, D.I. and C.E. Stephan.  1967.  A method for establishing acceptable
     toxicant limits for fish - Malathion and 2,4-D.  Trans. Am.  Fish. Soc.
     96:185-193.

Mount, D.I.  1968.  Chronic toxicity of copper to fathead minnows (Pimephales
     promelas, Rafinesque).  Water Res. 2:215-223.
                                        231

-------
Mount, D.I. and T.J. Norberg.  1984.  A seven-day life-cycle cladoceran test.
     Environ. Toxicol. Chem.  3:425-434.

Mount, D.I., T.J. Norberg-King, R. Keen, and J.T. Taraldsen.  1987.  A
     reference test water for cladocerans.  Abstract, llth Annual Symposium,
     Aquatic Toxicology and Hazard Assessment.  American Society for Testing
     and Materials, May 10-12, 1987. Cincinnati, OH.

Neiheisel, T.M., W.B. Horning, II, B.M. Austern, D.F. Bishop, T.L. Reed, and
     J.F. Estenik.  1988.  Toxicity reduction at municipal wastewater
     treatment plants.  J. Wat. Pollut. Control Fed. 60(1):57-67.

Norberg, T.J. and D.I. Mount. 1983.  A seven-day larval growth test.
     Presented at the Annual Meeting, Society of Environmental Toxicology and
     Chemistry, November 6-9, 1983, Arlington, VA.

Norberg, T. J. and D. I. Mount.   1985.  A new fathead minnow (Pimephales
     promelas) subchronic toxicity test.  Environ. Toxicol. Chem.
     4(5):711-718.

Norberg-King, T.J.  1991.  Calculation of ICp values of IC15, IC20, IC25,
     IC30, and IC50 for Appendix  A of the revised Technical Support Document.
     Memorandum to M. Heber, Office of Water, U.S. Environmental Protection
     Agency, Washington, D.C".

Pickering, Q.H.  1988.  Evaluation and Comparison of two short-term fathead
     minnow tests for estimating  chronic toxicity.  Water Res.  22(7):883-
     893.

Pearson, E.S. and T.O. Hartley.   1962.  Biometrika tables for statisticians.
     Vol. 1.  Cambridge Univ. Press, England,  pp. 65-70.

Pennak, R.W.  1978.  Freshwater invertebrates of the United States.  Protozoa
     to Mollusca.  Third Edition.  John Wiley & Sons, Co., New York, NY.

Pennak, R.W.  1989.  Freshwater invertebrates of the United States.  Protozoa
     to Mollusca.  Third Edition.  John Wiley & Sons,  Co., New York, NY.

Rehnberg, B.6., D.A. Schultz, and R.L. Raschke.  1982.  Limitations of
     electronic counting in reference to algal assays.  J. Wat. Pollut.
     Control Fed. 54(2):181-186.

Snedecor, G.W. and W.G. Cochran.  1980.  Statistical Methods.  Seventh
     Edition.  Iowa State University Press, Ames, Iowa.  593 pp.

Soderberg, R.W.  1982.  Aeration  of water supplies for fish culture in
     flowing water.  Prog. Fish-Cult. 44(2):89-93.

Steel, R.G.D.  1959.  A multiple  comparison rank sum test: treatments versus
     control.  Biometrics 15:560-572.
                                        232

-------
Taylor J.K.  1987.  Quality Assurance of chemical measurements.  Lewis Publ.,
     Inc. Chelsea, MI.

Thurston, R.V., R.C. Russo, and K. Emerson.  1974.  Aqueous ammonia
     equilibrium calculations.  Tech. Rep. No. 741.  Fish Bioassay
     Laboratory, Montana     State Univ., Bozeman, MT.  18 pp.,

USDA.  1989.  Methods which detect multiple residues.  Vol.! 1.  Pesticide
     analysis manual.  U.S. Department of Health and Human Services,
     Washington, D.C.

USEPA.  1971.  Algal assay procedures: Bottle test.  Miller, W.E., J.C.
     Greene, and T. Shiroyama (eds.).  National Eutrophication Program.
     Environmental Research Laboratory, U. S. Environmental Protection
     Agency, Corvallis, OR 97333.  82 pp.

USEPA.  1973.  Biological field and laboratory methods for measuring the
     quality of surface waters and effluents.  Weber, C.I. (ed.).
     Environmental Monitoring Series.  Office of Research and Development,  U.
     S. Environmental Protection Agency, Cincinnati, OH 45268.,  EPA/670/4-
     73/001.

USEPA.  1975.  Methods for acute toxicity tests with fish,
     macroinvertebrates, and amphibians.  Environmental Research Laboratory,
     U. S. Environmental Protection Agency, Duluth, MN 55804.
     EPA-660/3-75-009.

USEPA.  1978a.  The Selenastrum capn'cornutum Printz algal assay bottle test.
     Miller, W.E., J.C. Greene,, and T. Shiroyama  (eds.).  Environmental
     Research Laboratory, U. S. Environmental Protection Agency, Con/all is,
     OR 97333.  EPA-600/9-78-018.

USEPA.  1978b  Methods for measuring the acute toxicity of effluents to
     aquatic organisms. Second edition.  Peltier, W. (ed.}ซ  Environmental
     Monitoring and Support Laboratory - Cincinnati.  U. S,, Environmental
     Protection Agency, Cincinnati, OH 45268. EPA-600/4-78--012.

USEPA.  1979a.  Handbook for analytical quality control in water and
     wastewater laboratories.  U. S. Environmental Protection Agency,
     Environmental Monitoring and Support Laboratory, Cincinnati, OH 45268.
     EPA-600/4-79-019.

USEPA.  1979b.  Methods for chemical analysis of water and wastes.
     Environmental Monitoring and Support Laboratory, U. S,, Environmental
     Protection Agency, Cincinnati, OH 45268.  EPA-600/4-79-020.  Revised
     March 1983.

USEPA.  1979c.  Interim NPDES compliance biomonitoring inspection manual.
     Office of Water Enforcement, U. S. Environmental Protection Agency,
     Washington, DC 20460. (MCD-62).
                                        233

-------
USEPA.  1979d.  Good laboratory practice standards for health effects.
     Paragraph 772.110-1, Part 772 - Standards for development of test data.
     Fed. Reg. 44:27362-27375, May 9, 1979.

USEPA.  1980a.  Appendix B - Guidelines for Deriving Water Quality Criteria
      for the Protection of Aquatic Life and Its Uses.  Federal Register,
     Vol. 45, No. 231, Friday, November 28, 1980.

USEPA.  1980b.  Proposed good laboratory practice guidelines for toxicity
     testing.  Paragraph 163.60-6.  Fed. Reg, 45:26377-26382, April 18, 1980.

USEPA.  1980c.  Physical, chemical, persistence, and ecological effects
     testing; good laboratory practice standards (proposed rule).  40 CFR
     772, Fed. Reg. 45:77353-77365, November 21, 1980.

USEPA.  1981.  In situ acute/chronic toxicological monitoring of industrial
     effluents for the NPDES biomonitoring program using fish and amphibian
     embryo/larval stages as test organisms.  Birge, W.J. and J.A. Black.
     OWEP-82-001.  Office of Water Enforcement and Permits, U. S.
     Environmental Protection Agency, Washington, DC 20460.

USEPA.  1982.  Methods for organic chemical analysis of municipal and
     industrial wastewater.  Environmental Monitoring and Support Laboratory,
     U. S. Environmental Protection Agency, Cincinnati, OH 45268.  EPA/600/4-
     82-057.

USEPA.  1983.  Guidelines and format for EMSL-Cincinnati methods.  Kopp, J.F.
     Environmental Monitoring and Support Laboratory, U. S. Environmental
     Protection Agency, Cincinnati, OH 45268.  EPA-600/8-83-020.

USEPA.  1984.  Effluent and ambient toxicity testing and instream community
     response on the Ottawa River, Lima, Ohio.  Mount, D.I., N.A. Thomas,
     T.J. Norberg, M. T. Barbour, T.H. Roush, and W.F. Brandes. (eds.).
     Environmental Research Laboratory, U. S. Environmental Protection
     Agency, Duluth, MN 55804.  EPA/600/3-84/080.

USEPA.  1985a   Methods for measuring the acute toxicity of effluents to
     freshwater and marine organisms.  Third edition.  Peltier, W. and C.I.
     Weber (eds.).  Environmental Monitoring and Support Laboratory, U. S.
     Environmental Protection Agency, Cincinnati, OH 45268.  EPA
     600/4-85-013.

USEPA.  1985b.  Validity of effluent and ambient toxicity testing for
     predicting biological impact on Five Mile Creek, Birmingham, Alabama.
     Mount, D.I., A.E. Steen, and T.J. Norberg-King. (eds.).  Environmental
     Research Laboratory, U. S. Environmental Protection Agency, Duluth, MN
     55804.  EPA/600/8-85/015.
                                        234

-------
 USEPA.   1985c.   Validity of effluent and  ambient toxicity tests for
      predicting  biological  impact,  Scippo Creek, Circleville,  Ohio.  Mount,
      D.I.  and T.J.  Norberg-King.  (eds.).   Environmental  Research Laboratory,
      U.  S.  Environmental  Protection Agency,  Duluth,  MN 55804.
      EPA/600/3-85/044.

 USEPA.   1985d.   Validity of effluent and  ambient toxicity tests for
      predicting  biological  impact,  Ohio River,  near  Wheeling,  West Virginia.
      Mount, D.I., A.E.  Steen,  and T.J. Norberg-King.  (eds.)t   Environmental
      Research Laboratory, U. S. Environmental  Protection Agency,  Duluth,  MN
      55804. EPA/600/3-85/071.

 USEPA.   1985e.   Short-term  methods  for estimating the  chronic  toxicity of
      effluents and  receiving waters to freshwater organisms.   Horning,  W.B.,
      II  and C.I. Weber (eds.).  Environmental  Monitoring and  Support
      Laboratory, U. S.  Environmental  Protection Agency,  Cincinnati,  OH 45268.
      EPA/600/4-85/014.

 USEPA.   1986a.   Validity  of effluent and  ambient toxicity tests for
      predicting  biological  impact,  Kanawha River,  Charleston,  West Virginia.
      Mount, D.I. and T.J. Norberg-King. (eds.).   Environmental  Research
      Laboratory, U. S.  Environmental  Protection Agency,  Duluth,  MN 55804.
      EPA/600/3-86/006.

 USEPA.   1986b.   Validity  of effluent  and  ambient toxicity tests for
      predicting  biological  impact,  Naugatuck River, Connecticut.   Mount,
      D.I., T.J.  Norberg-King,  and A.E. Steen.  (eds.).  Environmental  Research
      Laboratory, U. S.  Environmental  Protection Agency,  Duluth,  MN 55804.
      EPA/600/8-86/005.

 USEPA.   1986c.   Validity of effluent  and  ambient  toxicity tests  for
      predicting  biological  impact,  Back River,  Baltimore  Harbor, Maryland.
     Mount, D.I., A.E.  Steen,  and T.J. Norberg-King. (eds.).   Environmental
      Research Laboratory, U. S. Environmental Protection  Agency, Duluth, MN
      55804.  EPA/600/8-86/001.

USEPA.   1986d.   Validity of effluent  and  ambient  toxicity  tests for
     predicting  biological  impact,  Skeleton Creek, Enid, Oklahoma.
     Norberg-King,  T.J. and D.I. Mount, (eds.).   Environmental  Research
      Laboratory, U.  S.  Environmental  Protection Agency, Duluth, MN 55804
     EPA/600/8-86/002.

USEPA.   1986e.  Occupational health and safety manual.   Office of
     Administration, U. S. Environmental  Protection Agency, Washington,
     DC 20460.

USEPA.  1988a.  NPDES compliance inspection manual.  Office of Water
     Enforcement and Permits (EN-338), U.  S.  Environmental Protection Agency,
     Washington,  DC 20460.
                                       235

-------
USEPA.  1988b.  Short-term methods for estimating the chronic toxicity of
     effluents and receiving waters to marine and  estuarine organisms.
     Weber, C.I., W.B. Horning, D.J. Klemm, T.W. Neiheisel, P.A. Lewis, E.L.
     Robinson, J. Menkedick, and F. Kessler (eds.).  Environmental Monitoring
     Systems Laboratory, U. S. Environmental Protection Agency, Cincinnati,
     OH 45268.  EPA/600/4-87/028.

USEPA.  1988c.  Methods for aquatic toxicity identification evaluations:
     Phase I toxicity characterization procedures.  D.I. Mount  and L.
     Anderson-Carnahan.  U. S. Environmental Protection Agency, Environmental
     Research Laboratory, Duluth, MN 55804.  EPA-600/3-88/034.

USEPA.  1988d.  An interpolation estimate  for chronic toxicity: The  ICp
     approach.  Norberg-King, T. J.  Technical  Report 05-88, National
     Effluent Toxicity Assessment Center,  Environmental Research Laboratory,
     U. S. Environmental Protection Agency, Duluth, MN 55804.

USEPA.  1988e.  Multi-laboratory study of  Ceriodaphm'a chronic  toxicity test.
     Neiheisel, T.W., J.R. Menkedick, F.A.  Kessler, W.B. Horning,  and  C.I.
     Weber.   In House Report. Environmental Monitoring Systems  Laboratory,  U.
     S. Environmental Protection Agency, Cincinnati, OH 45268.

USEPA.  1989a.  Short-term methods  for estimating  the chronic toxicity of
     effluents and receiving waters to freshwater  organisms.  Second edition.
     Weber, C.I., W.H.  Peltier, T.J. Norberg-king, W.B. Horning,  II, F.A.
     Kessler, J.R. Menkedick, T.W.  Neiheisel, P.A. Lewis,  D.J.  Klemm,  Q.H.
     Pickering, E.L.  Robinson, J.M. Lazorchak,  L.J. Wymer,  and  R.W.  Freyberg
      (eds.).  Second  Edition.  Environmental Monitoring Systems Laboratory,
     U.S.  Environmental  Protection  Agency,  Cincinnati, OH  45268.   EPA/600/4-
     89/001.

USEPA.  1989b.  Methods  for  aquatic toxicity  identification evaluations:
      Phase II toxicity  identification procedures.  D.I. Mount  and  L.
     Anderson-Carnahan.  Environmental Research Laboratory,  U.  S.
      Environmental  Protection Agency, Duluth, MN  55804.   EPA/600/3-88/035.

USEPA.  1989c.  Methods  for  aquatic toxicity  identification evaluations:
      Phase III toxicity confirmation  procedures.   D.I. Mount.   Environmental
      Research Laboratory,  U.  S.  Environmental  Protection  Agency,  Duluth,  MN
      55804.   EPA/600/3-88/036.

USEPA.  1989d.   Toxicity reduction  evaluation  protocol  for municipal
      wastewater  treatment  plants.   Botts,  J.A., J.W.  Braswell,  J.  Zyman,  W.
      L. Goodfellow,  and S.B.  Moore  (eds.).  Reduction  Engineering Laboratory,
      U. S. Environmental  Protection Agency, Cincinnati,  OH 45268.   EPA/600/2-
      88/062.

USEPA.  1989e.   Generalized  methodology  for conducting  industrial  toxicity
      reduction  evaluations (TREs).   Fava,  J.A., D. Lindsay, W.H.  Clement,  R.
      Clark,  G.M.  DeGraeve,  J.D.  Cooney,  S. Hansen, W.  Rue, S.  Moore, and  P.
      Lankford.   Risk Reduction Engineering Laboratory,  U.  S. Environmental
      Protection  Agency,  Cincinnati, OH  45268.   EPA/600/2-88/070.

                                         236

-------
 USEPA.  1990.  Macro-invertebrate field and laboratory methods for evaluating
      the biological integrity of surface waters.  Klemm, D.J., P.A. Lewis, F.
      hulk, and J.M. Lazorchak.   Environmental Monitoring Systems Laboratory
      U.  S. Environmental Protection Agency, Cincinnati, OH 45268. EPA/600/4-
      •/U/ UOU *

 USEPA.  1991a.  Technical  support document for water quality-based toxic
      controls.  Office of Water Enforcement and Permits and Office of Water
      Regulations and  Standards, U. S. Environmental Protection Agency,
      Washington, DC 20460.   EPA/505/2-90/001.

 USEPA.  1991b.  Methods for aquatic toxicity identification evaluations:
      Phase I,  toxicity characterization procedures.   2nd ed.,   Norberg-King
      T.J., D.I.  Mount,  E.J.  Durhan, G.T.  Ankley, L.P.  Burkhard,  J.R.  Amato,
      M.T.  Lukasewycz,  M.K.  Schubauer-Berigan,  and L. Anderson-Carnahan
      (eds.).   Environmental  Research Laboratory, U.  S.  Environmental
      Protection   Agency, Duluth, MN 55804.   EPA-600/6-91/003.

 USEPA.   1991c.   Manual  for  the evaluation of laboratories performing  aquatic
      toxicity tests.   Klemm,  D.J.,  L.B.  Lobring, and W.H.  Horning,  II.
      Environmental  Monitoring and Support Laboratory,  U.  S.  Environmental
      Protection  Agency,  Cincinnati,  OH 45268.   EPA/600/4-90/031.

 USEPA.   1992.  Toxicity identification evaluations:   characterization  of
      chronically toxic  effluents,  Phase  I:   Toxicity characterization
      procedures.  2nd  ed., T.J.  Norberg-King,  D.I. Mount,  J.R.  Amato,  D.A.
      Jensen,  and J.A.  Thompson.   Environmental  Research  Laboratory, U  S
      Environmental  Protection Agency,  Duluth,  MN 55804.   EPA/600/6-91/005F.

 USEPA.   1993a.   Short-term methods  for estimating the chronic  toxicity of
      effluents and  receiving  waters  to marine  and estuarine  organisms
      (Second Edition).   Klemm, D.J.  and G.E. Morrison  (eds.).  Environmental
      Monitoring  Systems  Laboratory,  U. S. Environmental  Protection Agency,
      Cincinnati,  OH 45268.  EPA/600/4-91/003.

 USEPA.   1993b.   Methods  for measuring  the acute  toxicity of  effluents  to
      freshwater  and marine organisms.  Weber,  C.I. (ed.).  Environmental
     Monitoring  Systems  Laboratory,  U. S. Environmental Protection Agency
     Cincinnati,  OH 45268.  EPA/600/4-90/027F.

USEPA.   1993c.   Fish field and laboratory methods for evaluating the
     biological  integrity of  surface waters.  Klemm,  D.J.,  Q.j*. Stober, and
     J.M.  Lazorchak.  Environmental Monitoring Systems Laboratory, U S
     Environmental Protection Agency,  Cincinnati, OH 45268. EPA/600/R-92/H1.

Vanhaecke,  P.  and Sorgeloos, P.  1980.  "International Study on Artemia.  IV
     The biometrics  of Artemia strains from different geographical origin "
     In:   G.  Persoone, P.  Sorgeloos, 0. Roels and E. Jaspers (eds.),  The'
     Brine Shrimp Artemia,  Vol. 3, Ecology,  Culturing Use in Aquaculture,
     Universal Press, Wetteren, Belgium, pp. 393-405.
                                       237

-------
Vanhaecke, P., Steyaert, H. and Sorgeloos, P.  1980.  "International Study on
     Artemia.  III.  The use of Coulter Counter equipment for the biometrical
     analysis of Artemia cysts.  Methodology and Mathematics," In:  G.
     Persoone, P. Sorgeloos, 0. Roels, and E. Jaspers (eds.), The Brine
     Shrimp  Artemia, Vol. 1, Morphology, Genetics, Radiobiology, Toxicology.
     Universal Press, Wettersen, Belgium, 1980, pp. 107-115.

Walsh, G.E., L.H. Bahner, and W.B. Horning.  1980.  Toxicity of textile mill
     effluents to freshwater and estuarine algae, crustaceans, and fishes.
     Environ. Pollut. (Ser. A.) 21:169-179.

Walters, D.B. and C.W. Jameson.  1984.  Health and  safety for toxicity
     testing.  Butterworth Pub!., Woburn, MA.

Ward, G.S. and P.R.  Parrish.   1980.   Evaluation of  early life stages
     toxicity tests  with embryos and  juveniles of sheepshead minnows.
     Eaton, J.G., P.R.  Parrish, and A.C.  Hendricks  (eds.).  Aquatic
     Toxicology, ASTM STP 707, American Society for Testing and Materials,
     Philadelphia, PA.  pp.  143-247.

Weltering, D.M.  1984.  The  growth response  in fish chronic and early life
     stage toxicity  tests: A critical  review.  Aquat. Toxicol. 5:1-21.
In:
                                         238

-------
                                    BIBLIOGRAPHY


 Adams, N., K.H. Golding, and A.J. Dodds.   1985.  Toxicity of eight
      Wa!ur:S0luble Or9anic chemicals to Selenastrum capricornutum: A study of
      methods for calculating toxic values using different growth parameters
      Arch. Environ. Contam.  Toxicol. 14:333.

 Adams, N., K.H. Golding, and A.J. Dodds.   1986.  Effect of acetone on the
      toxicity of four chemicals to Selenastrum capricornutum.   Bull.  Environ
         Contam. Toxicol. 36:254-259.

 Akasenova, Y.I.,  G.I.  Bogucharskova, and  M.I.  Zozulina.   1969.   The role  of
      phytoplankton and bacterioplankton in  the food of the dominant Cladocera
      of the Lower Don.   Hydrobiologia 5(5):19-26.

 ASTM.   1993.   Guide for conducting early  life-stage toxicity tests with
      fishes.   Annual  Book of ASTM Standards. Vol.  11.04.   American Society  of

      Steals?" PMSljMa/P™ ^^  ^""  Sฐ^ fฐr  Test1"9  and

 Atkins,  W.R.G.   1923.   The phosphate content of fresh  and  salt waters  in  its
                   to'the growth  of algal  Pla"kton.   J. Mar. Biol.  Assoc.
         inrซ
      lo : 119-150 .


 Brige, W.J., J.A. Black, A.G. Westerman, and B.A. Ramey.  1983.  Fish and
      amphibian embryos  - A model system for evaluating teratogenicity
      Fundam. Appl . Toxicol. 3: 237-242.


 Black, J.A., W.J. Birge, A.G. Westerman, and P.C. Francis.  :1983. Comparative
      a^aJrซ toxicolฐgy of aromatic hydrocarbons.  Fundam. Appl. Toxicol. 3:
      353-358.


 Blaise, C.R.  1986.  Micromethod for acute aquatic toxicity assessment using
      the green alga Selenastrum capricornutum.   Toxic. Assess. 1:377-385.

 Burton, G. A., Jr., D. Nimmo, D. Murphey, and F. Payne.  1987.  Stream profile
      determinations using microbial activity assays and Ceh'odaphnia
      Environ. Toxicol. Chem. 6:505-513.

 Burton, G. A., Jr., A. Drotar,  J.  Lazorchak, and L. Bahls.   1987
      Relationship of microbial activity and Ceriodaphnia responses  to mining
      impacts on the Clark Fork River,  Montana.   Arch.  Environ. Contam.
     Toxicol. 16:523-530.


Conit, R.J.   1972.  Phosphorus and  algal  growth in the Spokane River
     Northwest Sci.  46(3): 177-189.

Conover,  W.J   1980.   Practical  Nonparametric Statistics.  Second edition
     John Wiley and  Sons,  New  York, NY. pp. 466-467.
                                       239

-------
Cooney, J.D., G.M. DeGraeve, E.L. Moore, T.L. Pollock, and W.D. Palmer. 1986.
     Effects of food and water quality on the culturing and toxicity testing
     of Ceriodaphnia.  Draft Report, No. 526-P-5372, Battelle Laboratories,
     Columbus, OH.

Cowqill, U.M.  1987.  Critical analysis of factors affecting the sensitivity
     of zooplankton and the reproducibility of toxicity test results.  Water
     Res. 21(12):1453-1462.

Cowgill, U.M., K.I. Keating, and I.T. Tokahashi.  1985.  Fecundity and
     longevity of Ceriodaphnia dubia/affinis in relation to diet at two
     different temperatures.  J. Crust. Biol. 5(3):420-429.

Cowgill, U.M., D.P. Milazzo, and C.K. Meagher.  1988.  New diet for
     Ceriodaphnia dubia.   Bull.  Environ. Contam. Toxicol. 41:304-309.

Cowgill, U.M., I.T. Takahashi, and  S.L. Applegath.   1985.  A comparison of
     the effect of  four benchmark chemicals  on Daphnia magna and Cenodaphma
     dubia-affinis  tested  at two different temperatures.  Environ. Toxicol.
     Chem.   4:415-422.

DeGraeve, G.M. and  J.D. Cooney.  1987.  Ceriodaphnia: An  update on effluent.
     toxicity testing  and  research  needs.  Environ.  Toxicol. Chem.
     6:331-333.

Dixon,  W.J.  and F.J. Massey, Jr.   1983.   Introduction to  Statistical
     Analysis.  Fourth edition.  McGraw Hill,  New  York, NY.

Draper,  N.R. and  J.A.  John.   1981.   Influential observations and  outliers  in
     regression.   Technometrics  23:21-26.

Dunnett,  C.W.  1964.   New  table  for multiple comparisons  with  a  control.
      Biometrics 20:482.

 Efron, B.   1982.   The  Jackknife, the Bootstrap,  and other resampling plans.
      CBMS 38, Soc.  Industr.  Appl.  Math.,  Philadelphia,  PA.

 Eloranta,  V. and  0. Laitinen.   1982.  The usefulness of the Selenastrum
      capricornutum algal  assay to evaluate the toxic effects of pulp and
      paper mill  effluents  in lake water.   Vatten  38:317-331.

 Ferris, J.J., S.  Kobyashi, and N.L. Clesceri.  1974.  Growth of Selenastrum
      capricornutum in natural  waters augmented with detergent products in
      wastewaters.  Water Res. 8(12):1013-1020.

 Filip, D.S. and E.J. Middelbrooks.  1975.  Evaluation of sample preparation
      techniques for algal  bioassays.  Water Res.  9:581-585.

 Finney, D.J.  1978.  Statistical methods in biological assay.   3rd ed.
      Charles Griffin & Co. Ltd., London.  508 pp.
                                         240

-------
Finney, D.J.  1985.  The median lethal dose and its estimation.  Arch.
     Toxicology.  56:215-218.

Forsberg, C.G.  1972.  Algal assay procedure.  J. Wat. Pollut. Control Fed.
     44(8):1623-1628.

Forsberg, C.G. and A. Forsberg.  1972.  Algal growth potential test improves
     sewage settlement control.  Ambio 1(1):26-29.

Goldman, C.R., M.G. Tunzi, and R. Armstrong.  1969.  Carbon-14 uptake as a
     sensitive measure of the growth of algal cultures.  In: Middlebrooks, E.
     J., T. E. Maloney, C. F. Powers, and L. M. Kaack (eds.).  Proc. of the
     Eutroph. Bioassess. Workshop, 19-21 June, 1969.  U. S. Pacific
     Northwest Water Laboratory, Corvallis, OR 97333.  pp. 158-170.

Gophen, M.  1976.  Temperature dependence on food intake, ammonia excretion
     and respiration in Cen'odaphm'a reticulata (Jurine) (Lake Kinnere.t,
     Israel).  Freshwat. Biol. 6(5):451-455.

Gray, R.H., R.W. Hanf, D.D. Dauble, and J.R. Skalaski.  1982.  Chronic
     effects of a coal liquid on a freshwater alga, Selenastrum
     capn'cornutum.  Environ. Sci. Technol. 16(4):225-229.
                                                           i
Greene, J.C., W.E. Miller, T. Shiroyama, and E. Merwin.  1975.  Toxicity of
     zinc to the green alga Selenastrum capricornutum as a function of
     phosphorus or ionic strength.  In: Middlebrooks, E. J., T. E. Maloney,
     C. F. Powers, and L. M. Kaack (eds.).  Proc. of the Eutroph.
     Bioassessment Workshop, 19-21 June, 1969.  U. S. Pacific Northwest Water
     Laboratory, Corvallis, OR 97333.  pp. 28-43.  EPA-660/3-75-034.

Hamilton, M.A.  1986.  Statistical analysis of cladoceran reproductivity
     tests.  Environ. Toxicol. Chem. 5:205-212.

Hamilton, M.A., R.C. Russo, and R.V. Thurston.  1977.  Trimmed Spearman-
     Karber method for estimating median lethal concentrations.  Environ.
     Sci. Tech. 11(7):714-719.

Hedtke, S.F., C.W. West, K.N. Allen, T.J. Norberg-King, and D.I. Mount. 1986.
     Toxicity of pentachlorophenol to aquatic organisms under naturally
     varying and controlled environmental conditions.  Environ. Toxicol.
     Chem. 5:531-542.
                                                           i
Keller, A.E., R.J. Dutton, G. Bitton, and T.L. Crisman.  1988.  Chronic
     toxicity of hydrothol-191 to Cen'odaphm'a dubia at 25 and 15ฐC.  Bull.
     Environ. Contain. Toxicol. 41:233-240.

Knight, J.T. and W.T. Waller.  1987.  Incorporating Daphm'a magna into the
     seven-day Cen'odaphm'a effluent toxicity test method.  Environ. Toxicol.
     Chem. 6:635-645.                                      i
                                        241

-------
Lewis, M.A.  1986  Comparison of the effects of surfactants on freshwater
     phytoplankton communities in experimental enclosures and on algal
     population growth in the laboratory.  Environ. Toxicol. Chem.
     5:319-332.

Marcus, A.H. and A.P. Holtzman.  1988.  A robust statistical method for
     estimating effects concentrations in short-term fathead minnow toxicity
     tests.  Manuscript submitted to the Criteria and Standards Division, U.
     S. Environmental Protection Agency, by Battelle Washington Environmental
     Program Office, Washington, DC, June 1988, under EPA Contract No.
     69-03-3534.  39 pp.

Mayer, F.L., Jr., K.S. Mayer, and M.R. Ellersieck.  1986.  Relation of
     survival to other endpoints in chronic toxicity tests with fish.
     Environ. Toxicol. Chem. 5:737-748.

Mayes, M.A., H.C. Alexander, D.C. Hopkins, and P.B. Latavitis.  1986. Acute
     and chronic toxicity of ammonia to freshwater fish: A site-specific
     study.  Environ. Toxicol. Chem. 5:437-442.

McNaught, D.C. and D.I. Mount.  1985.  Appropriate durations and measures for
     Ceriodaphnia toxicity tests.  In: R. C. Bahner and D. J. Hansen
     (eds.).  Aquatic Toxicology and Hazard Assessment: Eighth Symposium.
     ASTM STP 891.  American Society for Testing and Materials, Philadelphia,
     PA.  pp. 375-381.

McPhee, C.  1961.  Bioassay of algal production in chemically altered waters.
     Limnol. Oceanogr. 6:416-422.

Norberg, T.J. and D.I. Mount.  1985.  Diets for Ceriodaphnia reticulata
     life-cycle tests.  In: Cardwell, R. D., R. Purdy, and R. C. Bahner
     (eds.).  Aquatic Toxicology and Hazard Assessment, Seventh Symposium,
     ASTM STP 854, American Society for Testing and Materials, PA.  pp.
     42-52.

Norberg-King, T.J.  1987.  Relative sensitivity of the fathead minnow
     (Pimephales promelas) seven day subchronic test to the early life stage
     and life cycle tests. M. S. Thesis, Montana State Univ, Bozeman, MT.

Norberg-King, T.J.  1989.  An evaluation of the fathead minnow seven-day
     subchronic test for estimating chronic toxicity.  Environ. Toxicol.
     Chem. 8:1075-1089.

Norberg-King, T.J. 1993.  A Linear Interpolation Method for Sublethal
     Toxicity:  The Inhibition Concentration (ICp) Approach. Version 2.0.
     National Effluent Toxicity Assessment Center Technical Report 03-93,
     Environmental Research Laboratory, Duluth, MN 55804.  June 1993.

Owsley, J.A. and D.E. McCauley.  1986.  Effect of extended sublethal exposure
     to sodium selenite on Ceriodaphnia affinis.  Bull. Environ. Contam.
     Toxicol. 36:876-880.
                                        242

-------
Parker, M.  1977.  The use of algal bioassays to predict the short and long
     term changes in algal standing crop which result from altered phosphorus
     and nitrogen loadings.  Water Res. 11:719-725.

Penak, R.W. 1978.  Freshwater Invertebrates of the United States.  Second
     Edition.  Ronald Press, New York, NY.  769 pp.

Pickering, Q.H.  1988.  Evaluation and comparison of two short-term fathead
     minnow tests for estimating chronic toxicity.  Water Res. 22(7):883-893.

Sachdev, D.R. and N.L. Clesceri.  1978.  Effects of organic fractions from
     secondary effluent on Selenastrum capn'cornutum (Kutz.).  J. Wat.
     Pol Tut. Control Fed. 52:1810-1820.

Scheffe, H.  1959.  The Analysis of Variance.  John Wiley and Sons, New York,
     NY.  477 pp.

Shapiro, S.S. and M.B. Wilk.  1965.  An analysis of variance test for
     normality (complete samples).  Biometrika 52:591-611. ;

Shuba, T. and R.R. Costa.  1982.  Development and growth of Ceriodaphm'a
     reticulata embryos.  Trans. Amer. Microsc. Soc. 9(3):429-435.

Skulberg, 0.  1964.  Algal problems related to the eutrophication of European
     water supplies, and a bioassay method to assess fertilizing influences
     of pollution on inland waters.  In: D. F. Jackson (ed.). Algae and man.
     Plenum Press, New York, NY.  pp. 262-299.

Steel, R.G.D. and J.H. Torrie.  1960.  Principles and procedures of
     statistics with special reference to biological sciences.  McGraw-Hill
     Pub!., New York, NY.

Stephens, M.A. 1974.  EOF statistics for goodness of fit and some
     comparisons.  J. Amer. Stat. Assoc.  (JASA) 69:730-373:

Takahashi, I.T., U.M. Cowgill, and P.G. Murphy.  1987.   Comparison of ethanol
     toxicity to Daphnia magna and Ceriodaphm'a dubia tested at two different
     temperatures: Static acute toxicity test results.   Bull. Environ.
     Contam. Toxicol. 39:229-236.

Tallqvist, T.  1973.  Algal assay procedure (Bottle Test) at the Norwegian
     Institute for Water Research.  In:  Algal Assays in Water Pollution
     Research:  Proceedings from a Nordic Symposium, Oslow.  October,  25-29,
     1972.  Nordforsk. Secretariat of Environm. Sci., Helsinki.  PublI.
     1973-2.  pp. 5-17.

Tarzwell, C.M.  1971.  Bioassays to determine allowable waste concentrations
     in the aquatic environment.  In: Cole, H.A. (Organizer).  A Discussion
     on Biological Effects of Pollution in the Sea.  Proc. Roy. Soc. Lond. B
     177(1048):279-285.
                                        243

-------
Thomas, R.E. and R.L. Smith.  1975.  Assessing treatment process efficiency
     with the algal assay test.  In:  Middlebrooks, E.J., I.E. Maloney, C.F.
     Powers, and L.M. Kaack (eds.).  Proc. of the Eutroph. Bioassessment
     Workshop, 19-21 June, 1969.  U. S. Pacific Northwest Water Laboratory,
     Con/all is, OR 97333.  pp. 244-248.  EPA-660/3-75-034.

Thurston, R.V., R.C. Russo, E.L. Meyn, and R.K. Zaidel.  1986.  Chronic
     toxicity of ammonia to fathead minnows.  Trans. Amer. Fish. Soc.
     115:196-207.

USEPA.  1985a.  Technical support document for water quality-based control.
     Office of Water, U. S. Environmental Protection Agency, Washington, DC
     20460.

USEPA.  1985b.  Ambient water quality criteria for ammonia - 1984.  Office of
     Water Regulations and Standards Criteria and Standards Division,
     Washington, DC. 20460.  EPA/440/5085-001.

USEPA.  1986.  Ambient water quality criteria for pentachlorophenol.
     Criteria and Standards Division, Office of Water  Regulations and
     Standards, U. S. Environmental Protection Agency, Washington, DC  20460.
           EPA/440/5-86/009.

USEPA,  1987a.  Ambient water quality criteria for zinc.  Criteria and
     Standards Division, Office of Water Regulations and Standards, U. S.
     Environmental Protection Agency, Washington, DC 20460.
     EPA/440/5-87/003.

USEPA.  1987b.  Permit writer's guide to water quality-based permitting for
     toxic pollutants.  Office of Water, U. S. Environmental Protection
     Agency, Washington, DC 20460.  EPA/440/4-87/005.

USEPA.  1988a.  Protocols for short-term toxicity screening of  hazardous
     waste sites.  Greene, J.C., C.L. Bartels,. W.J. Warren-Hicks, B.R.
     Parkhurst, 6.L. Linder, S. A. Peterson, and W.E.  Miller  (eds.).
     Environmental Research Laboratory, U. S. Environmental Protection
     Agency, Corvallis, OR 97333.  EPA/600/3-88/029.
USEPA.   1988b.
      153-164.
40 CFR Part 160 - Good laboratory practice standards,  pp.
Walsh, G.E.  and  S.V. Alexander.   1980.  A marine  algal  bioassay method:
      Results with  pesticides  and  industrial wastes.   Wat. Air  Soil  Pollut.
      13:45-55.

Weiss, C.M.  1975.   Field  investigation of the  algal  assay  procedure  on
      surface waters  of North  Carolina.  In: Middlebrooks, E.J., E.H.
      Falkenborg, and T.E.  Maloney (eds.).  Biostimulation and  Nutrient
      Assessment.   Ann  Arbor Science  Pub!., Ann  Arbor, MI.   pp. 93-126.
                                        244

-------
Weiss, C.M. and R.W. Helms.  1971.   Inter!aboratory precision test-  An
     eight-laboratory evaluation of  the Provisional Algal Assay Procedure
     Bottle Test.  National Eutrophication Research Program.  U  S
     Environmental Protection Agency, Con/all is, OR  97333.

Winner, R.W.  1988.  Evaluation of the relative sensitivities of 7-D Daphnia
     magna and Ceriodaphnia dubia toxicity tests for cadmium and sodium
     pentachlorophenate.  Environ. Toxicol. Chem. 7:153-159.

Wong, P T.S.,  Y.K. Chau, and P.L. Luxon.  1978.  Toxicity of a mixture of
     metals on freshwater algae.  J. Fish. Res. Board Can. 35:479-481.

Yoshioka,  Y.,  Y.  Ose, and T. Sato.  1986.   Correlation of the five test
     methods to assess chemical  toxicity and relation to physical  properties
     Ecotoxicol.  Environ. Safety 12:15.
                                       245

-------
                               APPENDICES
                                                                       Page
A.  Independence,  Randomization,  and Outliers  . . 	  248
    1.  Statistical  Independence   	  248
                                                                        y/\o
    2.  Randomization   	"ฐ
    3.  Outliers	  •  253
B.  Validating Normality and Homogeneity of Variance Assumptions  ...  254
    1.  Introduction	254
    2.  Tests for Normal Distribution of Data   	254
    3.  Test for Homogeneity of Variance	  •  266
    4.  Transformations of the Data   	268
C.  Dunnett's Procedure   	•  •  271
    1.  Manual Calculations   	•  •  271
    2.  Computer Calculations	277
D.  T test with the  Bonferroni Adjustment	283
E.  Steel's  Many-one Rank  Test  	  289
F.  Wilcoxon Rank Sum Test 	  293
G.  Fisher's Exact  Test with  the  Bonferroni  Adjustment	  299
H.  Single  Concentration Toxicity Test  -  Comparison  of  Control  with
    100% Effluent or Receiving Water 	  308
 I.  Probit  Analysis   ....-.'	312
J.  Spearman-Karber Method 	   315
 K.  Trimmed Spearman-Karber Method  	   32ฐ
 L.  Graphical Method  	   325
 M.   Linear Interpolation Method    	  	   329
     1.   General  Procedure    	• •   329
     2.   Data Summary and Plots	329
                                    246

-------
                         APPENDICES  (CONTINUED)
                                                                       Page
    3.  Monotonicity '....'	                  32g
    4.  Linear Interpolation Method    	!	330
    5.  Confidence  Intervals	    331
    6.  Manual  Calculations	      331
    7.  Computer  Calculations  	, .    	335
                                                       i
Cited References	i                 340
                                  247

-------
                                  APPENDIX A

                   INDEPENDENCE, RANDOMIZATION, AND OUTLIERS


1.  STATISTICAL INDEPENDENCE

1 1  Dunnett's Procedure and the t test with Bonferroni's adjustment are
parametric procedures based on the assumptions that (1) the observations
within treatments are independent and normally distributed, and (2) that the
variance of the observations is homogeneous across all toxicant concentrations
and the control.  Of the three possible departures from the assumptions, non-
normality, heterogeneity of variance, and lack of independence  those caused
by lack of independence are the most difficult to resolve  (see Scheffe, 1959).
For toxicity data, statistical independence means that given knowledge of the
true mean for a given concentration or control, knowledge  of the error in any
one actual observation would provide no information about  the error in any
other observation.  Lack of independence  is difficult to assess and difficult
to test for statistically.  It may also have serious effects on the true alpha
or beta level.  Therefore, it is of utmost  importance to be aware of the need
for statistical independence between observations and to be constantly
vigilant  in avoiding any patterned experimental procedure  that might
compromise independence.   One of the best ways to help ensure independence  is
to follow proper  randomization procedures throughout the test.

2.  RANDOMIZATION

2 1   Randomization of the  distribution of test organisms among test chambers
and the arrangement  of  treatments  and  replicate chambers is an important part
of conducting  a valid test.  The  purpose  of randomization  is to  avoid
 situations where  test organisms  are  placed  serially  into test chambers, or
where all replicates for  a test  concentration  are  located  adjacent  to  one
 another,  which could introduce  bias  into  the  test  results.

 2 2   An  example of randomization  of  the  distribution  of  test organisms among
 test  chambers,  and an  example  of randomization of arrangement  of treatments
 and  replicate chambers  are described using  the Fathead Minnow  Larval  Survival
 and  Growth  test.   For  the purpose of the example,  the test design is  as
 follows-   five effluent concentrations are tested in addition  to the  control.
 The  effluent concentrations are as follows:  6.25%,  12.5%, 25.0%, 50.0%,  and
 100.0%.   There are four replicate chambers per treatment.   Each  replicate
 chamber contains ten fish.

 2.3   RANDOMIZATION OF FISH TO REPLICATE CHAMBERS EXAMPLE

 2 3.1  Consider first the random assignment of the fish  to the replicate
 chambers   The first step is to label  each of the replicate chambers with  the
 control  or effluent concentration and the replicate number.  The next step is
 to assign each replicate chamber four double-digit numbers.  An example of
 this assignment  is provided in Table A.I.  Note that the  double digits 00 and
 97 through 99 were not used.

                                       248 '

-------
TABLE A
Assigned
01,
02,
03,
04,
05,
06,
07,
08,
09,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26,
27,
28,
29,
30,
31,
32,
33,
34,
35,
36,
37,
38,
39,
40,
41,
42,
43,
44,
45,
46,
47,
48,
.1.
RANDOM ASSIGNMENT OF FISH TO REPLICATE CHAMBERS EXAMPLE
ASSIGNED NUMBERS FOR EACH REPLICATE CHAMBER
Numbers
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96




6.
6.
6.
6.
12
12
12
12
25
25
25
25
50
50
50
50
100
100
100
100




25%
25%
25%
25%
.5%
.5%
.5%
.5%
.0%
.0%
.0%
.0%
.0%
.0%
.0%
.0%
.0%
.0%
.0%
.0%
Replicate Chamber
Control ,
Control ,
Control ,
Control ,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
2.3.2  The random numbers used to carry out the random assignment of fish to
replicate chambers are provided in Table A.2.  The third step is to choose a
starting position in Table A.2, and read the first double digit number.  The
first number read identifies the replicate chamber for the first fish taken
from the tank.  For the example, the first entry in row 2 was chosen as the
starting position.  The first number in this row is 37.  According to
Table A.I, this number corresponds to replicate chamber 1 of the 25.0%
effluent concentration.  Thus, the first fish taken from the tank is to be
placed in replicate chamber 1 of the 25.0% effluent concentration.

2.3.3  The next step is to read the double digit number to the right of the
first one.  The second number identifies the replicate chamber for the second
fish taken from the tank.  Continuing the example,  the second number read in
row 2 of Table A.2 is 54.  According to Table A.I,  this number corresponds to
replicate chamber 2 of the 6.25% effluent concentration.   Thus,  the second
fish taken from the tank is to be placed in replicate chamber 2  of the 6.25%
effluent concentration.
                                      249

-------
TABLE A.2.  TABLE OF RANDOM NUMBERS (Dixon and Massey, 1983)
10 09 73
37 54 20
08 42 26
99 01 90
12 80 79
66 06 57
31 06 01
85 26 97
63 57 33
73 79 64
98 52 01
11 80 50
83 45 29
88 68 54
99 59 46
65 48 11
80 12 43
74 35 09
69 91 62
09 89 32
91 49 91
80 33 69
44 10 48
12 55 07
63 60 64
61 19 69
15 47 44
94 55 72
42 48 11
23 52 37
04 49 35
00 54 99
35 96 31
59 80 80
46 05 88
32 17 90
69 23 46
19 56 54
45 15 51
94 86 43
98 08 62
33 18 51
80 95 10
79 75 24
18 63 33
74 02 94
54 17 84
11 66 44
48 32 47
69 07 49
25 33
48 05
89 53
25 29
99 70
47 17
08 05
76 02
21 35
57 53
77 67
54 31
96 34
02 00
73 48
76 74
56 35
98 17
68 03
05 05
45 23
45 98
19 49
37 42
93 29
04 46
52 66
85 73
62 13
83 17
24 94
76 54
53 07
83 91
52 36
05 97
14 06
14 30
49 38
19 94
48 26
62 32
04 06
91 40
25 37
39 02
56 11
98 83
79 28
41 38
76 52 01
64 89 47
19 64 50
09 37 67
80 15 73
34 07 27
45 57 18
02 05 16
05 32 54
03 52 96
14 90 56
39 80 82
06 28 89
86 50 75
87 51 76
17 46 85
17 72 70
77 40 27
66 25 22
14 22 56
68 47 92
26 94 03
85 15 74
11 10 00
16 50 53
26 45 74
95 27 07
67 89 75
97 34 40
73 20 88
75 24 63
64 05 18
26 89 80
45 42 72
01 39 09
87 37 92
20 11 74
01 75 87
19 47 60
36 16 81
45 24 02
41 94 15
96 38 27
71 96 12
98 14 50
77 55 73
80 99 33
52 07 98
31 24 96
87 63 79
35 86
42 96
93 03
07 15
61 47
68 50
24 06
56 92
70 48
47 78
86 07
77 32
80 83
84 01
49 69
09 50
80 15
72 14
91 48
85 14
76 86
68 58
79 54
20 40
44 84
77 74
99 53
43 87
87 21
98 37
38 24
81 59
93 45
68 42
22 86
52 41
52 04
53 79
72 46
08 51
84 04
09 49
07 74
82 96
65 71
22 70
71 43
48 27
47 10
19 76
34
24
23
38
64
36
35
68
90
35
22
50
13
36
91
58
45
43
36
46
46
70
32
12
40
51
59
54
16
68
45
96
33
83
77
05
15
40
43
34
44
89
20
69
31
97
05
59
02
35
67 35
80 52
20 90
31 13
03 23
69 73
30 34
66 57
55 35
80 83
-10 94
72 56
74 67
76 66
82 60
04 77
31 82
23 60
93 68
42 75
16 28
29 73
97 92
86 07
21 95
92 43
36 78
62 24
86 84
93 59
86 25
11 96
35 13
60 94
28 14
56 70
95 66
41 92
66 79
88 88
99 90
43 54
15 12
86 10
01 02
79 01
33 51
38 17
29 53
58 40
43 76
40 37
25 60
11 65
66 53
61 70
26 14
48 18
75 48
42 82
05 58
82 48
00 78
79 51
89 28
69 74
23 74
02 10
72 03
67 88
35 54
41 35
65 75
46 97
25 63
37 29
38 48
44 31
87 67
14 16
10 25
38 96
54 62
97 00
40 77
70 07
00 00
15 85
45 43
15 53
88 96
85 81
33 87
25 91
46 74
71 19
29 69
15 39
68 70
44 01
80 95
20 63
15 95
88 67
98 95
65 81
86 79
73 05
28 46
60 93
60 97
29 40
18 47
90 36
93 78
73 03
21 11
45 52
76 62
96 29
94 75
53 14
57 60
96 64
43 65
65 39
82 39
91 19
03 07
26 25
61 96
54 69
77 97
13 02
93 91
86 74
18 74
66 67
59 04
01 54
39 09
88 69
25 01
74 85
05 45
52 52
56 12
09 97
32 30
10 51
90 91
61 04
33 47
67 43
11 68
33 98
90 74
38 52
82 87
52 03
09 34
52 42
54 06
47 64
56 13
95 71
57 82
16 42
11 39
77 88
08 99
03 33
04 08
48 94
17 70
45 95
61 01
04 25
11 20
22 96
27 93
28 23
45 00
12 48
08 36
31 71
39 24
43 68
79 00
03 54
47 34
54 19
62 52
22 05
56 14
75 80
71 92
33 34
75 75
82 16
17
02
64
97
77
85
39
47
09
44
33
01
10
93
68
86
53
37
90
.22
23
40
81
39
82
93
18
92
59
63
35
91
24
92
47
57
23
06
33
56
07
94
98
39
27
21
55
40
46
15
39 29
00 82
35 08
04 43
12 27
11 19
23 40
18 62
83 49
35 27
50 50
52 77
68 71
29 60
23 47
40 21
14 38
96 28
94 40
54 38
37 08
42 05
22 22
28 70
07 20
42 58
33 21
92 92
25 70
05 52
65 33
23 28
90 10
78 56
70 61
85 39
97 11
84 96
20 82
05 01
35 44
37 54
94 62
00 38
77 93
80 81
36 04
88 46
15 02
01 84
27 49 45
29 16 65
03 36 06
62 76 59
17 68 33
92 91 70
30 97 32
38 85 79
12 56 24
38 84 35
07 39 98
56 78 51
17 78 17
91 10 62
83 41 13
81 65 44
55 37 63
60 26 55
05 64 18
21 45 98
92 00 48
08 23 41
20 64 13
72 58 15
73 17 90
26 05 27
15 94 66
74 59 73
14 66 70
28 25 62
71 24 72
72 95 29
33 93 33
52 01 06
74 29 41
41 18 38
89 63 38
28 52 07
66 95 41
45 11 76
13 18 80
87 30 43
46 11 71
75 95 79
89 19 36
45 17 48
09 03 24
12 33 56
00 99 94
87 69 38
                             250

-------
2.3.4  Continue in this fashion until all the fish have been randomly assigned
to a replicate chamber.   In order to fill each replicate chamber with ten
fish, the assigned numbers will be used more than once.  If a number is read
from the table that was not assigned to a replicate chamber, then  ignore it
and continue to the next  number.  If a replicate chamber becomes filled and a
number is read from the table that corresponds to it, then ignore  that value
and continue to the next  number.  The first ten random assignments of fish to
replicate chambers for the example are summarized in Table A.3.

   TABLE A.3.  EXAMPLE OF RANDOM ASSIGNMENT OF FIRST TEN FISH TO REPLICATE
               CHAMBERS
  Fish
Assignment
First
Second
Third
Fourth
Fifth
Sixth
Seventh
Eighth
Ninth
Tenth
fish
fish
fish
fish
fish
fish
fish
fish
fish
fish
taken
taken
taken
taken
taken
taken
taken
taken
taken
taken
from
from
from
from
from
from
from
from
from
from
tank
tank
tank
tank
tank
tank
tank
tank
tank
tank
25.
0%
6.25%
50.
100
0%
effluent,
effluent,
effluent
.0% effluent,
6.25%
25.
50.
100
50.
100
0%
0%
.0%
0%
.0%
effluent,
effluent,
effluent,
effluent,
effluent,
effluent,
repl
repl
repl
repl
repl
repl
repl
repl
repl
repl
icate
icate
icate
icate
icate
icate
icate
icate
icate
icate
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
chamber
1
2
4
4
1
4
1
3
2
4
2.3.5  Four double-digit numbers were assigned to each replicate chamber
(instead of one, two, or three double-digit numbers) in order to make
efficient use of the random number table (Table A.2).  To illustrate, consider
the assignment of only one double-digit number to each replicate chamber:  the
first column of assigned numbers in Table A.I.  Whenever the numbers 00 and 25
through 99 are read from Table A.2, they will be disregarded and the next
number will be read.

2.4  RANDOMIZATION OF REPLICATE CHAMBERS TO POSITIONS EXAMPLE

2.4.1  Next consider the random assignment of the 24 replicate chambers to
positions within the water bath (or equivalent).  Assume that the replicate
chambers are to be positioned in a four row by six column rectangular array.
The first step is to label the positions in the water bath.   Table A.4
provides an example layout.

2.4.2  The second step is to assign each of the 24 positions four double-digit
numbers.  An example of this assignment is provided in Table A.5.  Note that
the double digits 00 and 97 through 99 were not used.

2.4.3  The random numbers used to carry out the random assignment of replicate
chambers to positions are provided in Table A.2.  The third  step is to choose
a starting position in Table A.2, and read the first double-digit number.  The

                                      251

-------
  TABLE A.4.  RANDOM ASSIGNMENT OF REPLICATE CHAMBERS TO POSITIONS: EXAMPLE
              LABELLING THE POSITIONS WITHIN THE WATER BATH
1
7
13
19
2
8
14
20
3
9
15
21
4
10
16
22
5
11
17
23
6
12
18
24
    TABLE A.5.  RANDOM ASSIGNMENT OF REPLICATE CHAMBERS TO POSITIONS: EXAMPLE
                ASSIGNED NUMBERS FOR EACH  POSITION
Assigned Numbers
01, 25, 49, 73
02, 26, 50, 74
03, 27, 51, 75
04, 28, 52, 76
05, 29, 53, 77
06, 30, 54, 78
07, 31, 55, 79
08, 32, 56, 80
09, 33, 57, 81
10, 34, 58, 82
11, 35, 59, 83
12, 36, 60, 84
13, 37, 61, 85
14, 38, 62, 86
15, 39, 63, 87
16, 40, 64, 88
17, 41, 65, 89
18, 42, 66, 90
19, 43, 67, 91
20, 44, 68, 92
21, 45, 69, 93
22, 46, 70, 94
23, 47, 71, 95
24, 48, 72, 96
Position
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
first number read identifies the position for the first replicate chamber of
the control.  For the example, the first entry in row 10 of Table A.2 was
chosen as the starting position.  The first number in this row was 73.
According to Table A.5, this number corresponds to position 1.  Thus, the
first replicate chamber for the control will be placed in position 1.

                                      252

-------
2.4.4  The next step is to read the double-digit number to the right of the
first one.  The second number identifies the position for the second replicate
chamber of the control.  Continuing the example, the second number read in row
10 of Table A.2 is 79.  According to Table A.5, this number corresponds to
position 7.  Thus, the second replicate chamber for the control will be placed
in position 7.

2.4.5  Continue in this fashion until all the replicate chambers have been
assigned to a position.  The first four numbers read will identify the
positions for the control replicate chambers, the second four numbers read •
will identify the positions for the lowest effluent concentration replicate
chambers, and so on.  If a number is read from the table that was not assigned
to a position, then ignore that value and continue to the next number.  If a
number is repeated in Table A.2, then ignore the repeats and continue to the
next number.  The complete randomization of replicate chambers to positions
for the example is displayed in Table A.6.                !
    TABLE A.6.  RANDOM ASSIGNMENT OF  REPLICATE CHAMBERS TO POSITIONS: EXAMPLE
                ASSIGNMENT OF ALL 24 POSITIONS
Control
Control
100.0%
50.0%
100.0%
12.5%
50.0%
50.0%
6.25%
Control
100.0%
25.0%
6.25%
25.0%
Control
50.0%
6.25%
12.5%
100.0%
12.5% .
12.5%
25.0%
25.0%
6.25%
2.4.6  Four double-digit numbers were assigned to each position (instead of
one, two, or three) in order to make efficient use of the random number table
(Table A.2).  To illustrate, consider the assignment of only one double-digit
number to each position:  the first column of assigned numbers in Table A.5.
Whenever the numbers 00 and 25 through 99 are read from Table A.2, they will
be disregarded and the next number will be read.

3.  OUTLIERS
                                                          I
3.1  An outlier is an inconsistent or questionable data point that appears
unrepresentative of the general trend exhibited by the majority of the data.
Outliers may be detected by tabulation of the data, plotting, and by an
analysis of the residuals.  An explanation should be sought for any
questionable data points.  Without an explanation, data points should be
discarded only with extreme caution.  If there is no explanation, the analysis
should be performed both with and without the outlier, and the results of both
analyses should be reported.

3.2  Gentleman-Wilk's A statistic gives a test for the condition that the
extreme observation may be considered an outlier.  For a discussion of this,
and other techniques for evaluating outliers, see Draper and John (1981).

                                      253

-------
                                  APPENDIX B

         VALIDATING NORMALITY AND HOMOGENEITY OF VARIANCE ASSUMPTIONS


1.  INTRODUCTION

1.1  Dunnett's Procedure and the t test with Bonferroni's adjustment are
parametric procedures based on the assumptions that the observations within
treatments are independent and normally distributed, and that the variance of
the observations is homogeneous across all toxicant concentrations and the
control.  These assumptions should be checked prior to using these tests,  to
determine if they have been met.  Tests for validating the assumptions are
provided in the following discussion.  If the tests fail (if the data do not
meet the assumptions), a nonparametric procedure such as Steel's Many-one Rank
Test may be more appropriate.  However, the decision on whether to use
parametric or nonparametric tests may be a judgment call, and a statistician
should be consulted in selecting the analysis.

2.  TEST FOR NORMAL DISTRIBUTION OF DATA

2.1  SHAPIRO-WILK'S TEST

2.1.1  One formal test for normality is the Shapiro-Wilk's Test (Conover,
1980).  The test statistic is obtained by dividing the square of an
appropriate linear combination of the sample order statistics by the usual
symmetric estimate of variance.  The calculated W must be greater than zero
and less than or equal to one.  This test is recommended for a sample size of
50 or less.  If the sample size is greater than 50, the Kolmogorov "D"
statistic (Stephens, 1974) is recommended. An example of the Shapiro-Wilk's
test is provided below.

2.2  The example uses growth data from the Fathead Minnow Larval Survival  and
Growth Test.  The same data are used in the discussion of the homogeneity of
variance determination in Paragraph 3 and Dunnett's Procedure in Appendix C.
The data, the mean and variance of the observations at each concentration,
including the control, are listed in Table B.I.

2.3  The first step of the test for normality is to center the observations by
subtracting the mean of all the observations within a concentration from each
observation in that concentration.  The centered observations are listed in
Table B.2.

2.4  Calculate the denominator, D, of the test statistic:

                                     A     _ ,
                                D =  ฃ (Xd-X)2


    Where:  X,- -  the centered observations and X is the overall mean of
                  the centered observations.  For this set of data, X = 0,
                  and D = 0.0412.
                                      254

-------
       TABLE B.I.   FATHEAD LARVAL,  PIMEPHALES PROMELAS, LARVAL GROWTH  DATA
                    (WEIGHT  IN MG) FOR THE  SHAPIRO-MILK'S TEST
NaPCP Concentration (ua/L)
Replicate
A
B
C
D
Mean(Y,-)
Si
i
Control
0.711
0.662
0.718
0.767
0.714
0.0018
.1
32
0.646
0.626
0.723
0.700
0.674
0.0020
2
64
0.669
0.669
0.694
0.676
0.677
0.0001
3
128
0.629
0.680
0.513
0.672
0.624
0.0059
4
256
0.650
0.558
0.606
0.508
0.580
0.0037
5
     TABLE B.2.  EXAMPLE OF SHAPIRO-WILK'S TEST:   CENTERED  OBSERVATIONS
NaPCP Concentration
Replicate
A
B
C
D
Control
-0.003
-0.052
0.004
0.053
32
-0.028
-0.048
0.049
0.026
64
-0.008
-0.008
0.017
-0.001
128
0.005
0.056
-0.111
0.048
(ud/L)
256
0.070
-0.022
0.026
-0.072
2.5  Order the centered observations from smallest to  largest.
                     X(1)
-------
  TABLE B.3.   EXAMPLE OF THE SHAPIRO-MILK'S TEST:   ORDERED OBSERVATIONS
1
2
3
4
5
6
7
8
9
10
-0.111
-0.072
-0.052
-0.048
-0.028
-0.022
-0.008
-0.008
-0.003
-0.001
11
12
13
14
15
16
17
18
19
20
0.004
0.005
0.017
0.026
0.026
0.048
0.049
0.053
0.056
0.070
                         W=
                             D 1-1  1
2.7  Compute the test statistic, W, as follows:
    The differences, X
  Yo")
• A  ,
                                   are listed in Table B.5.
2.8  The decision rule for this test is to compare the critical value from
Table B.6 to the computed W.  If the computed value is less than the critical
value, conclude that the data are not normally distributed.  For this example,
the critical value at a significance level of 0.01 and 20 observations (n) is
0.868.  The calculated value, 0.959, is not less than the critical value.
Therefore, conclude that the data are normally distributed.

2.9  In general, if the data fail the test for normality, a transformation
such as to log values may normalize the data.  After transforming the data,
repeat the Shapiro-Milk's Test for normality.
2.10  KOLMOGOROV "D" TEST
                                                                           The
2.10.1  A formal two-sided test for normality is the Kolmogorov "D" Test.
test statistic is calculated by obtaining the difference between the
cumulative distribution function estimated from the data and the standard
normal cumulative distribution function for each standardized observation.
This test is recommended for a sample size greater than 50.  If the sample
size is less than or equal to 50, then the Shapiro Milk's Test is recommended.
An example of the Kolmogorov "D11- test is provided below.

2.10.2  The example uses reproduction data from the daphnid, Ceriodaphm'a
dubia, Survival and Reproduction Test.  The observed data and the mean of
                                      256

-------
TABLE B.4.  COEFFICIENTS FOR THE SHAPIRO-WILK'S TEST (Conover, 1980)
V
iV
1
2
3
4
5

2
0.7071
.

3
0.7071
0.0000

4
0.6872
0.1667

5
Number
6
0.6646 0
0.2413 0
0.0000 0
-
-
.
•
.
-
-
-
-
-
of Observations

.6431
.2806
.0875


7
0.6233
0.3031
0.1401
0.0000
•
8
0.6052
0.3164
0.1743
0.0561
"
<>
0.5888
0.3244
0.1976
0.0947
0.0000
10
0.5739
0.3291
0.2141
0.1224
0.0399
\ Number of Observations
f\ป
1
2
3
4
5
6
7
8
9
10
\
A-
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
11
0.5601
0.3315
0.2260
0.1429
0.0695
0.0000
.
.
.
-

21
0.4643
0.3185
0.2578
0.2119
0.1736
0.1399
0.1092
0.0804
0.0530
0.0263
0.0000
-
-
.
-
12
0.5475
0.3325
0.2347
0.1586
0.0922
0.0303
-
-
-
•

22
0.4590 '
0.3156
0.2571
0.2131
0.1764
0.1443
0.1150
0.0878
0.0618
0.0368
0.0122
-
.
-
-
13
0.5359
0.3325
0.2412
0.1707
0.1099
0.0539
0.0000
-
.
-

23
0.4542
0.3126
0.2563
0.2139
0.1787
0.1480
0.1201
0.0941
0.0696
0.0459
0.0228
0.0000
-
-
-
14
0.5251
0.3318
0.2460
0.1802
0.1240
0.0727
0.0240
-
-
-

24
0.4493
0.3098
0.2554
0.2145
0.1807
0.1512
0.1245
0.0997
0.0764
0.0539
0.0321
0.0107
-
-
-.
15
0.5150
0.3306
0.2495
0.1878
0.1353
0.0880
0.0433
0.0000
-
-
Number
25
0.4450
0.3069
0.2543
0.2148
0.1822
0.1539
0.1283
0.1046
0.0923
0.0610
0.0403
0.0200
0.0000
-
-
16
0.5056
0.3209
0.2521
0.1939
0.1447
0.1005
0.0593
0.0196
-
•
17
0.4968
0.3273
0.2540
0.1988
0.1524
0.1109
0.0725
0.0359
0.0000
'
18
0.4886
0.3253
0.2553
0.2027
0.1587
0.1197
0.0837
0.0496
0.0163
~
19
0.4808
0.3232
0.2561
0.2059
0.1641
0.1271
0.0932
ID. 0612
0.0303
0.0000
20
0.4734
0.3211
0.2565
0.2085
0.1686
0.1334
0.1013
0.0711
0.0422
0.0140
of Observations I
26
0.4407
0.3043
0.2533
0.2151
0.1836
0.1563
0.1316
0.1089
0.0876
0.0672
0.0476
0.0284
0.0094
-
-
27
0.4366
0.3018
0.2522
0.2152
0.1848
0.1584
0.1346
0.1128
0.0923
0.0728
0.0540
0.0358
0.0178
0.0000
-
28
0.4328
0.2992
0.2510
0.2151
0.1857
0.1601
0.1372
0.1162
0.0965
0.0778
0.0598
0.0424
0.0253
0.0084
-
29
0.4291
0.2968
0.2499
0.2150
0.1864
0.1616
0.1395
0.1192
0.1002
0.0822
0.0650
0.0483
0.0320
0.0159
0.0000
30
0.4254
0.2944
0.2487
0.2148
0.1870
0.1630
0.1415
0.1219
0.1036
0.0862
0.0697
0.0537
0.0381
0.0227
0.0076
                                    257

-------
TABLE B.4.  COEFFICIENTS FOR THE SHAPIRO MILK'S TEST (CONTINUED)
\ Number of Observations
l\"
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
tV
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
31
0.4220
0.2921
0.2475
0.2145
0.1874
0.1641
0.1433
0.1243
0.1066
0.0899
0.0739
0.0585
0.0435
0.0289
0.0144
0.0000
•
-
-

31
0.3940
0.2719
0.2357
0.2091
0.1876
0.1693
0.1531
0.1384
0.1249
0.1123
0.1004
0.0891
0.0782
0.0677
0.0575
0.0476
0.0379
0.0283
0.0188
0.0094
0.0000
-
-
-
-
32
0.4188
0.2898
0.2462
0.2141
0.1878
0.1651
0.1449
0.1265
0.1093
0.0931
0.0777
0.0629
0.0485
0.0344
0.0206
0.0068
-
-
-

32
0.3917
0.2701
0.2345
0.2085
0.1874
0.1694
0.1535
0.1392
0.1259
0.1136
0.1020
0.0909
0.0804
0.0701
0.0602
0.0506
0.0411
0.0318
0.0227
0.0136
0.0045
-
-
-
-
33
0.4156
0.2876
0.2451
0.2137
0.1880
0.1660
0.1463
0.1284
0.1118
0.0961
0.0812
0.0669
0.0530
0.0395
0.0262
0.0131
0.0000
-
-

33
0.3894
0.2684
0.2334
0.2078
0.1871
0.1695
0.1539
0.1398
0.1269
0.1149
0.1035
0.0927
0.0824
0.0724
0.0628
0.0534
0.0442
0.0352
0.0263
0.017.5
0.0087
0.0000
-
-
-
34
0.4127
0.2854
0.2439
0.2132
0.1882
0.1667
0.1475
0.1301
0.1140
0.0988
0.0844
0.0706
0.0572
0.0441
0.0314
0.0187
0.0062
-
-

34
0.3872
0.2667
0.2323
0.2072
0.1868
0.1695
0.1542
0.1405
9,1278
0.1160
0.1049
0.0943
0.0842
0.0745
0.0651
0.0560
0.0471
0.0383
0.0296
0.0211
0.0126
0.0042
-
-
-
35
0.4096
0.2834
0.2427
0.2127
0.1883
0.1673
0.1487
0.1317
0.1160
0.1013
0.0873
0.0739
0.0610
0.0484
0.0361
0.0239
0.0119
0.0000
-

Number of
35
0.3850
0.2651
0.2313
0.2065
0.1865
0.1695
0.1545
. 0.1410
0.1286
9.1170
0.1062
0.0959
0.0860
0.0765
0.0673
0.0584
0.0497
0.0412
0.0328
0.0245
0.0163
0.0081
0.0000
-
-
36
0.4068
0.2813
0.2415
0.2121
0.1883
0.1678
0.1496
0.1331
0.1179
0.1036
0.0900
0.0770
0.0645
0.0523
0.0404
0.0287
0.0172
0.0057
-

37
0.4040
0.2794
0.2403
0.2116
0.1883
0.1663
0.1505
0.1344
0.1196
0.1056
0.0924
0.0798
0.0677
0.0559
0.0444
0.0331
0.0220
0.0110
0.0000

Observations
36 37
0.3830
0.2635
0.2302
0.2058
0.1862
0.1695
0.1548
0.1415
0.1293
0.1180
0.1073
0.0972
0.0876
0.0783
0.0694
0.0607
0.0522
0.0439
0.0357
0.0277
0.0197
0.0118
0.0039
-
-
0.3808
0.2620
0.2291
0.2052
0.1859
0.1695
0.1550
0.1420
0.1300
0.1189
0.1085
0.0986
0.0892
0.0801
0.0713
0.0628
0.0546
0.0465
0.0385
0.0307
0.0229
0.0153
0.0076
0.0000
-
38
0.4015
0.2774
0.2391
0.2110
0.1881
0.1686
0.1513
0.1356
0.1211
0.1075
0.0947
0.0824
0.0706
0.0592
0.0481
0.0372
0.0264
0.0158
0.0053

38
0.3789
0.2604
0.2281
0.2045
0.1855
0.1693
0.1551
0.-1423
0.1306
0.1197
0.1095
0.0998
0.0906
0.0817
0.0731
0.0648
0.0568
0.0489
0.0411
0.0335
0.0259
0.0185
0.0111
0.0037
-
39
0.3989
0.2755
0.2380
0.2104
0.1880
0.1689
0.1520
0.1366
0.1225
0.1092
0.0967
0.0848
0.0733
0.0622
0.0515
0.0409
0.0305
0.0203
0.0101
0.0000
39
0.3770
0.2589
0.2271
0.2038
0.1851
0.1692
0.1553
0.1427
0.1312
0.1205
0.1105
0.1010
0.0919
0.0832
0.0748
0.0667
0.0588
0.0511
0.0436
0.0361
0.0288
0.0215
0.0143
0.0071
0.0000
40
0.3964
0.2737
0.2368
0.2098
0.1878
0.1691
0.1526
0.1376
0.1237
0.1108
0.0986
0.0870
0.0759
0.0651
0.0546
0.0444
0.0343
0.0244
0.0146
0.0049
40
0.3751
0.2574
0.2260
0.2032
0.1847
0.1691
0.1554
. 0.1430
0.1317
0.1212
0.1113
0.1020
0.0932
0.0846
0.0764
0.0685
' 0.0608
0.0532
0.0459
0.0386
0.0314
0.0244
0.0174
0.0104
0.0035
                                 258

-------
        TABLE B.5.  EXAMPLE OF THE SHAPIRO-MILK'S TEST:
                    TABLE OF COEFFICIENTS AND DIFFERENCES
                                      _ w(
1
2
3
4
5
6
7
8
9
10
0.4734
0.3211
0.2565
0.2085
0.1686
0.1334
0.1013
0.0711
0.0422
0.0140
0.181
0.128
0.105
0.097
0.076
0.048
0.034
0.025
0.008
0.005
x<20)
X(19)
X(18>
X(17)
XC16>
x<15)
X(14)
X(13)
X(12)

- ^
- x(2)
- x(3)
- XC4)
y(5)
~ A
y<6)
- A
- >;C7)
y(8)
- A
y(9)
'. ~ A
y(10)
- A
the observations at each concentration, including the control, are listed in
Table B.7.

2.10.3  The first step of the test for normality is to center the observations
by subtracting the mean of all the observations within a concentration from
each observation in that concentration.  The centered observations for the
example are listed in Table B.8.

2.10.4  Order the centered observations from smallest to largest:

                           yd) < y(2)         y(n)


where X(1) denotes the ith ordered observation, and n denotes the total number
of centered observations.  The ordered observations for the example are listed
in Table B.9.

2.10.5  The next step is to standardize the ordered observations.  Let z,-
denote the standardized value of the ith ordered observation.  Then,
                        Zi ~
and
                                               (n-l)
For the example, s = 5.3, and the standardized observations are listed in
Table B.9.

2.10.6  From Table B.10, obtain the value of the standard normal cumulative
distribution function (standard normal CDF) at z*.   Denote this value  as  p{.
Note that negative z are not listed in Table B.10.   The value of the standard
normal CDF at a negative number is one minus the value of the standard normal
                                      259

-------
TABLE B.6.  QUANTILES OF THE SHAPIRO-WILK'S TEST STATISTIC (Conover, 1980)
n
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
0.01
0.753
0.687
0.686
0.713
0.730
0.749
0.764
0.781
0.792
0.805
0.814
0.825
0.835
0.844
0.851
0.858
0.863
0.868
0.873
0.878
0.881
0.884
0.888
0.891
0.894
0.896
0.898
0.900
0.902
0.904
0.906
0.908
0.910
0.912
0.914
0.916
0.917
0.919
0.920
0.922
0.923
0.924
0.926
0.927
0.928
0.929
0.929
0.930
0.02
0.756
0.707
0.715
0.743
0.760
0.778
0.791
0.806
0.817
0.828
0.837
0.846
0.855
0.863
0.869
0.874
0.879
0.884
0.888
0.892
0.895
0.898
0.901
0.904
0.906
0.908
0.910
0.912
0.914
0.915
0.917
0.919
0.920
0.922
0.924
0.925
0.927
0.928
0.929
0.930
0.932
0.933
0.934
0.935
0.936
0.937
0.937
0.938
0.05
0.767
0.748
0.762
0.788
0.803
0.818
0.829
0.842
0.850
0.859
0.866
0.874
0.881
0.887
0.892
0.897
0.901
0.905
0.908
0.911
0.914
0.916
0.918
0.920
0.923
0.924
0.926
0.927
0.929
0.930
0.931
0.933
0.934
0.935
0.936
0.938
0.939
0.940
0.941
0.942
0.943
0.944
0.945
0.945
0.946
0.947
0.947
0.947
0.10
0.789
0.792
0.806
0.826
0.838
0.851
0.859
0.869
0.876
0.883
0.889
0.895
0.901
0.906
0.910
0.914
0.917
0.920
0.923
0.926
0.928
0.930
0.931
0.933
0.935
0.936
0.937
0.939
0.940
0.941
0.942
0.943
0.944
0.945
0.946
0.947
0.948
0.949
0.950
0.951
0.951
0.952
0.953
0.953
0.954
0.954
0.955
0.955
0.50
0.959
0.935
0.927
0.927
0.928
0.932
0.935
0.938
0.940
0.943
0.945
0.947
0.950
0.952
0.954
0.956
0.957
0.959
0.960
0.961
0.962
0.963
0.964
0.965
0.965
0.966
0.966
0.967
0.967
0.968
0.968
0.969
0.969
0.970
0.970
0.971
0.971
0.972
0.972
0.972
0.973
0.973
0.973
0.974
0.974
0.974
0.974
0.974
0.90
0.998
0.987
0.979
0.974
0.972
0.972
0.972
0.972
0.973
0.973
0.974
0.975
0.975
0.976
0.977
0.978
0.978
0.979
0.980
0.980
0.981
0.981
0.981
0.982
0.982
0.982
0.982
0.983
0.983
0.983
0.983
0.983
0.984
0.984
0.984
0.984
0.984
0.985
0.985
0.985
0.985
0.985
0.985
0.985
0.985
0.985
0.985
0.985
0.95
0.999
0.992
0.986
0.981
0.979
0.978
0.978
0.978
0.979
0.979
0.979
0.980
0.980
0.981
0.981
0.982
0.982
0.983
0.983
0.984
0.984
0.984
0.985
0.985
0.985
0.985
0.985
0.985
0.986
0.986
0.986
0.986
0.986
0.986
0.987
0.987
0.987
0.987
0.987
0.987
0.987
0.987
0.988
0.988
0.988
0.988
0.988
0.988
0.98
1.000
0.996
0.991
0.986
0.985
0.984
0.984
0.983
0.984
0.984
0.984
0.984
0.984
0.985
0.985
0.986
0.986
0.986
0.987
0.987
0.987
0.987
0.988
0.988
0.988
0.988
0.988
0.988
0.988
0.988
0.989
0.989
0.989
0.989
0.989
0.989
0.989
0.989
0.989
0.989
0.990
0.990
0.990
0.990'
0.990
0.990
0.990
0.990
0.99
1.000
0.997
0.993
0.989
0.988
0.987 '
0.986
0.986
0.986
0.986
0.986
0.986
0.987
0.987
0.987
0.988
0.988
0.988
0.989
0.989
0.989
0.989
0.989
0.989
0.990
0.990
0.990
0.990
0.990
0.990
0.990
0.990
0.990
0.990
0.990
0.990
0.991
0.991
0.991
0.991
0.991
0.991
0.991
0.991
0.991
0.991
0.991
0.991
                                    260

-------
               TABLE B.7.   CERIODAPHNIA DUBIA REPRODUCTION DATA
                           FOR THE KOLMOGOROV "D" TEST
                                   Effluent Concentration (%)
                Control
             1.56
            3.12
          6.25
          12.5
           25.0

1
2
3
4
5
6
7
8
9
10
27
30
29
31
16
15
18
17
14
27
32
35
32
26
18
29
27
16
35
13
39
30
33
33
36
33
33
27
38
44
27
34
36
34
31
27
33
31
33
31
19
25
26
17
16
21
23
15
18
10
10
13
7
7
7
10
10
16
12
2

CDF at the absolute value of that number.  For example, since the value of the
standard normal CDF at 3.21 is 0.9993, the value of the standard normal CDF at
-3.21 is 1 - 0.9993 = 0.0007.  The p1  values  for the example data are listed
in Table B.9.
         TABLE B.8.  CENTERED OBSERVATIONS FOR KOLMOGOROV "D" EXAMPLE

                                      Effluent Concentration (%)
  Replicate
Control
1.56
3.12
6.25
12.5
25.0
1
2
3
4
5
6
7
8
9
10
4.6
7.6
6.6
8.6
-6.4
-7.4
-4.4
-5.4
-8.4
4.6
5.7
8.7
5.7
-0.3
-8.3
2.7
0.7
-10.3
8.7
-13.3
4.4
-4.6
-1.6
-1.6
1.4
-1.6
-1.6
-7.6
3.4
9.4
-4.7
2.3
4.3
2.3
-0.7
-4.7
1.3
-0.7 ,
1.3
-0.7
0.0
6.0
7.0
-2.0
-3.0
2.0
4.0
-4.0
-1.0
-9.0
0.6
3.6
-2.4
-2.4
-2.4
0.6
0.6
6.6
2.6
-7.4
                                      261

-------
TABLE B.9. EXAMPLE CALCULATION OF THE KOLMOGOROV "D" STATISTIC
i
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
X(f)
-13.3
-10.3
-9.0
-8.4
-8.3
-7.6
-7.4
-7.4
-6.4
-5.4
-4.7
-4.7
-4.6
-4.4
-4.0
-3.0
-2.4
-2.4
-2.4
-2.0
-1.6
-1.6
-1.6
-1.6
-1.0
-0.7
-0.7
-0.7
-0.3
0.0
0.6
0.6
0.6
0.7
1.3
1.3
1.4
2.0
2.3
2.3
2.6
2.7
3.4
3.6
4.0
4.3
4.4
zi
-2.51
-1.94
-1.70
-1.58
-1.57
-1.43
-1.40
-1.40
-1.21
-1.02.
-0.89
-0.89
-0.87
-0.83
-0.75
-0.57
-0.45
-0.45
-0.45
-0.38
-0.30
-0.30
-0.30
-0.30
-0.19
-0.13
-0.13
-0.13
-0.06
0.00
0.11
0.11
0.11
0.13
0.25
0.25
0.26
0.38
0.43
0.43
0.49
0.51
0.64
0.68
0.75
0.81
0.83
Pi
0.0060
0.0262
0.0446
0.0571
0.0582
0.0764
0.0808
0.0808
0.1131
0.1539
0.1867
0.1867
0.1922
0.2033
0.2266
0.2843
0.3264
0.3264
0.3264
0.3520
0.3821
0.3821
0.3821
0.3821
0.4247
0.4483
0.4483
0.4483
0.4761
0.5000
0.5438
0.5438
0.5438
0.5517
0.5987
0.5987
0.6026
0.6480
0.6664
0.6664
0.6879
0.6950
0.7389
0.7517
0.7734
0.7910
0.7967
Di+
0.0107
0.0071
0.0054
0.0096
0.0251
0.0236
0.0359
0.0525
0.0369
0.0128
-0.0034
0.0133
0.0245
0.0300
0.0234
-0.0176
-0.0431
-0.0264
-0.0097
-0.0187
-0.0321
-0.0154
0.0012
0.0179
-0.0080
-0.0150
0.0017
0.0184
0.0072
0.0000
-0.0271
-0.0105
0.0062
0.0150
-0.0154
0.0013
0.0141
-0.0147
-0.0164
0.0003
-0.0046
0.0050
-0.0222
-0.0184
-0.0234
-0.0243
-0.0134
Dr
0.0060
0.0095
0.0113
0.0071
-0.0085
-0.0069
-0.0192
-0.0359
-0.0202
0.0039
0.0200
0.0034
-0.0078
-0.0134
-0.0067
0.0343
0.0597
0.0431
0.0264
0.0353
0.0488
0.0321
0.0154
-0.0012
0.0247
0.0316
0.0150
-0.0017
0.0094
0.0167
0.0438
0.0271
0.0105
0.0017
0.0320
0.0154
0.0026
0.0313
0.0331
0.0164
0.0212
0.0117
0.0389
0.0350
0.0401
0.0410
0.0300
262

-------
   TABLE B.9.  EXAMPLE CALCULATION OF THE KOLMOGOROV "D" STATISTIC (CONTINUED)

            i         X<0       Zi      '   p,         D,+         Dr
48
49
50
51
52
53
54
55
56
57
58
59
60
4.6
4.6
5.7
5.7
6.0
6.6
6.6
7.0
7.6
8.6
8.7
8.7
9.4
0.87
0.87
1*08
1.08
1.13
1.25
1.25
1.32
1.43
1.62
1.64
1.64
1.77
0.8078
0.8078
0.8599
0.8599
0.8708
0.8944
0.8944
0.9066
0.9236
0.9474
0.9495
0.9495
0.9616
-0.0078
0.0089
-0.0266
-0.0099
-0.0041
-0.0111
0.0056
0.0101
0.0097
0.0026
0.0172
0.0338
0.0384
0.0245
0.0078
0.0432
0.0266
0.0208
0.0277
0.0111
0.0066
0.0069
0.0141
-0.0005
-0.0172
-0.0217
2.10.7  Next, calculate the following differences for each ordered
observation:

                         D,+  =  (i/n)  -  Pi

                         Dr  =  P|  - [(i-l)/n]

The differences for the example are listed in Table B.9.

2.10.8  Obtain the maximum of the 0,-+,  and denote it  as D+.   Obtain  the
maximum of the D--,  and denote it as D-.   For the example,  D+ =  0.0525,  and  D-
= 0.0597.

2.10.9  Next, obtain the maximum of D+ and D-, and denote it as  D.  For the
example, D = 0.0597.

2.10.10  The test statistic, D*, is calculated as follows:

                          D* = Z>(/n-0.01+0.85/v/Ja)

For the example, D* = 0.4684.

2.10.11  The decision rule for the two tailed test is to  compare the critical
value from Table B.ll to the computed D*.  If the computed value is  greater
than the critical value, conclude that the data are not normally distributed.
For this example, the critical value at a significance level of  0.01 is 1.035.
The calculated value, 0.4684, is not greater than the critical value.  Thus,
the conclusion of the test is that the data are normally  distributed.

2.10.12  In general, if the data fail the test for normality,, a  transformation
such as the log transformation may normalize the data. After transforming the
data, repeat the Kolmogorov "D" test for normality.

                                     263

-------
TABLE B.10. P IS THE VALUE OF THE STANDARD NORMAL CUMULATIVE DISTRIBUTION
            AT Z
Z
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
0.11
0.12
0.13
0.14
0.15
0.16
0.17
0.18
0.19
0.20
0.21
0.22
0.23
0.24
0.25
0.26
0.27
0.28
0.29
0.30
0.31
0.32
0.33
0.34
0.35
0.36
0.37
0.38
0.39
0.40
P
0.5000
0.5040
0.5080
0.5120
0.5160
0.5199
0.5239
0.5279
0.5319
0.5359
0.5398
0.5438
0.5478
0.5517
0.5557
0.5596
0.5636
0.5675
0.5714
0.5753
0.5793
0.5832
0.5871
0.5910
0.5948
0.5987
0.6026
0.6064
0.6103
0.6141
0.6179
0.6217
0.6255
0.6293
0.6331
0.6368
0.6406
0.6443
0.6480
0.6517
0.6554
z
0.41
0.42
0.43
0.44
0.45
0.46
0.47
0.48
0.49
0.50
0.51
0.52
0.53
0.54
0.55
0.56
0.57
0.58
0.59
0.60
0.61
0.62
0.63
0.64
0.65
0.66
0.67
0.68
0.69
0.70
0.71
0.72
0.73
0.74
0.75
0.76
0.77
0.78
0.79
0.80
0.81
P
0.6591
0.6628
0.6664
0.6700
0.6736
0.6772
0.6808
0.6844
0.6879
0.6915
0.6950
0.6985
0.7019
0.7054
0.7088
0.7123
0.7157
0.7190
0.7224
0.7257
0.7291
0.7324
0.7357
0.7389
0.7422
0.7454
0.7486
0.7517
0.7549
0.7580
0.7611
0.7642
0.7673
0.7704
0.7734
0.7764
0.7794
0.7823
0.7852
0.7881
0.7910
z
0.82
0.83
0.84
0.85
0.86
0.87
0.88
0.89
0.90
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1.00
1.01
1.02
1.03
1.04
1.05
1.06
1.07
1.08
1.09
1.10
1.11
1.12
1.13
1.14
1.15
1.16
1.17
1.18
1.19
1.20
1.21
1.22
P
0.7939
0.7967
0.7995
0.8023
0.8051
0.8078
0.8106
0.8133
0.8159
0.8186
0.8212
0.8238
0.8264-
0.8289
0.8315
0.8340
0.8365
0.8389
0.8413
0.8438
0.8461
0.8485
0.8508
0.8531
0.8554
0.8577
0.8599
0.8621
0.8643
0.8665
0.8686
0.8708
0.8729
0.8749
0.8770
0.8790
0.8810
0.8830
0.8849
0.8869
0.8888
z
1.23
1.24
1.25
1.26
1.27
1.28
1.29
1.30
1.31
1.32
1.33
1.34
1.35
1.36
1.37
1.38
1.39
1.40
1.41
1.42
1.43
1.44
1.45
1.46
1.47
1.48
1.49
1.50
1.51
1.52
1.53
1.54
1.55
1.56
1.57
1.58
1.59
1.60
1.61
1.62
1.63
P
0.8907
0.8925
0.8944
0.8962
0.8980
0.8997
0.9015
0.9032
0.9049
0.9066
0.9082 .
0.9099
0.9115
0.9131
0.9147
0.9162
0.9177
0.9192
0.9207
0.9222
0.9236
0.9251
0.9265
0.9279
0.9292
0.9306
0.9319
0.9332
0.9345
0.9357
0.9370
0.9382
0.9394
0.9406
0.9418
0.9429
0.9441
0.9452
0.9463
0.9474
0.9484
                                   264

-------
TABLE B.10. P IS THE VALUE OF THE STANDARD NORMAL CUMULATIVE DISTRIBUTION
            AT Z (CONTINUED)
z
1.64
1.65
1.66
1.67
1.68
1.69
1.70
1.71
1.72
1.73
1.74
1.75
1.76
1.77
1.78
1.79
1.80
1.81
1.82
1.83
1.84
1.85
1.86
1.87
1.88
1.89
1.90
1.91
1.92
1.93
1.94
1.95
1.96
1.97
1.98
1.99
2.00
2.01
2.02
2.03
2.04
P
0.9495
0.9505
0.9515
0.9525
0.9535
0.9545
0.9554
0.9564
0.9573
0.9582
0.9591
0.9599
0.9608
0.9616
0.9625
0.9633
0.9641
0.9649
0.9656
0.9664
0.9671
0.9678
0.9686
0.9693
0.9699
0.9706
0.9713
0.9719
0.9726
0.9732
0.9738
0.9744
0.9750
0.9756
0.9761
0.9767
0.9772
0.9778
0.9783
0.9788
0.9793
z
2.05
2.06
2.07
2.08
2.09
2.10
2.11
2.12
2.13
2.14
2.15
2.16
2.17
2.18
2.19
2.20
2.21
2.22
2.23
2.24
2.25
2.26
2.27
2.28
2.29
2.30
2.31
2.32
2.33
2.34
2.35
2.36
2.37
2.38
2.39
2.40
2.41
2.42
2.43
2.44
2.45
P
0.9798
0.9803
0.9808
0.9812
0.9817
0.9821
0.9826
0.9830
0.9834
0.9838
0.9842
0.9846
0.9850
0.9854
0.9857
0.9861
0.9864
0.9868
0.9871
0.9875
0.9878
0.9881
0.9884
0.9887
0.9890
0.9893
0.9896
0.9898
0.9901
0.9904
0.9906
0.9909
0.9911
0.9913
0.9916
0.9918
0.9920
0.9922
0.9925
0.9927
0.9929
z
2.46
2.47
2.48
2.49
2.50
2.51
2.52
2.53
2.54
2.55
2.56
2.57
2.58
2.59
2.60
2.61
2.62
2.63
2.64
2.65
2.66
2.67
2.68
2.69
2.70
2.71
2.72
2.73
2.74
2.75
2.76
2.77
2.78
2.79
2.80
2.81
2.82
2.83
2.84
2.85
2.86
P
0.9931
0.9932
0.9934
0.9936
0.9938
0.9940
0.9941
0.9943
0.9945
0.9946
0.9948
0.9949
0.9951
0.9952
0.9953
0.9955
0.9956
0.9957
0.9959
0.9960
0.9961
0.9962
0.9963
0.9964
0.9965
0.9966
0.9967
0.9968
0.9969
0.9970
0.9971
0.9972
0.9973
0.9974
0.9974
0.9975
0.9976
0.9977
0.9977
0.9978
0.9979
z
2.87
2.88
2.89
2.90
2.91
2.92
2.93
2.94
2.95
2.96
2.97
2.98
2.99
3.00
3.01
3.02
3.03
3.04
3.05
3.06
3.07
3.08
3.09
3.10
3.11
3.12
3.13
3.14
3.15
3.16
3.17
3.18
3.19
3.20
3.21
3.22
3.23
3.24
3.25
3.26
3.27
P
0.9979
0.9980
0.9981
0.9981
0.9982
0.9982
0.9983
0.9984
0.9984
0.9985
0.9985
0.9986
0.9986
0.9987
0.9987
0.9987
0.9988
0.9988
0.9989
0.9989
0.9989
0.9990
0.9990
0.9990
0.9991
0.9991
0.9991
0.9992
0.9992
0.9992
0.9992
0.9993
0.9993
0.9993
0.9993
0.9994
0.9994
0.9994
0.9994
0.9994
0.9995

                                    265

-------
  TABLE B.10. P IS THE VALUE OF THE STANDARD NORMAL CUMULATIVE DISTRIBUTION
              AT Z (CONTINUED)
z
3.28
3.29
3.30
3.31
3.32
3.33
3.34
3.35
3.36
3.37
3.38
3.39
3.40
3.41
3.42
3.43
3.44
3.45
P
0.9995
0.9995
0.9995
0.9995
0.9995
0.9996
0.9996
0.9996
0.9996
0.9996
0.9996
0.9997 '
0.9997
0.9997
0.9997
0.9997
0.9997
0.9997
z
3.46
3.47
3.48
3.49
3.50
3.51
3.52
3.53
3.54
3.55
3.56
3.57
3.58
3.59
3.60
3.61
3.62
3.63
P
0.9997
0.9997
0.0997
0.9998
0.9998
0.9998
0.9998
0.9998
0.9998
0.9998
0.9998
0.9998
0.9998
0.9998
0.9998
0.9998
0.9999
0.9999
z
3.64
3.65
3.66
3.67
3.68
3.69
3.70
3.71
3.72
3.73
3.74
3.75
3.76
3.77
3.78
3.79
3.80
3.81
P
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
z
3.82
3.83
3.84
3.85
3.86
3.87
3.88
3.89
3.90
3.91
3.92
3.93
3.94
3.95
3.96
3.97
3.98
3.99
P
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
      TABLE B.ll.  CRITICAL VALUES FOR THE KOLMOGOROV "D" TEST
Alpha
Level
0.010
0.025
0.050
0.100
0.150
Critical
Value
1.035
0.955
0.895
0.819
0.775
3.  TEST FOR HOMOGENEITY OF VARIANCE

3.1  For Dunnett's Procedure and the t test with Bonferroni's adjustment, the
variances of the data obtained from each toxicant concentration and the
control are assumed to be equal.  Bartlett's Test is a formal test of this
assumption.  In using this test, it is assumed that the data are normally
distributed.

                                      266

-------
3.2  The data used in this example are growth data from a Fathead Minnow
Larval Survival and Growth Test, and are the same data used in Appendices C
and D.  These data are listed in Table B.12, together with the calculated
variance for the control and each toxicant concentration.

    TABLE B.12.  FATHEAD LARVAL GROWTH DATA (WEIGHT IN MG) USED FOR
                 BARTLETT'S TEST FOR HOMOGENEITY OF VARIANCE
NaPCP Concentration (uq/L)
Rep! i cate
A
B
C
D
M|an(Y,-)
Si
1
Control
0.711
0.662
0.718
0.767
0.714
0.0018
1
32
0.646
0.626
0.723
0.700
0.674
0.0020
2
64
0.669
0.669
0.694
0.676
0.677
0.0001
3
128
0.629
0.680
0.513
0.672
0.624
0.0059
4
256
0.650
0.558
0.606
0.508
0.580
0.0037
5
3.3  The test statistic for Bartlett's Test (Snedecor and Cochran, 1980) is as
follows:
                     rj) InS2- EVjlnS'f]
          S              —ฃ

Where:  V,.  =  degrees of freedom for each toxicant concentration and control

        p   =  number of levels of toxicant concentration including the
               control

        In  =  loge

         i  =  1, 2,  ..., p where p is the number of concentrations

         n, = .the number of replicates for concentration i.
                                      267

-------
         C - 1+ (3 (p-1) ) -1 [ ฃ 1/Vฑ - (
3.4  Since B is approximately distributed as chi -square with p - 1 degrees of
freedom when the variances are equal, the appropriate critical value is
obtained from a table of the chi-square distribution for p-1 degrees of
freedom and a significance level of 0.01.  If B is less than the critical
value then the variances are assumed to be equal.

3.5  For the data in this example, V,- = 3,  p = 5,  32  = 0.0027,  and
C ซ 1.133.  The calculated B value is:

                          (15) [In (0.0027 )] -3 fin (Si)
                     B =
                                      1.133                "  '


                           15 (-5.9145) -3 (-32.4771)
                                      1.133

                       = 7.691

3.6  Since B is approximately distributed as chi-square with p-1  degrees  of
freedom when the variances are equal, the appropriate critical  value for the
test is 13.277 for a significance level  of 0.01.   Since B = 7.691  is less than
the critical value of 13.277, conclude that the variances are not  different.

4.  TRANSFORMATIONS OF THE DATA

4.1  When the assumptions of normality and/or homogeneity of variance are not
met, transformations of the data may remedy the problem,  so that the data can
be analyzed by parametric procedures, rather than by nonparametric  technique
such as Steel's Many-one Rank Test or Wilcoxon's  Rank Sum Test.  Examples of
transformations include log, square root, arc sine square root,  and
reciprocals.  After the data have been transformed,  Shapiro-Milk's  and
Bartlett's tests should be performed on  the transformed observations to
determine whether the assumptions of normality and/or homogeneity  of variance
are met.

4.2  ARC SINE SQUARE ROOT TRANSFORMATION (USEPA,  1993)

4.2.1  For data consisting of proportions from a  binomial  (response/no
response; live/dead) response variable,  the variance within the  ith treatment
is proportional to P,- (1  -  P.), where  P,-  is the expected proportion  for the
treatment.  This clearly violates the homogeneity of variance assumption
required by parametric procedures such as Dunnett's  Procedure or the t test
with Bonferroni's adjustment, since the  existence of a treatment effect
implies different values of Pi for different  treatments,  i.   Also,  when  the
observed proportions are based on small  samples,  or  when  P,-  is close  to  zero
                                     268

-------
or one, the normality assumption may be invalid.  The arc sine square root
(arc sine J P ) transformation is commonly used for such data to stabilize the
variance and satisfy the normality requirement.
                                                        i
4.2.2  Arc sine transformation consists of determining the angle (in radians)
represented by a sine value.  In the case of arc sine square root
transformation of mortality data, the proportion of dead (or affected)
organisms is taken as the sine value, the square root of the sine value is
calculated, and the angle (in radians) for the square root of the sine value
is determined.  Whenever the proportion dead is 0 or 1, a special modification
of the arc sine square root transformation must be used (Bartlett, 1937).  An
explanation of the arc sine square root transformation aind the modification is
provided below.

4.2.3  Calculate the response proportion (RP) at each effluent concentration,
where:                                                  !

    RP = (number of surviving or "unaffected" organisms)/(number exposed)

    Example;  If 12 of 20 animals in a given treatment replicate survive:
                                                        i
              RP = 12/20

                 = 0.60

4.2.4  Transform each RP to its arc sine square root, as follows:

4.2.4.1  For RPs greater than zero or less than one:


         Angle (radians) = arc sine


         Example: If RP = 0.60:


                  Angle = arc sine
                                                        i

                        = arc sine 0.7746

                        = 0.8861 radians

4.2.4.2  Modification of the arc sine square root when RP = 0:


         Angle (in radians) = arc sine  ^1/4 #


         Where: N = Number of animals/treatment replicate
                                      269

-------
         Example: If 20 animals are used:
                  Angle = arc sine  /i/so
                        = arc sine 0.1118
                        = 0.1120 radians
4.2.4.3  Modification of the arc sine square root when RP
         Angle - 1.5708 radians - (radians for RP = 0)
         Example: Using above value:
                  Angle = 1.5708 - 0.1120
                        = 1.4588  radians
= 1.0:
                                     270

-------
                                  APPENDIX C

                              DUNNETT'S PROCEDURE
1.  MANUAL CALCULATIONS
                                                          j
1.1  Dunnett's Procedure (Dunnett, 1955; Dunnett, 1964) is used to compare
each concentration mean with the control mean to decide if any of the
concentrations differ from the control.  This test has an overall error rate
of alpha, which accounts for the multiple comparisons with the control.  It is
based on the assumptions that the observations are independent and normally
distributed and that the variance of the observations is homogeneous across
all concentrations and control (see Appendix B for a discussion on validating
the assumptions).  Dunnett's Procedure uses a pooled estimate of the variance,
which is equal to the error value calculated in an analysis of variance.
Dunnett's Procedure can only be used when the same number of replicate test
vessels have been used at each concentration and the control.  When this
condition is not met, a t test with Bonferroni's adjustment is used (see
Appendix D).

1.2  The data used in this example are growth data from a Fathead Minnow
Larval Survival and Growth Test, and are the same data used in Appendices B
and D.  These data are listed in Table C.I.
   TABLE C.I.  FATHEAD MINNOW, PlMEPHALES PROMELAS, LARVAL GROWTH DATA
               (WEIGHT IN MG) USED FOR DUNNETT'S PROCEDURE
  Replicate
Control
                                     NaPCP Concentration fyq/L)
32
64
128
256
A
B
C
D
Mean (7,-)
Total (Tf)
0.711
0.662
0.646
0.690
0.677
2.709
0.517
0.501
0.723
0.560
0.575
2.301
0.602
0.669
0.694
0.676
0.660
2.641
0.566
0.612
0.410
0.672
0.565
2.260
0.455
0.502
0.606
0.254
0.454
1.817
1.3  One way to obtain an estimate of the pooled variance is to construct an
ANOVA table including all sums of squares, using the following formulas:
                                      271

-------
    Where:  p - number of effluent  concentrations including:
      SST = E Yi:j2-G2/N    Total Sum of Squares
      3SB = ETj2/n.i-G2/N  Between Sum of Squares
      SSW = SST-SSB
Within Sum of Squares
       G  -  the grand total of all sample observations;  c? = E r^

      T5  -  the total of the replicate measurements for  concentration i
       N  =  the total sample size;   N = ~
      n,-  -  the number of replicates for concentration i
     Y,j  =  the jth observation for concentration i
1.4  For the data in this example:
       n1 = n2 - n3 = n4 = n5 = 4
       n  =  20
       TI s  Yn +  Y12 + Y13 + YM = 2.709
       %:  fe:fe:fe:fe:l:Si
       T4 =  Y41 +  Y42 + Y43 + Y44 = 2 • 26ฐ
T             V   _i_  V   j- V   j. V   — 1 Q17
        5    Y51 +  Y52 + Y53 + Y54 " 1'bl/
       G  =  T, -J- T2  + T3 + T4 + T5 = 11.728
     SST = E Yi:j2-Gz/N

          =  7.146 -  (11.728)2/20
          =  0.2687
                                  272

-------
       SSB = ETi2/ni-G2/N






                 1/4 (28.017 - 11.728)2/20



                 0.1270




       SSW = SST-SSB




                 0.2687 - 0..1270



                 0.1417



1.5  Summarize these data in the ANOVA table (Table C.2).



               TABLE C.2.  ANOVA TABLE FOR DUNNETT'S PROCEDURE
Source df
Between p - 1
Within N - p
Total N - 1
Sum of
Squares (SS)
SSB
SSW
SST
Mean Square (MS)
(SS/df)
i
s2 ;- SSB/(P-D
S2 = SSW/(N-p)
1 '
1
i
1.6  Summarize data for ANOVA (Table C.3).



        TABLE C.3.   COMPLETED ANOVA TABLE FOR DUNNETT'S  PROCEDURE
Source
Between
Within
df
5-1=4
20 - 5 = 15
SS
0.1270
0.1417
•
Mean Square
0.0318
i
0.0094
  Total
19
0.2687
                                     273

-------
1.7  To perform the individual comparisons, calculate the t statistic for each
concentration and control combination, as follows:
   Where:  Y{ ซ=  mean for concentration i

           Y, =  mean for the control

           S  =  square root of the within mean square

           n., -  number of replicates in the control

           n- =  number of replicates for concentration i.

1.8  Table C.4 includes the calculated t values for each concentration and
control combination.
                      TABLE C.4.  CALCULATED T VALUES.
                NaPCP
            Concentration
32
64
128
256
2
3
4
5
1.487
0.248
1.633
3.251
 1.9   Since  the  purpose  of the  test  is  only  to  detect  a  decrease  in growth  from
 the  control,  a  one-sided  test  is  appropriate.   The  critical  value for  the
 one-sided comparison  (2.36), with an overall alpha  level  of  0.05, 15 degrees
 of freedom  and  four concentrations  excluding the  control  is  read from  the
 table of Dunnett's  "T"  values  (Table C.5; this table  assumes an  equal  number
 of replicates in  all  treatment concentrations  and the control).  The mean
 weight for  concentration  i  is  considered significantly  less  than the mean
 weight for  the  control  if t,. is greater  than the  critical value.  Since T5 is
 greater than  2.36,  the  256 M9/L concentration  has significantly  lower  growth
 than the control.   Hence  the NOEC and  LOEC  for growth are 128 ng/L and 256
 /*g/L, respectively.
                                      274

-------
CO
 S-
 O)
=

00
CO

-------
1.10  To quantify the sensitivity of the test, the minimum significant
difference (MSD) may be calculated.  The formula is as follows:
                          MSD = d StfjCL/nJ + (1/23)

      Where:   d   =  critical  value for the Dunnett's Procedure

               Sw  =  the square root of the within mean square

               n   -  the number of replicates at each concentration,
                      assuming an equal number of replicates at all treatment
                      concentrations

               n.,  =  number of replicates in the control

    For example:

        MSD = 2.36 (0.097) [3^(1/4)+(1/4)]  =  2.36 (0.097) (^2/4)

             =  2.36  (0.097)(0.707)

             -• 0.162

 1.11  For this set of data, the minimum difference  between the control mean
 and a concentration mean  that can be detected as statistically significant is
 0.087 mg.   This  represents a decrease  in growth of  24%  from the control.

 1.11.1   If  the data have  not been transformed, the  MSD  (and the percent
 decrease from the control mean that  it represents)  can  be reported as is.

 1.11.2   In  the case where the data have been transformed, the MSD would be in
 transformed units.  In  this case  carry out the following conversion to
 determine the MSD in untransformed units.

 1.11.2.1  Subtract  the  MSD from the  transformed control mean.  Call this
 difference  D.  Next, obtain untransformed values for  the control mean and the
 difference, D.
        MSD,

Where:  MSDt

    Control,
controlu - Du
the minimum significant difference for untransformed data

the untransformed control mean

the untransformed difference
                                      276

-------
                                                          I

1.11.2.2  Calculate the percent reduction from the control that MSD
rpnyp^pntc ac-                        ,                             u
 represents as:
                                    MSDU
          Percent Reduction  = 	 X  100
                                  Control,
                                         u
 1.11.3   An  example  of a conversion of the MSD to untransformed units,  when the
 arc  sine square  root  transformation was  used on the data,  follows:

     Step 1.  Subtract  the MSD from the transformed control  mean.   As an
             example,  assume  the  data in  Table C.I were transformed  by  the arc
             sine square root transformation.   Thus:
                       .
                 0.677 - 0.162 =  0.515

     Step 2.  Obtain  untransformed values  for  the control  mean  (0.677) and  the
             difference  (0.515) obtained  in Step 1 above.  [
                                2
                 [ Sine  (0.677)];  =  0.392
                 [ Sine  (0.515)]2  =  0.243
                                                          I
    Step 3. The  untransformed MSD  (MSDJ is determined by subtracting the
            untransformed values obtained  in Step 2.

                  MSDU  =  0.392 - 0.243  =  0.149
                                                          i
In this case, the MSD would represent a 38.0% decrease in survival
from the control [ (0.149/0. 392) ( 100) ].                    i

2. COMPUTER CALCULATIONS

2.1  This computer program incorporates two analyses:  an analysis of variance
(ANOVA), and a multiple comparison of treatment means with the control mean
(Dunnett's Procedure).  The ANOVA is used to obtain the error value.
Dunnett's Procedure indicates which toxicant concentration means (if any) are
statistically different from the control mean at the 5% level of significance.
The program also provides the minimum difference between the control and
treatment means that could be detected as statistically significant, and tests
the validity of the homogeneity of variance assumption by Bartlett's Test.
The multiple comparison is performed based on procedures described by Dunnett
2.2  The source code for the Dunnett's program is structured into a series of
subroutines, controlled by a driver routine.  Each subroutine has a specific
function in the Dunnett's Procedure, such as data input, transforming the
data, testing for equality of variances, computing p values, and calculating
the one-way analysis of variance.

2.3  The program compares up to seven toxicant concentrations against the
control, and can accommodate up to 50 replicates per concentration.


                                     277

-------
2 4  If the number of replicates at each toxicant concentration and control
are not equal, a t test with Bonferroni's adjustment is performed instead of
Dunnett's Procedure (see Appendix D).

2 5  The program was written in IBM-PC FORTRAN by Computer Sciences
Corporation, 26 W. Martin Luther King Drive, Cincinnati, OH 45268.  A compiled
executable version of the program can be obtained from EMSL-Cincinnati by
sending a written request to EMSL at 3411 Church Street, Cincinnati, OH
45244.

2.6  DATA INPUT AND OUTPUT

261  Reproduction data from a daphnid, Ceriodaphnia dubia, survival and
reproduction test (Table C.6) are used to illustrate the data  input and output
for this program.

          TABLE C.6.  SAMPLE DATA FOR DUNNETT'S  PROGRAM
                      CERIODAPHNIA DUBIA REPRODUCTION DATA
   Replicate     Control
                                      Effluent  Concentration  (%)
1.56
3.12
6.25
                                                                12.5
1
2
3
4
5
6
7
8
9
10
27
30
29
31
16
15
18
17
14
27
32
35
32
26
18
29
27
16
35
13
39
30
33
33
36
33
33
27
38
44
27
34
36
34
31
27
33
31
33
31
10
13
7
7
7
10
10
16
12
2
 2.6.2  Data Input

 2.6.2.1  When the program is entered, the user is asked to select the type of
 data to be entered:

    1. Response proportions, like survival or fertilization proportions.
    2. Counts and measurements, like offspring counts, cystocarp counts'or
       weights.
                                       278

-------
 2.6.2.2   After  the  type  of data  is  chosen,  the  user has  the  following  options:


    1.  Create  a  data file
    2.  Edit  a  data file
    3.  Perform analysis on  existing  data  set             !
    4.  Stop

 2.6.2.3   When Option  1 (Create a data  file)  is  selected  for  counts  and
 measurements, the program  prompts the  user  for  the  follov/ing information:

    1.  Number  of concentrations, including control
    2.  For each  concentration:
       - number  of observations
       - data  for each observation
                                                         i
 2.6.2.4   After  the  data  have been entered,  the  user may  save the file on a
 disk,  and the program returns to the menu (see  below).
                                                         t
 2.6.2.5   Sample data input  is shown in Figure C.I.

 2.6.3   Program Output

 2.6.3.1  When Option 3 (Perform analysis on existing data set) is selected
 from the menu, the  user  is  asked to select the transformation desired,  and
 indicate whether they expect the means of the test groups to be less or
greater than the mean for the control  group (see Figure 0.2).

2.6.3.2  Summary statistics (Figure C.3)  for the raw and transformed data,  if
applicable,  the ANOVA table, results of Bartlett's Test,  the results of the
multiple comparison procedure and the minimum detectable difference are
included in  the program output.
                                     279

-------
                     EMSL Cincinnati  Dunnett Software
                               Version 1.5
      1)    Create a data file
      2)    Edit a data file
      3)    Perform ANOVA on existing data
      4)    Stop
Your choice ? 1
Number of groups, including control ? 5
Number of observations for group  1 ? 10
Enter the data for group  1 one observation at a time,
NO.  1?  27
NO.  2?  30
NO.  3?  29
NO.  4?  31
NO.  5?  16
NO.  6?  15
NO.  7?  18
NO.  8?  17
NO.  9? 14
NO.  10? 27
Number of  observations for group  2   ?  10

 Do you wish  to save the data on disk  ?y
 Disk file  for output  ?  cerio
 Figure C.I.  Sample Data Input for Dunnett's Program for Reproduction
              Data from Table C.6.
                                     280

-------
                      EMSL  Cincinnati  Dunnett  Software
                                Version 1.5
       1)     Create  a data  file
       2)     Edit  a  data  file
       3)     Perform analysis  on  existing data  set
       4)     Stop
Your choice  ?  3


File name  ?  cerio


Available Transformations

      1)    no transform
      2)    square root
      3)    loglO

Your choice ? 1
Dunnett's test as implemented in this program is       i
a one-sided test. You must specify the direction
the test is to be run; that is, do you expect the
means for the test groups to be less than or
greater than the mean for the control group mean.

Direction for Dunnett's test : L=less than, G=greater than ? L
Figure C.2.  Example of  Choosing Option  3 from  the Menu  of the  Dunnett
            Program.
                                   281

-------
   Group
Ceriodaphnia Reproduction Data from Table C.6

         Summary  Statistics  and ANOVA
   Transformation =       None
 n             Mean            s.d.             CV%
1




- control
2
3
4
5*
10
10
10
10
10
22.4000
26.3000
34.6000
31.7000
9.4000
6.9314
8.0007
4.8351
2.9458
3.8930
30.9
30.4
14.0
9.3
41.4
*) the mean for this group is significantly less than the control
   mean at alpha - 0.05 (1-sided) by Dunnett's test
Minimum detectable difference for Dunnett's test = -5.628560
This difference corresponds to -25.13 percent of control
Between concentrations
Sum of squares - 3887.880000 with  4 degrees of freedom.
Error mean square = 31.853333 with 45 degrees of freedom.
Bartlett's test p-value for equality of variances =  .029
Do you wish to restart the program ?
 Figure  C.3.  Example  of Program Output for the Dunnett's Program Using  the
             Reproduction  Data from Table C.6.
                                    282

-------
                                  APPENDIX D

                     T TEST  WITH BONFERRONI'S  ADJUSTMENT
1. The t test with Bonferroni's adjustment is used as an alternative to
Dunnett's Procedure when the number of replicates is not the same for all
concentrations.  This test sets an upper bound of alpha on the overall error
rate, in contrast to Dunnett's Procedure, for which the overall error rate is
fixed at alpha.  Thus, Dunnett's Procedure is a more powerful test.

2. The t test with Bonferroni's adjustment is based on the same assumptions of
normality and homogeneity of variance as Dunnett's Procedure (see Appendix B
for testing these assumptions), and, like Dunnett's Procedure, uses a pooled
estimate of the variance, which is equal to the error value calculated in an
analysis of variance.

3. An example of the use of the t test with Bonferroni's adjustment is
provided below.  The data used in the example are the same as in Appendix C,
except that the third replicate from the 256 fj.g/1 concentration is presumed to
have been lost.  Thus, Dunnett's Procedure cannot be used.  The weight data
are presented in Table D.I.


   TABLE D.I.  FATHEAD MINNOW, PIMEPHALES PROMELAS, LARVAL GROWTH DATA
               (WEIGHT IN MG) USED FOR THE T-TEST WITH BONFERRONI'S
               ADJUSTMENT
  Replicate
Control
                                     NaPCP Concentration f,itq/L)
32
64
128
256
A
B
C
D
Mean(Y,.)
Total (Tf)
0.711
0.662
0.646
0.690
0.677
2.709
0.517
0.501
0.723
0.560
0.575
2..301
0.602
0.669
0.694
0.676
0.660
2.641
0.566
0.612
0.410
0.672
0.565
2.260
i
0.455
0.502
(LOST)
0.254
0.404 .
1.211
                                      283

-------
3.1  One way to obtain  an estimate of the pooled variance is to construct an
ANOVA table including all sums of squares, using the  following formulas:

        Where:     p  = number of effluent concentrations  including the control
                  N  =  the total sample size;   N =
                 n.- = the number of replicates  for concentration i
               SST = ฃ Yi12-G2/N
                     ij
                        Total Sum of Squares
               SSB = T,Ti2/ni-G2/N
                        Between Sum of Squares
               SSW = SST-SSB
                        Within Sum of Squares
   Where:  G   -  The  grand total of all sample observations;   G = ฃ Tฑ
                                                                  1=1

          J.   -  The  total of the replicate measurements  for concentration i

         Yj-    -  The  jth  observation for concentration  i

3.2  For the data in this  example:
            '1
n2 =
ฃ: $;
ฃ:ฃ:
       12
                      '.32
                            = 4
                            13
                            23
                          T3 + T4
Y14 =2.709
Y24 = 2.301
YM = 2.641
   = 2.260
Y54= 1.211

+ T5 = 11
          SSB = ^Ti2/ni-G2/N
               =  6.668 -  (11.122)2 /19

               =  0.158
                                     284

-------
       SST = E YฑJ2-G2/N


            =  6.779 - (11.122)2/19

            =  0.269


       SSW = SST-SSB


            =  0.269 - 0.158                              i

            =  0.111

3.3  Summarize these data in the ANOVA table (Table D.2):

            TABLE D.2.  ANOVA TABLE FOR BONFERRONI'S ADJUSTMENT
  Source          df           Sum of              Mean Square (MS)
                            Squares (SS)               (SS/df)
  Between       p - 1            SSB               S*  = SSB/(p-l)

  Within        N - p            SSW               SJ;  = SSW/(N-p)
  Total         N - 1            SST
3.4  Summarize these data in the ANOVA table (Table D.3):

3.5  To perform the individual comparisons, calculate the t statistic for each
concentration and control combination, as follows:        i
   Where:  Y,-  =  mean for each concentration
                                                          i
           Y1  =  mean for the control                     <

           Su  =  square root of the within mean square
            "                                             i

           n,  =  number of replicates  in the control.

           n,-  =  number of replicates  for concentration i.
                                                          i
                                     285

-------
        TABLE D.3.  COMPLETED ANOVA TABLE FOR THE T-TEST WITH
                    BONFERRONI'S ADJUSTMENT
  Source           df               SS            Mean Square
Between
Within
5-1=4
19 - 5 = 14
0.158
0.111
0.0395
0.0029
  Total           18              0.269
3.6  Table D.4 includes the calculated t values for each concentration and
control combination.

                       TABLE D.4.  CALCULATED T VALUES
                NaPCP
            Concentration         i                   t-
32
64
128
256
2
3
4
5
1.623
0.220
1.782
4.022
3.7  Since the purpose of the test is only to detect a decrease in growth from
the control, a one-sided test is appropriate.  The critical value for the
one-sided comparison (2.510), with an overall alpha level of 0.05, fourteen
degrees of freedom and four concentrations excluding the control, was obtained
from Table D.5.  The mean weight for concentration "i" is considered
significantly less than the mean weight for the control if t,-  is greater than
the critical value.  Since t5 is greater than 2.510,  the 256 jug/L
concentration has significantly lower growth than the control.  Hence the NOEC
and LOEC for growth are 128 /jg/L and 256 ng/L, respectively.
                                      286

-------
CO
ง

CO
O
O
CO
CO
LU
:r
0:2=
o o
   UJ
01
o
CO t_>
JevJcoo
CTiocjioevii—iซMoocoo^-ซnir>eMa>r^.i<*oO'-icnoQii^.ii>in^-rovo^f-eMr—ii—iococx3ooooooi~-r~~ i^i-~.i^r^i~ปi—^r~^

i— o^in^cocococococococN4eviivjesjevje\jcvjeNjcvjcsiioji>Jojcsje>jcsjc\i
in



ป—icvivocO'—ioปi—imt-^ซei-t^ปa-coLnCT>ir)cMii-Hio>-HCM'ooi^-r-~
enoococoooincooji—iocno>oooooooor-~t^t--.i^iซ-. rซซi>~r^voioซ3iD

OOOi^^cococococooooJcsJcvjcvjcvjcvjcvjevjevioJcviojicvjcvjesjcsjcvJCM
to



ior^oooOt-HCvicr>oOCTปi—ii^cot~~i*~t—iCT>o^evjt^^i-cvit—i^-ieviKt-r^oco
ir>i~~co'd-oO'-HrO'—ievjvoou3e\jCT>COCOr^r^l^t-ป.l>ซ.lOlฃ>lOVO'lOlOVOlQVOlO

•*ooirt'*cococococoeซjesjeMcvjesi<>jevjcvicvjeviesjcvie\i'Csjcjevjevicvic>j
•<*•



O^O^h^i—ILOOOOOIOซ*>—ii—CDlOCOCslC>JiPOซ*lฃ>COI^
00'^miocooocsji—iroi~~cvJcoซ*>-HCTปi"~.Lnซ3-cM!—ioerปiooi^vomunซ*
ป—iiooOCTปuncviป-H^CT>oooor^r~r^ioiovovovo^oiOLOiLnLOir>minin

00^^'*cofOcocotoc^Je^Je^JcsJe^le^JC^JCvlc^Jc^Jc^lc>Jc^lcvJ'l^lcvJe>Je^le^l<^^
ro



r-Hin^Hi^iocoooi^c\j>ei-cy><-H^Hir>co'*i^coooocciCT>ocoioeT>coex3
e>juDซ*ซ!j-to^-oปo^csjvoi—lOOtncvJooovOLO^-evji—io^cj>ooป^t^io
oooiunr^coi—*CTปOOOOI— ^ป.^olo^ovOLnlnLnLnmLr)LOlLnซd••!t•ซsJ"'!i••*

f^U3ซ^cocococvie\icซje\icje^eMesjevJc\jcvie\Jcซjevjcvie\icvicvjoJCMOJcsi
CO



CMiDt^voซd-ey>cvjesjvoซet-ti-i—icooococftioซ*'d-ซd-iooO'—-.o>tovo^t-moococr>vocO'—iCTir^unซ*cocvj<—iOCTiCT>ooi~~i~^vo
ซ* CM >-< ซ*i^ericor^usioiOLnLOLn'st-ซ*'!j-<<3-ซa-ซ>i->d-ซd-rococococoro

uiiO^*coc^e\jevieMc\ic^cMeMesiesje\jcvic>jcvjcjCNjcsicsicvicvic^c>jcsJc\i
CM

                                                          \

c\JO>-csjoesj(^CDioeMซd-C3C3coo>voir)tnioaปi—i-ei-oocooocoen
CDซ*ซd-CO>—•lO'd-tD'—iOCOOOOVOซ!l-0>Ji—iOCT!Or>. t^lDLOlOซ*ซd-CO
ocor~-r^o>i^u>u^uiซ^-ซj-ซtrcococococococ\jCNjc\ievjc\jcsjc\ic\ievje\i

o>i^cocoeMCvjesjcsie\icMcvjcMe\jeMCMOJCMc\icvjcMc\icvicMCJC\icvi
i—i



f^COCOI-~>—cr^UP>l-~COO>i—ICT*'—iLOCMOCSi—lซ3-lDO^-Cr>ซ3-OlOCMCTป
oooor-ป.r>.'a-iooiocsJor^iO'*cocsj^HOoocc)i^-ioioioir)LOซd-
t^cot-Hi^.tO'd-cocoeMCMcO'-i'—i •—11—ii—ii—it—10000000000

evj^coesicMe\jeJOJCMOJCMcvi
1—4



^s-oซ*cviioซd-ioo>ซ*coiocO'—iojซd-iooLoc3LOi-lOCOi—iCTปOOI~>-lOLOซS-t!t-COCOCVIC\Jr—i^Hr-HOOOO
coeyjcO"—lOCTiooooooooi^i^i^f^r^i^i^t^r^i— r>. i-~.t-^i^.r^i^r^r^

tOCMevlCVjexJi-Hr—Ir—Ii—Ii—(i—Ii—I i—Ii—ii—I i—I i—I i—Ii—Ii—If— ) i—Ii—I i—I i—I i—I i—I i-H
                                                  287

-------

O Q
CD UJ
1-
00-—
uu
UJ ซ=C

ซa
o o
 	

CO O
—I LO
ซd-a^"d-oiocvicOLocOr—i
F—1    LTjLO^l-OOOOOJCVJCVJi—li——II--
      r~- i— r-~.r~-t~.t--r— r--r--.r--.r--r--.iQioioioioioioiOLr>






      co r~- F-H to r—iiocsjoo<ซa-or--oocO'-H(7>ooocooooo
er>    r—icsc3cncnooeor-.r--r--.<ฃ>ioroojc3C3ey>cooooo^ป-
      t~-r-.r~-iDioioioioiDiDioioioioioiOLnLOLnLOLr>
 11      	
      CMCMCVJCMCMCMCVJCMOdCVJCVICVJCMCVJCMCMCVJOJO'JCSJCV]




CO







t~-    oocncncocor-.!— i— IDIDIDCOOJI—iocncnCT>com
      tojcsjoJC3CT>oor~-ioiD
      ซd-ซd-^3-ซd-ซd-ซa-^i->^-ปd-ซi-ซ^-ซi-ซi-oocoooooco






ซ3-
      nnrococococooocorocofOcocoojcsicvjevjevjojcM






      LOt—lOO-sJ-i—IO11OCO'—r--.LOcyiCTli—IIOCVJOOIOCOCT>
co    cococvjcsjcvii—IT—enoooooooovo


      CMC>JCMCVJC


                                                                              C
                                                                                        a>
                                                                                      LU 4J
                                                                                      00
                                                                                      s:  M
                                                                                          c
                                                                                       S-  O
                                                                                       O •!-
                                                                                      M- 4J
                                                                                          to
                                                                                       E  S-
                                                                                       O)  Cl)
                                                                                       a;  o
                                                                                          o
                                                                                      q-
                                                                                       o t-
                                                                                          o
                                                                                       CO
                                                                                       QJ  %-
                                                                                       O)  O)
                                                                                       S- J2
                                                                                      O -
                                                                                       II  U
                                                    288

-------
                                  APPENDIX E

                          STEEL'S MANY-ONE RANK TEST

1.  Steel's- Many-one Rank Test is a nonparametric test for comparing
treatments with a control.  This test is an alternative to Dunnett's
Procedure, and may be applied to data when the normality assumption has not
been met.  Steel's Test requires equal variances across the treatments and the
control, but it is thought to be fairly insensitive to deviations from this
condition (Steel, 1959).  The tables for Steel's Test require an equal number
of replicates at each concentration.  If this is not the case, use Wilcoxon's
Rank Sum Test, with Bonferroni's adjustment (see Appendix F).

2.  For an analysis using Steel's Test, for each control and concentration
combination, combine the data and arrange the observations in order of size
from smallest to largest.  Assign the ranks to the ordered observations (1 to
the smallest, 2 to the next smallest, etc.).  If ties occur in the ranking,
assign the average rank to the observation.  (Extensive ties would invalidate
this procedure).  The sum of the ranks within each concentration and within
the control is then calculated.  To determine if the response in a
concentration is significantly different from the response in the control, the
minimum rank sum for each concentration and control combination is compared to
the significant values of rank sums given later in this section.  In this
table, k equals the number of treatments excluding the control and n equals
the number of replicates for each concentration and the control.
                                                         i
3.  An example of the use of this test is provided below.,  The test employs
reproduction data from a Ceriodaphnia dubia 7-day, chronic test.  The data are
listed in Table E.I.  Significant mortality was detected via Fisher's Exact
Test in the 50% effluent concentration.  The data for this concentration is
not included in the reproduction analysis.

     TABLE E.I.  EXAMPLE OF STEEL'S MANY-ONE RANK TEST:  DATA FOR THE
                 .DAPHNID, CERIODAPHNIA DUBIA, 7-DAY CHRONIC TEST

Effluent
Concentration
Control
3%
6%
12%
25%
50%







i


Replicate
1
20
13
18
14
9
0
2
26
15
22
22
0
0
3
26
14
13
20
9
0
4
23
13
13
23
7
0
5
24
23
23
20
6
0
6
27
26
22
23
10
0
7
26
0
20
25
12
0
8
I
23
25
22
24
14
0
9
27
26
23
25
9
0
10
24
27
22
21
13
0
No.
Live
Adults
10
9
10
10
8
0
                                      289

-------
4.  For each control and concentration combination, combine the data and
arrange the observations in order of size from smallest to largest.  Assign
ranks  (1, 2, 3,..., 16) to the ordered observations (1 to the smallest, 2 to
the next smallest, etc.)-  If ties occur in the ranking, assign the average
rank to each tied observation.

5.  An example of assigning ranks to the combined data for the control and 3%
effluent concentration is given in Table E.2.  This ranking procedure is
repeated for each control and concentration combination.  The complete set of
rankings is listed in Table E.3.  The ranks are then summed for each effluent
concentration, as shown in Table E.4.
       TABLE E.2.  EXAMPLE OF STEEL'S MANY-ONE RANK TEST:  ASSIGNING
                   RANKS TO THE CONTROL AND 3% EFFLUENT CONCENTRATION
         Rank
Number of Young
   Produced
Control or % Effluent
1
2.5
2.5
4
5
6
8
8
8
10.5
10.5
12
15
15
15
15
15
19
19
19
0
13
13
14
15
20
23
23
23
24
24
25
26
26
26
26
26
27
27
27
3
3
3
3
3
Control
Control
Control
3
Control
Control
3
Control
Control
Control
3
3
Control
Control
3
6.  For this set of data, determine if the reproduction in any of the effluent
concentrations is significantly lower than the reproduction by the control
organisms.  If this occurs, the rank sum at that concentration would be
significantly lower than the rank sum of the control.  Thus,  compare the rank
sums for the reproduction of each of the various effluent concentrations with
some "minimum" or critical rank sum, at or below which the reproduction would
be considered to be significantly lower than the control.  At a probability

                                      290

-------
TABLE E.
Repl icate
(Organism)
1
2
3
4
5
6
7
8
9
10
20
26
26
23
?4
27
26
23
27
24
3.
Control1

(6,4.5
(15,17
(15,17
(8,11.
(10.5,
(19,19
(15,17
(8,11.
(19,19
(10.5,

,3,11)
,17,17)
,17,17)
5,8.5,12
14.5,12,
.5,19.5,
,17,17)
5,8.5,12
.5,19.5,
14.5,12,




.5)
14.5)
19.5)
.5)
19.5)
14.5)

13
15
14
13
23
26
0
25
26
27
TABLE OF RANKS
Effluent Concentration (%)
3
(2.5)
(5)
(4)
(2.5)
(8)
(15)
(1)
(12)
(15)
(19)

18
22
13
13
23
22
20
22
23
22
6
(3)
(7.5)
(1.5)
(1.5)
(11.5)
(7.5)
(4.5)
(7.5)
(11.5)
(7.5)

14
22
20
23
20
23
25
24
25
21
12
(1)
(6)
(3)
(8.5)
(3)
(8.5)
(14.5)
(12)
(14.5)
(5)
25
9 (5)
0 (1)
9 (5)
7 (3)
6 (2)
10 (7)
12 (8)
14 (10)
9 (5)
13 (9)
  1  Control  ranks are given  in the order of the concentration with which  they
    were ranked.
                           TABLE E.4.   RANK SUMS
                    Effluent                      Rank Sum
                 Concentration
                       3                             84
                       6                             64
                      12                             76
                      25                             55
level of 0.05, the critical rank in a test with four concentrations and ten
replicates is 76 (see Table E.5, forR=4).

7.  Comparing the rank sums in Table E.4 to the appropriate critical rank, the
6%, 12% and 25% effluent concentrations are found to be significantly
different from the control.  Thus the NOEC and LOEC for reproduction are 3%
and 6%, respectively.
                                      291

-------
TABLE E.5.  SIGNIFICANT VALUES OF RANK SUMS: JOINT CONFIDENCE
            COEFFICIENTS OF 0.95 (UPPER) and 0.99 (LOWER) FOR
            ONE-SIDED ALTERNATIVES (Steel, 1959)
k = number of treatments (excludina
'
I 2
4 1 11
5 18

6

7

8

9

10

11

12

13

15
27
23
37
32
49
43
63
56
79
71
97
87
116
105
138
125
14 161
1 147
15 1 186
1 170
16 1 213
1
17

18

19

20

196
241
223
272
252
304
282
339
315
3
10
17
--
26
22
36
31
48
42
62
55
77
69
95
85
114
103
135
123
158
144
182
167
209
192
237
219
267
248
299
278
333
310
4
10
17
—
25
21
35
30
47
41
61
54
76
68
93
84
112
102
133
121
155
142
180
165
206
190
234
217
264
245
296
275
330
307
5
10
16
—
25
21
35
30
46
40
60
53
75
67
92
83
111
100
132
120
154
141
178
164
204
188
232
215
262
243
294
273
327
305
6
10
16
	
24

34
29
46
40
59
52
74
66
91
82
110
99
130
119
153
140
177
162
203
187
231
213
260
241
292
272
325
303
7
--
16
_ —
24
—
34
29
45
40
59
52
74
66
90
81
109
99
129
118
152
139
176
161
201
186
229
212
259
240
290
270
323
301
control )
8
--
16
— _
24
- -
33
29
45
39
58
51
73
65
90
81
108
98
129
117
151
138
175
160
200
185
228
211
257
239
288
268
322
300
9
--
15
__
23

33
29
44
39
58
51
72
65
89
80
108
98
128
117
150
137
174
160
199
184
227
210
256
238
287
267
320
299
                           292

-------
                                  APPENDIX F

                            WILCOXON RANK SUM TEST
1.  Wilcoxon's Rank Sum Test is a nonparametric test, to be used as an
alternative to Steel's Many-one Rank Test when the number of replicates are
not the same at each concentration.  A Bonferroni's adjustment of the pairwise
error rate for comparison of each concentration versus the control is used to
set an upper bound of alpha on the overall error rate, in contrast to Steel's
Many-one Rank Test, for which the overall error rate is fixed at alpha.  Thus,
Steel's Test is a more powerful test.

2.  An example of the use of the Wilcoxon Rank Sum Test is provided in Table
F.I.  The data used in the example are the same as in Appendix E, except that
two males are presumed to have occurred, one in the control and one in the 25%
effluent concentration.  Thus, there is unequal replication for the
reproduction analysis.

3.  For each concentration and control combination, combine the data and
arrange the values  in order of size, from smallest to largest.  Assign ranks
to the ordered observations (a rank of 1 to the smallest, 2 to the next
smallest, etc.).   If ties in rank occur, assign the average rank to each tied
observation.

     TABLE F.I.   EXAMPLE OF WILCOXON'S RANK SUM TEST:  DATA FOR THE
                  DAPHNID, CERIODAPHNIA DUBIA,  7-DAY CHRONIC TEST

Effluent
Concentration
Cont
3%
6%
12%
25%
50%








l

Replicate
1
M
13
18
14
9
0
2
26
15
22
22
0
0
3
26
14
13
20
9
0
4
23
13
13
23
7
0
5
24
23
23
M
6
0
6
27
26
22
23
10
0
7
26
0
20
25
12
0
8
23
25
22
24
14
0
9
27
26
23
25
i 9
0
10
24
27
22
21
13
0
No.
Live
Adults
10
9
10
10
8
0
 4.   An example of assigning ranks to the combined data for the control  and
 3% effluent concentration is given in Table F.2.   This ranking procedure is
 repeated for each of the three remaining control  versus test concentration
 combinations.  The complete set of ranks is listed in Table F.3.   The ranks
 are then summed for each effluent concentration,  as shown in Table F.4.

 5.   For this set of data, determine if the reproduction in any of the effluent
 concentrations is significantly lower than the reproduction by the control

                                       293

-------
         TABLE F.2.   EXAMPLE OF WILCOXON'S RANK SUM TEST:   ASSIGNING
                     RANKS TO THE CONTROL AND EFFLUENT CONCENTRATIONS
          Rank      Number of Young         Control  or % Effluent
                       Produced
1
2.5
2.5
4
5
7
7
7
9.5
9.5
11
14
14
14
14
14
18
18
18
0
13
13
14
15
23
23
23
24
24
25
26
26
26
26
26
27
27
27
3
3
3
3
3
Control
Control
3
Control
Control
3
Control
Control
Control
3
3
Control
Control
3
organisms.  If this occurs, the rank sum at that concentration would be
significantly lower than the rank sum for the control.  Thus, compare the rank
sums for the reproduction of each of the various effluent concentrations with
some "minimum" or critical rank sum, at or below which the reproduction would
be considered to be significantly lower than the control.  At a probability
level of 0.05, the critical rank in a test with four concentrations and nine
replicates in the control is 72 for those concentrations with ten replicates,
and 60 for those concentrations with nine replicates (see Table F.5, for K =
4).

6.  Comparing the rank sums in Table F.4 to the appropriate critical rank, the
6%, 12% and 25% effluent concentrations are found to be significantly
different from the control.  Thus, the NOEC and LOEC for reproduction are 3%
and 6%,  respectively.
                                     294

-------
                     TABLE  F.3.  TABLE OF  RANKS
Replicate Control1
(Organism)
1
2
3
4
5
6
7
8
9
10
M
26 (14,16,15,16)
26 (14,16,15,16)
23 (7,10.5,6.5,11.5)
24 (9.5,13.5,10,13.5)
27 (18,18.5,17.5,18.5)
26 (14,16,15,16)
23 (7,10.5,6.5,11.5)
27 (18,18.5,17.5,18.5)
24 (9.5,13.5,10,13.5)
13
15
14
13
23
26
0
25
26
27
Effluent
3
(2.5)
(5)
(4)
(2.5)
(7)
(14)
(1)
(11)
(14)
(18)

18
22
13
13
23
22
20
22
23
22
6
(3)
(6.
(1.
(1.
(10
(6.
(4)
(6.
(10
(6.
Concentration (%)


5)
5)
5)
.5)
5)

5)
.5)
5)
1
14
22
20
23
M
23
25
24
25
:21
12
(1)
(4)
(2)
(6.

(6.
(12




5)

5)
.5)
(10)
(12
(3)
.5)


9
0
9
7
6
10
12
14
9
13
25
(b)
(1)
(5)
(3)
(2)
(7)
(8)
(10)
(5)
(9)

Control ranks are given in the order of the concentration! with which they
 were  ranked.
                          TABLE F.4.   RANK SUMS
            Eff1uent
         Concentration
Rank Sum
No. of
Replicates
Critical
Rank Sum
               3
               6
              12
              25
   79
   57
   58
   55
  10
  10
   9
  10
   72
   72
   60
   72
                                  295

-------
TABLE F.5.  CRITICAL VALUES FOR WILCOXON'S RANK SUM TEST WITH
            BONFERRONI'S ADJUSTMENT OF ERROR RATE FOR COMPARISON
            OF "K" TREATMENTS VERSUS A CONTROL FIVE PERCENT CRITICAL
            LEVEL (ONE-SIDED ALTERNATIVE: TREATMENT CONTROL)
K No. Reolicates No. of Reolicates Per Effluent
in Control

1 3
4
5
6
7
8
9
10
2 3
4
5
6
7
8
9
10
3 3
4
5
6
7
8
9
10

3
6
6
7
8
8
9
10
10
„_
--
6
7
7
8
8
9

--
--
6
7
7
7
8

4
10
11
12
13
14
15
16
17
..
10
11
12
13
14
14
15

10
11
11
12
13
13
14

5
16
17
19
20
21
23
24
26
15
16
17
18
20
21
22
23

16
17
18
19
20
21
22

6
23
24
26
28
29
31
33
35
22
23
24
26
27
29
31
32
21
22
24
25
26
28
29
31

7
30
32
34
36
39
41
43
45
29
31
33
34
36
38
40
42
29
30
32
33
35
37
39
41

8
39
41
44
46
49
51
54
56
38
40
42
44
46
49
51
53
37
39
41
43
45
47
49
51
Concentration

9
49
51
54
57
60
63
66
69
47
49
52
55
57
60
62
65
46
48
51
53
56
58
61
63

10
59
62
66
69
72
72
79
82
58
60
63
66
69
72
75
78
57
59
62
65
68
70
73
76
                                296

-------
TABLE F.5.  CRITICAL VALUES FOR WILCOXON'S RANK SUM TEST WITH BONFERRONI'S
            ADJUSTMENT OF ERROR RATE FOR COMPARISON OF '"K" TREATMENTS
            VERSUS A CONTROL FIVE PERCENT CRITICAL LEVEL (ONE-SIDED
            ALTERNATIVE:  TREATMENT CONTROL) (CONTINUED)
K No. Replicates No. of Reol
in Control

4 3
4
5
6
7
8
9
10
5 3
4
5
6
7
8
9
10
6 3
4
5
6
7
8
9
10
7 3
4
5
6
7
8
9
10

3
...
--
--
6
6
7
7
7
..
_-
--
--
6
6
7
7
..
--
--
_-
6
6
6
7
._
.--
--
--
--.
6
6
7

4
...
--
10
11
12
12
13
14
..
--
10
11
11
12
13
13
— _
--
10
11
11
12
12
13
..
--
--
10
11
11
12
13
icates Per Effl

5
..
15
16
17
18
19
20
21
— —
15
16
17
18
19
20
21
_ —
15
16
16
17
18
19
20
„_
--
15
16
17
18.
19
20

6
21
22
23
24
26
27
28
30
_ _
22
23
24
25
27
28
29
_ _
21
22
24
25
26
27
29
..
21
22
23
25
26
27
28

7
28
30
31
33
34
36
38
40
28
29
31
32
34
35
37
39
28
29 ,
30 '
32
33
35
37
38
..
29
30
32
33
35
36
38
uent

8
37
38
40
42
44
46
48
50
36
38
40
42
43
45
47
49
36
38
39
41
43
:45
47
:49
36
37
39
41
43
44
46
48
Concentration

9
46
48
50
52
55
57
60
62
46
48
50
52
54
56
59
61
45
47
49
51
54
56
58
60
45
47
49
51
53
55
58
60

10
56
59
61
64
67
69
72
75
56
58
61
63
66
68
71
74
56
58
60
63
65
68
70
73
56
58
60
62
65
67
70
72
                                    297

-------
TABLE F.5.  CRITICAL VALUES FOR WILCOXON'S RANK SUM TEST WITH BONFERRONI'S
            ADJUSTMENT OF ERROR RATE FOR COMPARISON OF "K" TREATMENTS
            VERSUS A CONTROL FIVE PERCENT CRITICAL LEVEL (ONE-SIDED
            ALTERNATIVE:  TREATMENT CONTROL) (CONTINUED)
K No. Replicates No. of Reolicate Per
in Control

8 3
4
5
6
7
8
9
10
9 3
4
5
6
7
8
9
10
10 3
4
5
6
7
8
9
10

3

--
--
--
--
6
6
6
..
--
--
--
--
--
6
6
„.
--
--
--
--
--•
6
6

4

--
--
10
11
11
12
12

--
--
10
10
11
11
12
..
--
--
10
10
11
11
12

5

--
15
16
17
18
19
19

--
15
16
17
18
18
19
..
--
15
16
16
17
18
19

6

21
22
23
24
25
27
28

21
22
23
24
25
26
28
..
21
22
23
24
25
26
27
Effluent Concentration

7

29
30
31
33
34
36
37

28
30
31
33
34
35
37
..
28
29
31
32
34
35
37

8
36
37
39
40
42
44
46
48

37
39
40
42
44
46
47

37
38
40
42
43
45
47

9
45
47
49
51
53
55
57
59
45
46
48
50
52
55
57
59
45
46
48
50
52
54
56
58

10
55
57
59
62
64
67
69
72
55
57
59
62
64
66
69
71
55
57
59
61
64
66
68
71
                                   298

-------
                                  APPENDIX G

                              FISHER'S EXACT TEST

1.  Fisher's Exact Test (Finney, 1948; Pearson and Hartley, 1962) is a
statistical method based on the hypergeometric probability distribution that
can be used to test if the proportion of successes is the same in two
Bernoulli (binomial) populations.  When used with the Ceriodaphnia dubia data,
it provides a conservative test of the equality of any two survival
proportions assuming only the independence of responses from a Bernoulli
population.  Additionally, since it is a conservative test, a pair-wise
comparison error rate of 0.05 is suggested rather that ati experiment-wise
error rate.

2.  The basis for Fisher's Exact Test is a 2x2 contingency table.  However, in
order to use this table the contingency table must be arranged in the format
shown in Table G.I.  From the 2x2 table, set up for the control and the
concentration you wish to compare, you can determine statistical significance
by looking up a value in the table provided later in this section.

                       TABLE G.I.  FORMAT FOR CONTINGENCY TABLE
                              Number of
                         Successes
            Failures
                Number of
              Observations
            Row 1

            Row 2
            Total
  a

  b
A - a

B - b
A

B
a + b    [(A + B) - a - b]
                  A + B
3.  Arrange the table so that the total number of observations for row one is
greater than or equal to the total for row two (A > B).  Categorize a success
such that the proportion of successes for row one is greater than or equal to
the proportion of successes for row two (a/A > b/B).  For the Ceriodaphnia
dubia survival data, a success may be 'alive' or 'dead', whichever causes
a/A > b/B.  The test is then conducted by looking up a value in the table of
significance levels of b and comparing it to the b value given in the
contingency table.  The table of significance levels of b is Table G.5.   Enter
Table G.5 in the section for A, subsection for B, and the line for a.  If the
b value of the contingency table is equal to or less than the integer in the
column headed 0.05 in Table G.5, then the survival  proportion for the effluent
concentration is significantly different from the survival  proportion for the
control.  A dash or absence of entry in Table G.5 indicates that no
contingency table in that class is significant.
                                     299

-------
4.  To illustrate Fisher's Exact Test, a set of survival data (Table G.2) from
the daphnid, Ceriodaphnia dubia, survival and reproduction test will be used.
                 TABLE G.2.  EXAMPLE OF FISHER'S EXACT TEST:
                             CERIODAPHNIA DUBIA MORTALITY DATA
        Effluent
      Concentration (%)
No. Dead
Total1
Control
1
3
6
12
25
1
0
0
0
0
10
9
10
10
10
10
10
  1  Total number of live adults at the beginning of the test.
5.  For each control and effluent concentration construct a 2x2 contingency
table.

6.  For the control and effluent concentration of 1% the appropriate
contingency table for the test is given in Table G.3.

7.  Since 10/10 > 8/9, the category  'alive' is regarded as a success.  For A =
10, B = 9 and, a = 10, under the column headed 0.05, the value from Table G.5
is b = 5.  Since the value of b (b = 8) from the contingency table
(Table G.3), is greater than the value of b (b = 5) from Table G.5, the test
concludes that the proportion of survival is not significantly different for
the control and 1% effluent.

8.  The contingency tables for the combinations of control and effluent
concentrations of 3%, 6%, 12% are identical to Table G.3.  The conclusion of
no significant difference in the proportion of survival for the control and
the level of effluent would also remain the same.

9.  For the combination of control and 25% effluent, the contingency table
would be constructed as Table G.4.  The category 'dead' is regarded as a
success, since 10/10 > 1/9.  The b value  (b = 1) from the contingency table
(Table G.4) is less than the b value (b = 5) from the table of significance
levels of b (Table G.5).
                                      300.

-------
Thus, the percent mortality for 25% effluent is significantly greater than the
percent mortality for the control.  Thus, the NOEC and LOEC for survival  are
12% and 25%, respectively.

           TABLE G.3.  2x2 CONTINGENCY TABLE FOR CONTROL AND 1% EFFLUENT



1% Effluent
Control
Total
Number

Alive
10
8
18
of

Dead
0
1
1
i
Number of
Observations
10
9
19
'
         Table G.4.  2x2 CONTINGENCY TABLE FOR CONTROL AND 25% EFFLUENT

25% Effluent
Control
Total
Number
Dead
10
1
11
of
Alive
0
8
8
Number of
Observations
10
9
| 19
                                     301

-------
             TABLE  6.5.   SIGNIFICANT  LEVELS OF B:  VALUES OF  B  (LARGE  TYPE)
                            AND CORRESPONDING PROBABILITIES (SMALL  TYPE)1



A*3 B=3


A=4 B=4
3

A*5 B-5

4

3
2


Ar6 B=6


5


4

3

2

A=7 B*7



6



5








a


3


4
4

5
4
5
4
5
5


6
5
4
6
5
4
6
5
6
5
6

7
6
5
4
7
6
5
4
7
6
5
7
6
5
7
6
7
Probability
0-05

0 -050


1 -014
0 429

1 -02*
0 424.
1 -040
0 440
0 418
0 -049


2 430
1 440
0 430
1 416 +
0 413
0 445 +
1 433
0 424
0 412
0 448
0 438

3 435-
1 415-
1 410+
0 435 •
2 421
1 425+
0 -016
0 449
2 445+
1 445+
0 427
1 424
0 415+
0 446+
0 408
0 433
0 428
0-025

	


1 414
	

1 424
1 424
0 408
	
0 418
	


1 408
0 408
	
0 415 +
0 413
	
0 -005-
0 424
0 412
	
	

2 410+
1 415"
0 410+
	
2 421
0 404
0 418
	
1 410+
0 408
	
1 424
0 415+
	
0 408
	
	
0-01

—


—
—

0 -004
—
0 408
	
	
	


1 -008
0 -COB
	
0 402
	
	
0 405-
	
	
	
	

1 -002
0 -002
	
	
1 405-
0 -004
	
	
0 -001
0 -008
	
	
0 403
	
0 -008
	
	
0-005

—


—
—

0 -004
	
	
	
	
	


0 401
	
	
0 402
	
	
0 -006-
	
	
	
	

1 -002
0 402
	
	
1 -005"
0 -004
	
	
0 401
	
	
0 403
	
	
	
	
	


A=8 B=8




7



6



5



4


3

2
A=9 B=9





8




7




6




a

8
7
6
5
4
8
7
6
5
8
7
6
5
8
7
6
5
8
7
6
8
7
8
9
8
7
6
5
4
9
8
7
6
5
9
8
7.
6
5
9
8
7
6
5
Probability
0-05
4 -038
2 420
1 -020
0 -013
0 438
3 426
2 -035"
1 -032
0 419
2 -015'
1 -016
0 -009
0 428
2 435'
1 -032
0 416
0 -044
1 -018
0 410 +
0 -030
0 -006
0 424
0 -022
5 -041
3 425-
2 -028
1 -025-
0 -015-
0 -041
4 -029
3 443
2 444
1 436
0 -020
3 -019
2 -024
1 -020
0 -010+
0 429
3 -044
2 -047
1 -035-
0 417
0 442
0-025
3 -013 '
2 -020
1 -020
0 -013
	
2 -007
1 -oos
0 -006
0 -019
2 -015-
1 -016
0 409
	
1 -007
0 -005"
0 -016
	
1 -018
0 -010 +
	
0 -006
0 -024
0 -022
4 415-
3 -025-
1 -008
1 -025-
0 -015-
	
3 -009
2 413
1 412
0 -007
0 420
3 -019
2 -024
1 -020
0 -010+
	
2 -011
1 411
0 -006
0 417
	
0-01
2 -003
1 -005 +
0 -003
—
—
2 -007
1 -009
0 -006
	
1 -003
0 -002
0 -009
	
1 -007
0 -DOS"
	
	
0 -002
	
	
0 -006
	
	
3 -005"
2 -008
1 -008
0 -005-
	
	
3 -009
1 -003
0 -002
0 -007
	
2 -005
1 -006
0 403
	
	
1 -002
0 -001
0 -006
	
	
0-005
2 -003
0 -001
0 403
	
	
1 -001
0 -001
—
—
1 -003
0 -002
	
	
0 -001
0 -oos'
—
—
0 -002
—
—
—
—
—
3 -005"
1 -002
0 -001
0 -005"
	
	
2 -002
1 -003
0 -002
	
	
2 -005-
0 -001
0 -003
	
	
1 -002
0 -001
	
	
	
The table shows:(1) In bold type, for given a, A and B, the value of b (Ca) which is  just significant
at the probability level quoted  (one-tailed test); and (2)  In small type,  for given A, B and r = a +
b, the exact probability (if there is independence) that b is equal to or less than the integer shown
in bold type.  From Pearson and  Hartley (1962).
                                                 302

-------
TABLE G.5.
SIGNIFICANT  LEVELS  OF  B:  VALUES  OF  B  (LARGE  TYPE)  AND
CORRESPONDING PROBABILITIES (SMALL TYPE)1 (CONTINUED)



A=9 B=5



4



3


2


A=10 B=10






9





8





7





6




5





a

9
8
7
6
9
8
7
6
9
8
7
9


10
9
8
7
6
5
4
10
9
8
7
6
5
10
9
8
7
6
5
10
9
8
7
6
5
10
9
8
7
6
10
9
8
7
6
Probability

0-05
2 -027
1 -023
0 -010+
0 -028
1 -014
0 -007
0 -021
0 -049
1 -045+
0 -018
0 -045+
0 -018


6 -043
4 -029
3 -035-
2 -035-
1 -029
0 -016
0 -043
5 -033
4 -050-
2 -019
1 -015-
1 -040
0 -022
4 -023
3 -032
2 -O31
1 -023
0 -Oil
0 -029
3 -015-
2 -018
1 -013
1 -038
0 -017
0 -041
3 -036
2 -033
1 -024
0 -010+
0 -026 '
2 -022
1 -017
1 -047
0 -019
0 -042-
0-025
1 -005-
1 -023
0 -010+
	
1 -014
0 -007
0 -021
—
0 -005-
0 -018
	
0 -018


5 -016
3 -oio-
2 -012
1 -010"
0 -006+
0 -016
—
4 -011
3 -017
2 -019
1 -015-
0 -008
0 -022
4 -023
2 -009
1 -O08
1 -023
0 -Oil
	
3 -015-
2 -018
1 -013
0 -006
0 -017
	
2 -008
1 -008
1 -024
0 -010+
	
2 -022
1 -017
0 -007
0 -019
	
0-01
1 -005-
0 -003
—
—
0 -001
0 -007
—
0 -005
0 -005-
—
—
—


4 -005+
3 -010
1 -003
1 -010-
0 -005+
—
—
3 -003
2 -oos-
1 -004
0 -002
. 0 -008
	
3 -007
2 -009
1 -008
0 -004
—
	
2 -003
1 -004
0 -002
0 -003
	
	
2 -003
1 -oos
0 -003 .
- —
—
1 -004
0 -ooz
0 -007
- —
—
0-005
1 -006-
0 -003
	
	
0 -001
—
—
—
0 -005-
_
	
	


3 -002
2 -003
1 -003
0 -002
	
	
	
3 -003
2 -005-
1 -004
0 -002 .
	
	
2 -002
1 -002
0 -001
0 -004
	
	
2 -003
1 -004
0 -002
	
	
. 	
1 -001
0 -001
0 -003
—
—
1 -004
0 -002
	
	 .
	



A=10 B=4



3


2



A=11 B=11







10






9






8






7





6



a.

10
9
8
7
10
9
8
10
9


11
10
9
8
7
6
5
4
11
10
9
8
7
6
5
11
10
9
8
7
6
5
11
10
9
8
7
6
5
11
10
9
8
7
6
11
10
9
Probability

0-05
1 -Oil
1 -041
,0 -016-
0 -035-
1 -038
0 -014
0 -035-
0 -015+
0 -045+


7 -04S+
5 -032
4 -040
3 -043
2 -040
1 -032
0 -018
0 -045+
6 -035+
4 -021
3 -024
2 -023
1 -017
1 -043
0 -023
5 -026
4 -038
3 -040
ซ 2 -035-
1 -025 "
0 -012
0 -030
4 -018
3 -024
2 -022
1 -015-
1 -037
0 -017
0 -040
4 -043
3 -047
2 -039
1 -026-
0 -010+
0 -025-
3 -029
2 -028
1 -018
0-025
1 -011
0 -005-
0 -015-
	
0 -003
0 -014
	
0 -015+
	


6 -018
4 -012
3 -015"
2 -015-
1 -012
0 -006
0 -018
	
5 -012
4 -021
3 -024
2 -023
1 -017
0 -009
0 -023
4 -008
3 -012
2-012
1 -OO9
1 -02S-
0 -012
' 	
4 -018
3 -024
2 -022
1 -015-
0 -007
0 -017
	
3 -011
2 -013
1 -009
1 -025-
0 -010+
0 -025-
2 006
1 -005+
1 -018
0-01
0 -001
0 -005"
—
—
0 -003
—
—
—
—


5 -006
3 -004
2 -004
1 -004
0 -002
0 -006
	
	
4 -004
3 -007
2 -007
1 -006
0 -003
0 -009
	
4 -008
2 -003
1 -003
1 -009
0 -004
	
	
3 -005"
2 -009
1 -005-
0 -002
0 -007
	
	
2 -002
1 002
1 -009
0 -004
	
'. 	
2 -006
1 -005+
0 -002
0-005
0 -001
0 -005-
—
—
0 -003
	
	
	
	


4 -002
3 -004
2 -004
1 -004 .
0 -002
	
	
	
4 -004
2 -002
1 -002
0 -001
0 -003
	
'. 	
3 -002
2 -003
1 -003
0 , -001
0 -004
	
	
3 -005"
1 -001
1 -005"
0 -002
	
	
	
2 -002
1 -002
0 -001
0 -004
—
—
1 -001
0 -001
0 -002
                                    303

-------
TABLE G.5.  SIGNIFICANT LEVELS OF B: VALUES OF B (LARGE TYPE)
            AND CORRESPONDING PROBABILITIES (SMALL TYPE)1
            (CONTINUED)


.
As11 B=6


5




4



3


2



A*12 B=12








11







10







9





o:

8
7
6
11
10
9
8
7
11
10
9
8
11
10
9
11
10


12
11
10
9
8
7
6
5
4
12
11
10
9
8
7
6
5
12
11
10
9
8
7
6
5
12
11
10
9
8
Probability

0-05
1 -043
0 417
0 -O37
2 -018
1 -013
1 -038
0 -013
0 428
1 -009
1 -033
0 -on
0 428
1 433
0 .011
0 -027
0 .013
0 .038


8 .047
6 -034
5 445-
4 450-
3 450"
2 445-
1 -034
0 .018
0 447
7 437
5 -024
4 429
3 -030
2 -029
1 419
1 445-
0 424
6 429
5 443
4 448
3 -046
2 438
it
0 412
0 430
5 421
4 429
3 428
2 424
1 418
0-025
0 407
0 417
	
2 418
1 413
0 405-
0 413
	
1 409
0 404
0 -011
	
0 .003
0 411
	
0 413
	


7 419
5 414
4 -018
3 -020
2 418
1 -014
0 -007
0 419
	
6 414
5 424
3 410+
2 -009
1 407
1 419
0 409
0 424
5 410
4 -01 6+
3 417
2 415"
1 410+
0 405*
0 412
	
5 421
3 -009
2 408
2 424
1 -016
0-01
0 407
	
	
1 403
0 -001
0 -006-
	
	
1 409
0 404
	
	
0 403
	
	
	
	


6 407
4 -DOS"
3 -006
2 -008
1 40B-
0 402
0 -007
	
	
5 -cos-
4 -008
2 -003
2 -009
1 407
0 -003
0 -009
	
5 -010"
3 405-
2 -005-
1 -004
0 -002
0 405-

	
4 -008
3 -009
2 408
1 -006
0 402
0-005
_
—
—
1 -003
0 401
0 405-
	
	
0 401
0 -004
	
	
0 403
	
	
	
	


5 402
4 -oos-
2 -002
1 -001
1 405-
0 -002
	
	
	
5 405-
3 -002
2 -003
1 -002
0 -001
0 -003
	
	
4 -003
3 -005-
2 -005-
1 -004
0 -002
0 40S-

	
3 -002
2 -002
1 -002
0 -001
0 -002



A=12 B=9


8






7






6






5





4




3



2



A=13 B=13






a

7
6
5
12
11
10
9
.8
7
6
12
11
10
9
8
7
6
12
11
10
9
8
7
6
12
11
10
9
8
7
12
11
10
9
8
12
11
10
9
12
11


13
12
11
10
9
8
Probability

0-05
1 -037
0 -017
0 439
5 449
3 418
2 -015+
2 -040
1 -025 "
0 410+
0 424
4 438
3 -038
2 429
1 -017
1 -040
0 -018
0 -034
3 -025"
2 -022
1 -013
1 -032
0 411
0 426-
0 4SO-
2 -oi$-
1 -010-
1 428
0 -009
0 420
0 441
2 -050
1 427
0 -008
0 -019
0 -038
1 -029
0 -009
0 422
0 -044
0 411
0 433


9 448
7 -037
6 -048
4 -024
3 424
2 -021
0-025
0 407
0 -017
—
4 414
3 418
2 415+
1 410-
1 -025-
0 -010+
0 424
3 -009
2 -010-
1 406
1 -017
0 -007
0 -016
	
3 425-
2 422
1 -013
0 405"
0 -Oil
0 -02B"
	
2 -015
1 410-
0 -003
0 -009
0 -020
	
1 407
0 -003
0 -008
0 419
	
0 402
0 -009
0 -022
	
0 411
	


8 420
6 -015+
5 -021
4 -024
3 -024
2 -021
0-01
0 -007
—
	
.3 -004
2 404
1 -003
1 -010-
0 -004
	
	
3 -009
2 -010-
1 -006
0 -002
0 -007
—
—
2 405-
1 -004
0 -002
0 -005-
	
	
	
1 -002
1 -010"
0 -003
0 -009
	 •
	
1 -007
0 403
0 -008
	
	
0 -002
0 -009
	
	
	
	


7 -007
5 -006
4 408
3 -008
2 -008
1 -006
0-005
	
—
—
3 -004
2 -004
1 -003
0 -001
0 -004
—
—
2 -002
1 -002
0 401
0 -002
	
	
—
2 -005-
1 -004
0 -002
0 405"
	
	
	
1 -002
0 -001
0 -003
	
	
	
0 -001
0 -003
	
	
	 '
0 -002
	
	
	
	
	


6 -003
4 -002
3 -002
2 -002
1 -002
0 -001
                             304

-------
TABLE 6.5.  SIGNIFICANT  LEVELS  OF  B:   VALUES  OF  B  (LARGE  TYPE)
            CORRESPONDING PROBABILITIES  (SMALL TYPE)r
            (CONTINUED)
AND

A=13 B=13



12








11








10








9








8







7

at
7
6
5
4
13
12
11
10
9
8
7
6
5
13
12
11
10
9
8
7
6
5
13
12
11
10
9
8
7
6
5
13
12
11
10
9
8
7
6
5
13
12
11
10
9
8
7
6
13
12
Probability
0-05
2 -048
1 -037
0 -020
0 -048
8 -039
6 -027
5 -033
4 -036
3 -034
2 -029
1 -020
1 -048
0 -024
7 -031
6 -048
4 -021
3 -021
3 -060-
2 -040
1 -027
0 -O13
0 -030
6 -024
5 -03S-
4 -037
3 -033
2 -028
1 -017
1 -038
0 -017
0 -038
5 -017
4 -023
3 -022
2 -017
2 -040
1 -02B-
0'-010+
0 -023
0 -049
5 -042
4 -047
3 -041
2 -029
1 -017
1 -037
0 -015-
0 -032
4 -031
3 -031
0-025
1 -016 +
0 -007
0 -020
—
7 -01 6-
5 -DID-
4 -013
3 -013
2 -011
1 -008
1 -020
0 -010-
0 -024
6 -011
5 -018
4 -021
3 -021
2 -017
1 -011
0 -006-
0 -013
	
6 -024
4 -012
3 -012
2 -010+
1 -006
1 -017
0 -007
0 -017
	
5 -017
4 -023
3 -022
2 -017
1 -010 +
1 -02S-
0 -010 +
0 -023
	
4 -012
3 -014
2 -011
1 -007
1 -017
0 -006
0 -P1B-
' 	
3 .-007
2 -007
0-01
0 -003
0 -007
	
—
6 -005 -f
5 -010-
3 -004
2 -004
1 -003
1 -008
0 -004
0 -010-
—
5 -003
4 -008
3 -007
2 -008
1 -004
0 -002
0 -DOS-
	
—
5 -007
3 -003
2 -003
1 -002
1 -006
0 -003
0 -007
	
	
4 -008"
3 -007
2 -006
1 -004
0 -001
0 -004
—
— •
—
3 -003
2 -003
1 -002
1 -007
0 -002
0 -006
' • 	
	 .
3 -007
2 -007
0-005
0 -003
—
—
—
5 -002
4 -003
3 -004
2 -004
1 -003
0 -001
0 -004
	
	
5 -003
3 -002
2 -002
1 -001
1 -004
0 -002
0 -006-
	
	
4 -002
3 -003
2 -003
1 -002
0 -001
0 -003
	
	
	
4 -OOB-
2 -001
1 -001
1 -004
0 -001
0 -004
—
—
—
3 -003
2 -003
1 -002
0 -001
0 -002
—
— '
—
2 -001
1 -001

A=13 B=7





6






5





4




3



2



A=14 B=14










13






a
11
10
9
8
7
6
13
12
11
10
9
8
7
13
12
11
10
9
8
13
12
11
10
9
13
12
11
10
13
12


14
13
12
11
10
9
8
7
6
5
4
14
13
12
11
10
9
8
; Probability
0-05
2 -022
1 -012
1 -029
0 -010 +
0 -022
0 -044
3 -021
2 -017
2 -046
1 -024
1 -060-
0 -017
0 -034
2 -012
2 -044
1 -022
1 -047
0 -01B-
0 -029
2 -044
1 -022
0 -006
0 -01 S-
0 -029
1 -02B
0 -007
0 -018
0 -036
0 -010-
0 -029


10 -049
8 -038
6 -023
5 -027
4 -028
3 -027
2 -023
1 -018
1 -038
0 -020
0 -049
9 -041
7 -029
6 -037
5 -041
4 -041
3 -038
2 -031
0-025
2 -022
1 -012
0 -004
O1 -010 +
0 -022

3 -021
2 -017
1 -010-
1 -024
0 -008
0 -017

2 -012
1 -008
1 -022
0 -007
0 -01 6-

1 -006
1 -022
0 -006
0 -01 B-

1 -02B
0 -007
0 -018

0 -010-



9 -020
7 -016
6 -023
4 -011
3 -011
2 -009
2 -023
1 -018
0 -008
0 -020

8 -018
6 -011
5 -015 +
4 -017
3 -016
2 -013
1 -009
0-01
1 -004
0 -002
0 -004

__
	
2 -004
1 -003
1 -010-
0 -003
0 -008

	
1 -002
1 -008
0 -002
0 -007

__
1 -006
0 -002
0 -006

__
0 -002
0 -007

__
0 -010-



8 -008
6 -006
5 -009
3 -004
2 -003
2 -009
1 -006
0 -003
0 -008

__
7 -006
5 -004
4 -005 +
3 -006
2 -005-
1 -003
1 -009
0-005
1 -004
0 -002
0 -004

	
___
2 -004
1 -003
0 -001
0 -003

	
	
1 -002
0 -001
0 -002

_
__
0 -000
0 -002

	
___
0 -002

	
, 	
	
, 	 ,_


7 -003
5 -002
4 -003
3 -004
2 -003
1 -002
0 -001
0 -003


	
6 -002
5 -004
3 -002
2 -001
2 -005-
1 -003

                                   305

-------
TABLE G.5.  SIGNIFICANT LEVELS OF B: VALUES OF B (LARGE TYPE)
            AND CORRESPONDING PROBABILITIES (SMALL TYPE)1
            (CONTINUED)


A*14 8=13

12










11









10









9








8










7
6
5
14
13
12
11
10
9
8
7
6


14
13
12
11
10
9
8
7
6
5
14
13
12
11
10
9
8
7
6
5
14
13
12
11
m
1U
9
8
7
6
14
13
12
11
10
9
8
7
A
Probability
0-05
1 -021
1 -048
0 426-
8 -033
6 -021
5 -025+
4 -028
3 424
2 -oia
2 -042
1 -028
0. 413
0__n
430
7 -028
6 -039
5 -043
4 -042
3 -038
2 -027
1 -O17
1 -038
0 -017
0 -038
6 -020
5 428
4 -O28
3 -O24
2 -018
2 -O40
1 -024
0 -010-
0 -022
0 -047
6 .047
4 418
3 417
3 442
2___
429
1 417
1 438
0 414
0 430
5 438
4 -039
3 432
2 422
2 448
1 428
0 409
0 420

0-025
1 421
0 410 +
0 425-
7 412
6 421
4 409
3 409
3 424
2 410
1 412
0 405+
0 413


6 409
5 414
4 418
3 418-
2 411
1 407
1 417
0 407
0 417
	
6 420
4 409
3 409
3 424
2 418
1 411
1 424
0 410-
0 422
„
5 414
4 418
3 417
2 412
1 -nn-7
1 407
1 417
0 -000
0 414

4 410-
, 3 411
2 -ooo
2 422
1 412
0 404
0 409
0 420
	
0-01
0 404
	
6 404
5 -007
4 -000
3 -009
H -007
1 -006-
0 -002
0 405 +
	
__

6 409
4 404
3 405-
2 404
1 -003
1 407
0 403
0 407
	
	
5 -008
4 -009
3 409
2 -O07
1 -004
0 -002
0 -004
0 410-
	
_
4 404
3 -005-
2 -004
1 402
Innl
•OO7
0 -002
0 -008
	 •
	
4 -010-
2 402
2 -009
1 405-
0 -002
0 404
0 409
	
	
0-005
0 -004
	
6 -004
4 -002
3 -003
2 -002
1 -002
1 -005-
0 -002
	
	
___

5 -003
4 -004
3 405-
2 -004
1 -003
0 -001
0 403
	
	
	
4 -002
3 -002
2 402
1 -001
1 -004
0 -002
0 -004
—
—
—
4 -004
3 -005-
2 -004
1 -002
fi nni
0 -002
	
	
	
3 -002
2 -002
1 -001
1 405-
0 -002
0 -004
	
	
	


A=14 B=7






6








5






4





3



2





A=15 8=15













14
13
12
11
10
9
8
7
14
13
12
11
10

9
8
7
14
13
12
11
10
9
8
14
13
12
11
10
9
14
13
12
11
14
13
12



15
14
13
12
11
•10
9
8
7
6
5
4
Probability
0-05
4 -026
3 -025
2 417
2 441
1 -021
1 -043
0 -015-
0 -030
3 -018
2 414
2 -037
1 -018
1 438

0 412
0 -024
0 444
2 410 +
2 -037
1 -017
1 -038
0 -011
0 -022
0 -040
2 -039
1 419
1 -044
0 -011
0 -023
0 441
1 -022
0 408
0 -015-
0 429
0 408
0 425
0 450



11 450-
9 -040
7 425+
6 -030
5 -033
4 -033
3 430
2 425+
1 -018
1 440
0 421
0 -050-
0-025
3 -006
2 -006
2 417
1 -009
1 -021
0 -007
0 415-
	
3 -018
2 414
1 -007
1 418
0 405 +

0 -012
0 -024
*^~
2 410 +
1 406
1 -017
0 406-
0 -011
0 -022
	
1 -005-
1 419
0 -005-
0 -011
0 423
	
1 -022
0 -008
0 -015-
	
0 408
0 -025
	



10 421
8 -018
6 410+
5 -013
4 -013
3 -013
2 410+
1 -007
1 418
0 -008
0 -012
— ~
0-01
3 -006
2 -006
1 -003
1 -009
0 -003
0 -007
~~
: —
2 -003
1 -002
1 -007
0 -002
0 -005 +

~~
~~-
— ~~
1 -001
1 -008
0 -002
0 -005-
— •"
—
—
1 -005-
0 -002
0 -005- '
—
_—
^~
0 -001
0 -006
"•""•
1 —
0 -008
—
_



9 -008
7 -007
5 -004
4 -005-
3 -005"
2 404
1 -003
1 -007
0 -003
0 -008
	
^^
0-005
2 -001
1 -001
1 -003
0 -001
0 -003
"•
1 "
*— " '
2 -003
1 -002
0 -001
0 -002
—

-
~~

1 -001
0 -001
0 -002
0 -005-
—
"~~
	
1 -005-
0 -002
0 -005-
—
	
-
0 -001
-—
•
—
_—
„
"~~



8 -003
6 -003
5 -004
4 -006-
3 -005-
2 -004
1 -003
0 -001
0 -003
—
•~—
— — 	
                              306

-------
TABLE G.5.
SIGNIFICANT LEVELS OF B: VALUES OF B (LARGE TYPE)
AND CORRESPONDING PROBABILITIES (SMALL TYPE)1
(CONTINUED)

A=15 B=14



13




12






11





10




9

0!
15
14
13
12
11
10

8
7
5
15
14
13
12
11
10
9
8
7
6
15
14
13
12
11
10
9
8
7

15
14
13
12
11
10
9
8
7
6
•
15
14
13
12
11
10 .
9
8
7
6
15
14
Probability
0-05
10 -042
8 -031
7 -041
6 -O4S
5 -048
4 -048
3 -041
2 -033
1 -022
-049
0 -025+
9 -03B-
7 -023
6 -029
5 -031
4 -030
3 -026
2 -020
2 -043
1 -029
0 -013
0 -031
8 -028
7-043
6 -049
5 -049
4 -045+
3 -038
2 -028
1 -018
1 -038
•017
-037
7 -022
6 -032
5 -034
4 -032
3 -028
2 -019
2 -040
1 -024
1 -049
0 -022
0 -046
6 -017
.5-023
4 ' -022
3 -018
3 -042
2 -029
1 -016
1 -034
0 -013
0 -028
6 -042
5 -047
0-025
9 -017
7 -013
6 -O17
5 -020
4 -020
3 -018
2 -014
1 -009
1 -022
0 -011
8 -013
7 ;023
5 -011
4 -012
3 -011
2 -008
2 -020
1 -013
0 -005+
0 -O13
7 -010-
6 -016
5 -019
4 -019
3' -017
2 -012
1 -007
1 -018
0 -007
0 -017
7 -022
5 -011
4 -012
3 -010+
2 -008
2 -O19
1 -011
1 -024
0 -010-
0 -022
6 -O1 7
5 -023
4 -022
3 -018
2 -013
1 -007
0 -016
0 -006
1 -013
5 -012
4 -015-
0-01
8 -O08
6 -oos"
5 -007
4 -007
3 -007
2 -006
1 -004
,1 -oos
0 -004
7 -005"
6 -009
4 -004
3 -004
2 -003
2 -008
1 -005+
0 -002
0 -006+
7 -010 -
5 -006
4 -007
3 -006
2 -006"
1 -003
1 -007
0 -003
0 -007
"^
6 -007
4 -003
3 -003
2 -003
2 -008
1 -004
0 -002
0 -004
0 -010-
—
5 -oos-
4 -007
3 -007
2 -oos-
1 ' -003
1 -007
0 -002
0 -006
—
4 -003
3 -004
0-005
7 -002
6 -005-
4 -002
3 -002
2 -002
1 -001
1 -004
0 -001
0 -004
7 -005-
5 -003
4 -004
3 -004
2 -003
1 -002
0 -001
0 -002
6 -003
4 -002
3 -002
2 -002
2 -006-
1 -003
0 -001
0 -O03
	
	
5 -002
4 -003
3 -003
2 -003
1 002
1 -004
0 -002
0 -004
	
	
5 -005-
3 -002
2 -001
2 -oos-
1 -003
0 .001
0 -002
—
•—
4 -003


A=15 B=9


8




7



6





5


4


3


2

a
13
12
11
10
9
8
7
6
15
14
13
12
11
10
9
8
7
6
15
14
13
12
11
10
9
8
7
15
14
13
12
11
10
9
8
15
14
13
12
11
10
9
15
14
13
12
11
10
15
14
13
12
11
15
14
13
Probability
0-05
4 -042
3 -032
2 -021
2 -045-
1 -024
1 -048
0 -019
0 -037
5 -032
4 -033
3 -026
2 -017
2 -037
1 -019
1 -038
0 -013
0 .026
0 -oso-
4 -023
3 -021
2 -O14
2 -032
1 -015+
1 -032
0 -010+
0 -020
0 -038
3 -01 6+
2 -011
2 -031
1 -014
1 -029
0 -009
0 -017
0 -032
2 -coo
2 -032
1 -014
1 -031
0 -008
0 -016
0 -030
2 -035+
1 -016
1 -03'7
0 -009
0 -018
0 -033
1 -020
0 -005-
0 -012
0 -025-
0 -043
0 -007
0 -022
0 -044
0-025
3 -013
2 -009
2 -021
, 1 -01 1
1 -O24
0 -009
0 -019

4 -008
3 -009
2 -006
2 -017
1 -008
1 -019
0 -006
0 -013

__
4 -023
3 -021
2 -014
1 -007
1 -015+
0 -006-
0 -010+
0 -O20

3 -016+
2 -Oil
1 -008
1 -014
0 -004
0 -009
0 -017

; 2 -009
1 -005 "
1 -014
0 -004
0 -008
0 -cue

1 -004
1 -018
0 -004
0 -009
0 -018

1 -020
0 -oos-
0 -012
0 -025-

0 -007
0 -022

0-01
2 -003
2 -009
1 -oos-
0 -002
0 -004
0 -009

	
4 -008
3 -009
2 -006
1 -003
1 -008
0 -003
0 -006

^_
	
3 -006-
2 -004
1 -002
1 -007
0 -002
0 -005-


2 -003
1 -002
1 -006
0 -002
0 -004
0 -009
	
2 -009
1 -oos-
0 -001
0 -004
0 -008

	
1 -004
0 -001
0 -004
0 -009
	
0 -001
0 -oos-
__
,^_
0 -007

0-005
2 -003
1 -002
1 -005 -
0 -002
0 -004
	

3 -002
2 -002
1 -001
1 -003
0 -001
0 -003
	


3 -005-
2 -004
1 -002
0 -001
0 -002
0 -COS'


2 -003
1 -002
0 -001
0 -002
0 -004


1 -001
1 -005-
0 -001
0 -004
	

1 -004
0 -001
0 -004

0 -001
0 -005"
	
	
—

                            307

-------
                                  APPENDIX H

          SINGLE CONCENTRATION TOXICITY TEST - COMPARISON OF CONTROL
                    WITH  100%  EFFLUENT OR  RECEIVING WATER


1.  To statistically compare a control  with one concentration,  such as 100%
effluent or the instream waste concentration, a t-test is the recommended
analysis.  The t-test is based on the assumptions that the observations are
independent and normally distributed and that the variances of the
observations are equal between the two groups.

2   Shapiro Wilk's test may be used to test the normality assumption (see
Appendix B for details).  If the data do not meet the normality assumption,
the nonparametric test, Wilcoxon's Rank Sum Test, may be used to analyze the
data   An example of this test is given in Appendix F.  Since a control and
one concentration are being compared, the K = 1 section of Table F.5 contains
the needed critical values.

3   The F test for equality of variances is used to test the homogeneity of
variance assumption.  When conducting the  F test, the alternative hypothesis
of interest is that the variances are not  equal.

4.  To make the two-tailed F test at the 0.01 level of significance, put the
larger of the two variances in the numerator of F.
F =
                                     where Sฑ2 > S22
 5.  Compare  F with the  0.005 level of  a tabled F value with n, - 1 and n2 -  1
 degrees  of freedom, where n., and n2 are the number of replicates for each of
 the two  groups.

 6   A set of Cen'odaphm'a dubia  reproduction data  from an effluent  screening
 test  will be used to  illustrate  the  F  test.  The raw data, mean  and  variance
 for the  control  and 100% effluent  are  given in Table H.I.

 7.  Since the variability of the 100%  effluent is  greater than  the  variability
 of the control,  S  for  the 100%  effluent concentration is placed in the
 numerator of the F  statistic and S  for the control  is placed in the
 denominator.
                                  F =
            36.61
            14.55
 8.  There are 9 replicates for the effluent concentration and 8 replicates  for
 the control.  Thus,  the numerator degrees of freedom is 8 and the denominator
 degrees of freedom is 7.  For a two-tailed test at the 0.01  level  of
 significance, the critical F value is obtained from a table  of the F

                                      308

-------
                TABLE H.I.  CERIODAPHNIA DUBIA REPRODUCTION DATA
                            FROM AN EFFLUENT SCREENING TEST
                                       Replicate
                                                      8
                                               10
   Control        36   38   35    35   28   41   37   33
   100% Effluent  23   14   21     7   12   17   23    8   18
                                                   35.4   14.5
                                                   15.9   36.6
 distribution (Snedecor and Cochran,  1980).   The critical  F value for this test
 is 8.68.   Since 2.52 is not greater  than 8.68,  the conclusion is that the
 variances of the control  and 100% effluent  are  homogeneous.

 9.  EQUAL VARIANCE T-TEST

 9.1   To perform the t-test,  calculate  the following test  statistic:

                                t = 	?1~I*—
   Where:
                                     s I  A+_L
                                      "N  *i  *2
Y,   =  Mean for the control

Y2   =  Mean for the effluent concentration
                                ni+n2-2
                S,   =  Estimate of the variance for the control
                 2                  '                      '
                S2   =  Estimate of the Variance for the effluent
                        concentration

                n,   =  Number of replicates  for the control

                n2   =  Number of replicates  for the effluent
                        concentration

9.2  Since we are usually concerned with a decreased response  from the
                                     309

-------
control, such as a decrease in survival  or a decrease in reproduction,  a
one-tailed test is appropriate.  Thus, compare the calculated t with a
critical t, where the critical t is at the 5% level of significancewithn, +
n  - 2 degrees of freedom.   If the calculated t exceeds  the  critical  t,  the
mean responses are declared different.
9.3  Using the data from Table H.I to illustrate the t-test, the calculation
of t is as follows:
                                35.4-15.9  = 7 go
                           Cป ~~             *"*    "
Where:
                 (8-1)14.5+(9-l)36.6  = 5 13
                        (8+9-2)
 9  4   For  an  0.05  level  of  significance  test with  15 degrees of freedom the
 critical  t  is  1.754  (Note:   Table  D.5 for  K =  1 includes the critical t values
 for  comparing  two groups).   Since  7.82  is  greater than  1.754, the conclusion
 is that the  reproduction in  the  100% effluent  concentration is significantly
 lower than  the control  reproduction.
 10.   UNEQUAL VARIANCE T-TEST
 10 1  If  the F test  for equality of variance  fails, the t-test is still a
 valid test.   However, the denominator of the  t statistic is adjusted  as
 f ol1ows:
                                 t =
                                             n2
      Where:     Y.,  =  Mean for the control
                        Mean for the effluent concentration
                        Estimate of the variance for the control
                        Estimate of the variance for the effluent
                        concentration
                 n1  =  Number of replicates for the control
                 n2  =  Number of replicates for the effluent
                        concentration
                                       310

-------
10.2  Additionally,  the degrees of freedom for the  test  are  adjusted using the
following formula:                                         i
                       df =
                                     ca+(l-O2(n1-l)
         Where:
                                 C =
10.3  The modified degrees of freedom is  usually  not  an  integer.
practice is to round down to the nearest  integer.
Common
10.4  The t-test is then conducted as  the  equal  variance t-test.  The
calculated t is compared to the critical t at  the 0.05  significance level with
the modified degrees of freedom.   If the calculated t exceeds the critical t,
the mean responses are found to be statistically different.
                                     311

-------
                                  APPENDIX I

                                PROBIT ANALYSIS

1.  This program calculates the EC1 and EC50 (or LCI and LC50), and the
associated 95% confidence intervals.

2.  The program is written in IBM PC Basic for the IBM compatible PC by
Computer Sciences Corporation, 26 W. Martin Luther King Drive, Cincinnati, OH
45268.  A compiled, executable version of the program can be obtained from
EMSL-Cincinnati by sending a written request to EMSL at 3411 Church Street,
Cincinnati, OH  45244.

2.1  Data input is illustrated by a set of total mortality data (Figure I.I)
from a fathead minnow embryo-larval survival and teratogenicity test.  The
program requests the following input:

   1. Desired output of abbreviated (A) or full (F) output?  (Note: only
      abbreviated output is shown below.)
   2. Output designation (P = printer, D = disk file).
   3. Title for the output.
   4*. The number of exposure concentrations.
   5. Toxicant concentration data.

2.2  The program output for the abbreviated output includes the following:

   1. A table of the observed proportion responding and the proportion
      responding adjusted for the controls (see Figure 1.2).
   2. The calculated chi-square statistic for heterogeneity and the tabular
      value.  This test is one indicator of how well the data fit the model.
      The program will issue a warning when the test indicates that the data
      do not fit the model.
   3. Estimated LCI and LC50 values and associated 95% confidence intervals
      (see Figure 1.2).
                                      312

-------

                                    USEPA PROSIT ANALYSIS PROGRAM
                                  USED FOR CALCULATING  LC/EC VALUES
                                             Version 1.5
 Do you wish abbreviated (A)  or full ? P
 Title ? Example of Probit Analysis

 Number responding in the control group = ? 2
 Number of animals exposed in the concurrent control  group = ? 20
 Number of exposure concentrations, exclusive of controls ? 5

 Input data starting with the lowest exposure concentration

 Concentration = ? 0.5
 Number responding = ? 2
 Number exposed = ? 20

 Concentration = ? 1.0
 Number responding = ? 1
 Number exposed = ? 20

 Concentration = ? 2.0
 Number responding = ? 4
 Number exposed = ? 20

 Concentration = ? 4.0
 Number  responding = ? 16
 Number  exposed = ? 20

 Concentration = ? 8.0
 Number  responding = ? 20
 Number exposed = ? 20

Number
1
2
3
4
5

Cone.
0.5000
1.0000
2.0000
4.0000
8.0000
Number
Resp.
2~
1
4
16
20
Number
Exposed
20 —
20
20
20
20
Do you wish to modify your data ? N

The number of control animals which  responded =  2
The number of control animals exposed  =  20
Do you wish to modify these values ? N
Figure1.I.  Sample Data  Input  for USEPA Probit Analysis Program,  Version
                J. •, 0- •                               •                      i
                                               313

-------
Example of Probit Analysis



Cone.
Control
0.5000
1.0000
2.0000
4.0000
8.0000


Number
Exposed
20
20
20
20
20
20


Number
Resp.
2
2
1
4
16
20

Observed
Proportion
Responding
0.1000
0.1000
0.0500
0.2000
0.8000
1.0000
Proportion
Responding
Adjusted for
Controls
0.0000
0.0174
-.0372
0.1265
0.7816
1.0000
Chi - Square for Heterogeneity  (calculated)
Chi - Square for Heterogeneity
         (tabular value at 0.05  level)
                                    0.441

                                    7.815
Example of  Probit Analysis

       Estimated  LC/EC Values and Confidence Limits
 Point

 LC/EC   1.00
 LC/EC  50.00
  Exposure
    Cone.

     1.346
     3.018
  Lower         Upper
95% Confidence Limits
 0.453
 2.268
1.922
3.672
 Figure  1.2.
USEPA  Probit  Analysis Program  Used for  Calculating  LC/EC
Values, Version 1.5.
                                     314

-------
                                  APPENDIX J

                            SPEARMAN-KARBER METHOD
                                                          i

1.  The Spearman-Karber Method is a nonparametric statist!ca'l procedure for
estimating the LC50 and the associated 95% confidence interval (Finney, 1978).
The Spearman-Karber Method estimates the mean of the distribution of the Iog10
of the tolerance.  If the log tolerance distribution is symmetric, this
estimate of the mean is equivalent to an estimate of the median of the log
tolerance distribution.

2.  If the response proportions are not monotonically non-decreasing with
increasing concentration (constant or steadily increasing with concentration),
the data must be smoothed.  Abbott's procedure is used to "adjust" the
concentration response proportions for mortality occurring in the control
replicates.                                               i

3.  Use of the Spearman-Karber Method is recommended when partial mortalities
occur in the test solutions, but the data do not fit the Probit model.

4.  To calculate the LC50 using the Spearman-Karber Method, the following must
be true:  1) the smoothed adjusted proportion mortality for the lowest
effluent concentration (not including the control) must be zero, and 2) the
smoothed adjusted proportion mortality for the highest effluent concentration
must be one.

5.  To calculate the 95% confidence interval for the LC50 estimate, one or
more of the smoothed adjusted proportion mortalities must be between zero and
one.                                                      !
                                                          i

6.  The Spearman-Karber Method is illustrated below using a set of mortality
data from a Fathead Minnow Larval Survival and Growth test.  These data are
listed in Table J.I.

         TABLE J.I. EXAMPLE OF SPEARMAN-KARBER METHOD:  MORTALITY DATA FROM
                    A FATHEAD MINNOW LARVAL SURVIVAL AND GROWTH TEST
                    (40 ORGANISMS PER CONCENTRATION)
Eff 1 uent
Concentration
Control
6.25%
12.5%
25.0%
50.0%
100.0%
Number of
Mortalities
2
2
0
0
26
40
Mortality
Proportion
0.05
0.05
0.00
0.00
0.65
1.00
                                      315

-------
7.  Let p0,  pr  ..., pk denote the observed response proportion mortalities
for the control and k effluent concentrations.   The first step is to smooth
the p,. if they do not  satisfy p0 < p1 < ... < pk.   The smoothing process
replaces any adjacent p('s that  do not  conform  to  p0 < p., < ... < pk with  their
average.  For example, if p,- is  less than  p,-.., then:


                           P/-I  -Pi = (Pi+Pi-i)/2


Where:  p?   -= the smoothed observed proportion mortality for  effluent
               concentration i.

7.1  For the data in this example,  because the  observed mortality proportions
for the control and the 6.25% effluent concentration are greater than the
observed response proportions for the 12.5% and 25.0% effluent concentrations,
the responses for these four groups  must be averaged:

        Pcf  = P/ = P/ = P/ = 0-05+0-05*ฐ-00+0-00 -  -^ =0.025
7.2  Since p4 - 0.65 is larger than pj,  set p^ = 0.65.  Similarly, p5 =  1.00  is
larger than p^, so set p^  =  1.00.   Additional smoothing is not necessary.  The
smoothed observed proportion mortalities are shown in Table J.2.

8.  Adjust the smoothed observed proportion mortality in each effluent
concentration for mortality in the control group using Abbott's formula
(Finney, 1971).  The adjustment takes the form:

Where:  p? = (p* -  p*} / (1  - PQ)

        Po = the smoothed  observed proportion mortality for  the control

        p? = the smoothed  observed proportion mortality for  effluent
             concentration  i .

8.1  For the data in this example, the data for each effluent concentration
must be adjusted for control mortality using Abbott's formula, as follows:

         ซa     a _  _a _ „& _  Pi~PoS _  0.025-0.025 _   0.0   _
        Po  =Pi  -P2  -Pa  -
                                           i-o.025         975
                    a_P4-Po_ 0.650-0.025, 0 . 0625 ,
                  P4     -       1-0.025   "  0.975
                                      316

-------
                         ~~ t j*   n  nr\n_n  nr>c   n  a *~7 si
                                                        1.000
a_^s-Po    1.000-0.025  _  0.975
                       l-p<
                                  1-0.025     0.975
The smoothed, adjusted response proportions for the effluent concentrations
are shown in Table J.2.  A plot of the smoothed, adjusted data is shown in
Figure J.I.


      TABLE J.2.  EXAMPLE OF SPEARMAN-KARBER METHOD:  SMOOTHED,  ADJUSTED
                  MORTALITY DATA FROM A FATHEAD MINNOW LARVAL SURVIVAL
                  AND GROWTH TEST
Effluent
Concentration
Control
6.25%
12.5%
25.0%
50.0%
100.0%
Mortality
Proportion
0.05
0.05
0.00
0.00
0.65
1.00
Smoothed
Mortality
Proportion
0.025
0.025
0.025
0.025
0.650
1.000
Smoothed ,
| Adjusted
Mortal i ty
Proportion
0.000
; 0.000
0.000
0.000
0.641
1.000
9. Calculate the Iog10  of  the  estimated LC50, m, as follows:
Where:  p? = the smoothed adjusted proportion mortality at  concentration  i

        X,- = the Iog10 of concentration i

        k  =  the number of effluent concentrations tested,  not  including the
             control.

9.1  For this example, the Iog10  of the estimated  LC50, m,  is calculated  as
follows:

      m = [(0.000 - 0.000) (0.7959 + 1.0969)]/2 +
          [(0.000 - 0.000) (1.0969 + 1.3979)]/2 +
          [(0.641 - 0.000) (1.3979 + 1.6990)]/2 +
          [(1.000 - 0.641) (1.6990 + 2.0000)3/2

        = 1.656527


                                      317

-------
                                                                     CO
                                                                     O>
*
                                                     *
                                                           ง

                                                           8
ง
                                                            .
                                                           to
                                                           CM
         111
         o
         o
         o
      T-  111
                                                              ti-
                                                              ll.
                                                              Ill
 in
UCM
 
                                                                     re
                                                                     C4-

                                                                     cu
                                                                     s-
                                                                     o
                                                                     ns
                                                                     +->
                                                                     nj
                                                                     O)
                                                                     CO
                                                                     =3
                                                                     TJ
                                                                     OJ
                                                                     O
                                                                     o
                                                                     oo

                                                                     cu
                                                                     O


                                                                     51
                s-
                3
                en

-------
10.    Calculate the estimated variance  of m  as  follows:
                              i=2
                                        4
Where:  X,-  = the  Iog10 of concentration i
        n,-  = the  number  of  organisms  tested at effluent concentration i
        p*  = the  smoothed adjusted  observed proportion mortality at effluent
             concentration  i
        k  = the number of  effluent concentrations tested,  not  including  the
             control.                                      !
10.1  For this example,  the estimated variance of m,  V(m),  is calculated  as
fol1ows:
      V(m)  = (0.000)(1.000)(1.3979 - 0.7959)?/4(39) +
             (0.000)(1.000)(1.6990 - 1.0969)V4(39) +
             (0.641)(0.359)(2.0000 - 1.3979)74(39)
           = 0.00053477
                                                          QjV(m)
                                                          is calculated as
11.    Calculate the 95% confidence  interval  for m:    mฑ2.
11.1  For this example, the 95% confidence  interval  for m
follows:
             1.656527ฑ2^0-. 00053477 = (1.610277, 1.702777)
12.    The estimated LC50 and a 95% confidence  interval for  the  estimated  LC50
can be found by taking base10 antilogs of the above values,,
12.1  For this example, the estimated LC50  is  calculated as follows:
             LC50 = antilog(m) = antilog(1.656527)  = 45.3%.
12.2  The limits of the 95% confidence interval  for the estimated  LC50  are
calculated by taking the antilogs of the upper and  lower limits of the  95%
confidence interval for m as follows:
            lower limit:   antilog(l.610277) = 40.8%
            upper limit:   antilog(l.702777) = 50.4%
                                      319

-------
                                  APPENDIX K

                        TRIMMED SPEARMAN-KARBER METHOD
1.  The Trimmed  Spearman-Karber Method  is  a modification  of the Spearman-
Karber Method, a nonparametric statistical procedure for  estimating the  LC50
and the associated 95% confidence  interval (Hamilton et al ; 1977).  Appendix
The Trimmed Spearman-Karber Method estimates the trimmed  mean of the
distribution of  the Iog10 of the  tolerance.  If the log tolerance  distribution
is symmetric, this estimate of the trimmed mean is equivalent to an estimate
of the median of the log tolerance distribution.

2.  If the response proportions are not monotonically  non-decreasing with
increasing concentration (constant or steadily increasing with concentration),
the data must be smoothed.  Abbott's procedure is used to "adjust" the
concentration response proportions for mortality occurring in the control
replicates.

3.  Use of the Trimmed Spearman-Karber Analysis is recommended only when the
requirements for the Probit Method and the Spearman-Karber Method are not met.

4.  To calculate the LC50 using the Trimmed Spearman-Karber Method, the
smoothed, adjusted, observed proportion mortalities must  bracket 0.5.

5.  To calculate the 95% confidence interval for the LC50 estimate, one or
more of the smoothed, adjusted, observed proportion mortalities must be
between zero and one.

6.  Let p0,  pซ,  ...,  p. denote the  observed proportion  mortalities for the
control and the  k effluent concentrations.  The first  step is to smooth the p,-
if they do not satisfy p0 < p, <  ... < pk.  The smoothing  process replaces any'
adjacent p,-'s  that do not conform to p0  <  p1 <  . . . < pk, with their average.
For example, if  p,. is less than p,..,, then:

Where:  p'.,  ซ   p? = (p,  + p^J/Z

        p*   -   the smoothed observed proportion mortality for effluent
                 concentration i .

7.  Adjust the smoothed observed proportion mortality  in each effluent
concentration for mortality in the control group using Abbott's formula
(Finney,  1971).  The adjustment takes the form:

Where:  p*   =   (p? -  p*) /  (1 - p*)

        Po   =  the smoothed observed proportion mortality for the control

        pฎ   =  the smoothed observed proportion mortality for effluent
                concentration i.
                                      320

-------
8.  Calculate the amount of trim to use in the estimation of the LC50 as
follows:

Where:  Trim =  max(p*,  l-pฃ)

        p^   =  the smoothed, adjusted proportion mortality for the lowest
                effluent concentration, exclusive of the control

        pฃ   =  the smoothed, adjusted proportion mortality for the highest
                effluent concentration

        k    =  the number of effluent concentrations, exclusive of the
                control.

The minimum trim should be calculated for each data set rather than using a
fixed amount of trim for each data set.

9.  Due to the intensive nature of the calculation for the estimated LC50 and
the calculation of the associated 95% confidence interval using the Trimmed
Spearman-Karber Method, it is recommended that the data be analyzed by
computer.

10.  A computer program which estimates the LC50 and associated 95% confidence
interval using the Trimmed Spearman-Karber Method, can be obtained from EMSL-
Cincinnati by sending a written request to EMSL, 3411 Church Street,
Cincinnati, OH  45244.

11.  The Trimmed Spearman-Karber program automatically performs the following
functions:

             a. Smoothing.
             b. Adjustment for mortality in the control.
             c. Calculation of the necessary trim.
             d. Calculation of the LC50.
             e. Calculation of the associated 95% confidence interval.

12.  To illustrate the Trimmed Spearman-Karber method using the Trimmed
Spearman-Karber computer program, a set of data from a Fathead Minnow Larval
Survival and Growth test will be used.  The data are listed in Table K.I.
                                                          i

12.1  The  program requests the following input  (Figure K.I):
      au   Output destination  (D = disk file, P = printer).
      b.   Control data.
      c.   Data for each toxicant concentration.
       •'         •                                         i

12.2  The  program output includes the following  (Figure K.2):
      a.   A table of the concentrations tested, number of organisms
           exposed, and mortalities.
      b.   The amount of trim  used in the calculation.
      c.   The estimated LC50  and the associated 95% confidence  interval.
                                      321

-------
TABLE K.I.  EXAMPLE OF TRIMMED SPEARMAN-KARBER METHOD:  MORTALITY
            DATA FROM A FATHEAD MINNOW LARVAL SURVIVAL AND GROWTH
            TEST (40 ORGANISMS PER CONCENTRATION)

      Effluent               Number of            Mortality
    Concentration           Mortalities           Proportion
Control
6.25
12.5
25.0
50.0
100.0
2
0
2
0
0
32
0.05
0.00
0.05
0.00
0.00
0.80
                               322

-------
A:>spearman

 TRIMMED SPEARMAN-KARBER METHOD.  VERSION 1.5

ENTER DATE OF TEST:
1
ENTER TEST NUMBER:
2
WHAT IS TO BE ESTIMATED?
(ENTER "L" FOR LC50 AND "E" FOR EC50)
L '
ENTER TEST SPECIES NAME:
Fathead minnow
ENTER TOXICANT  NAME:
Eff1uent
ENTER UNITS FOR EXPOSURE CONCENTRATION OF TOXICANT:
f
ENTER THE NUMBER OF INDIVIDUALS IN THE CONTROL:
40
ENTER THE NUMBER OF MORTALITIES IN THE CONTROL:
2
ENTER THE NUMBER OF CONCENTRATIONS
(NOT INCLUDING THE CONTROL;  MAX = 10):
5
ENTER THE  5 EXPOSURE CONCENTRATIONS  (IN INCREASING ORDER):
6.25  12.5  25  50  100
ARE THE NUMBER OF  INDIVIDUALS AT EACH EXPOSURE CONCENTRATION EQUAL(Y/N)?

ENTER THE NUMBER OF INDIVIDUALS AT EACH EXPOSURE CONCENTRATION:
40
ENTER UNITS FOR DURATION OF EXPERIMENT
 (ENTER  "H" FOR HOURS, "D"  FOR DAYS,  ETC.):
Days
ENTER DURATION OF  TEST:
7
ENTER THE NUMBER OF MORTALITIES AT EACH EXPOSURE CONCENTRATION:
0  2 0 0 32
WOULD YOU LIKE THE AUTOMATIC TRIM CALCULATION(Y/N)?
y                                                         !
         Figure K.I.   Example input for Trimmed Spearman-Karber Method.


                                      323

-------
TRIMMED SPEARMAN-KARBER METHOD.  VERSION 1.5
 DATE:   1
 TOXICANT:  effluent
 SPECIES:   fathead minnow

 RAW DATA:  Concentration
TEST NUMBER: 2
DURATION:    7 Days
                .00
               6.25
              12.50
              25.00
              50.00
             100.00

  SPEARMAN-KARBER TRIM:

  SPEARMAN-KARBER ESTIMATES:
Number
Exposed
40
40
40
40
40
40
20.41%
LC50:
Mortal i

2
0
2
0
0
32

77.;
                                         95% CONFIDENCE LIMITS
                                         ARE NOT RELIABLE.
 NOTE:  MORTALITY PROPORTIONS WERE NOT MONOTONICALLY INCREASING.
        ADJUSTMENTS WERE MADE PRIOR TO SPEARMAN-KARBER ESTIMATION.
        Figure K.2.   Example output for Trimmed Spearman-Karber Method.
                                     324

-------
                                  APPENDIX L

                               GRAPHICAL METHOD

1.  The Graphical Method is used to calculate the LC50.  It is a mathematical
procedure which estimates the LC50 by linearly interpolating between points of
a plot of observed percent mortality versus the base 10 logarithm (Iog10)  of
percent effluent concentration.  This method does not provide a confidence
interval for the LC50 estimate and its use is only recommended when there are
no partial mortalities.  The only requirement for the Graphical Method is that
the observed percent mortalities bracket 50%.

2.  For an analysis using the Graphical Method the data must first be smoothed
and adjusted for mortality in the control replicates.  The procedure for
smoothing and adjusting the data is detailed in the following steps.
                                                         I
3.  The Graphical Method is illustrated below using a set of mortality data
from an Fathead Minnow Larval Survival and Growth test.  These data are listed
in Table L.I.

         TABLE L.I.  EXAMPLE OF GRAPHICAL METHOD:  MORTALITY DATA FROM A
                     FATHEAD MINNOW LARVAL SURVIVAL AND GROWTH TEST (40
                     ORGANISMS PER CONCENTRATION)
Effluent
Concentration
%
Control
6.25
12.5
25.0
50.0
100.0
Number of
Mortalities
2
0
0
0
40
40
Mortality
Proportion
0.05
0.00
0.00
0.00
1.00
1.00
 4.   Let p0, p., ..., p,< denote the observed proportion mortalities for the
 control and the k effluent concentrations.  The first step is to smooth the p,
 if they do not satisfy p0 < p, <  ... < pk.  The smoothing process replaces any
 adjacent pf's  that do  not conform to  p0 <  p., < ... < pk with their average.
 For example,  if p,- is  less than  pf., then:                i
 Where:
                           P/-I = Pi
         p?   = the smoothed observed proportion mortality for effluent
                concentration i.

 4.1  For the data in this example, because the observed mortality proportions
 for the 6.25%, 12.5%, and 25.0% effluent concentrations are less than the

                                       325               !

-------
 observed response proportion for the control, the values for these four groups
 must be averaged:
       Po
P!S =
-  0.05+0.00+0.00+0.00    0.05    _ ...-_
-	 -	—  = 0 . 0125
             4                 4
4.2  Since p4 = p5 - 1.00 are larger then  0.0125,  set  p^ = pg = 1.00.
Additional smoothing is  not  necessary.  The smoothed observed proportion
mortalities are shown in Table L.2.

         TABLE L.2.  EXAMPLE OF GRAPHICAL METHOD:  SMOOTHED, ADJUSTED
                     MORTALITY DATA FROM A FATHEAD MINNOW LARVAL
                     SURVIVAL AND GROWTH TEST


Effluent
Concentration
V
/o
Control
6.25
12.5
25.0
50.0
100.0


Mortality
Proportion

0.05
0.00
0.00
0.00
1.00
1.00

Smoothed
Mortality
Proportion

0.0125
0.0125
0.0125
0.0125
1.0000
1.0000
Smoothed,
Adjusted
Mortality
Proportion

0.00
0.00
0.00
0.00
1.00
1.00
5.  Adjust the smoothed observed proportion mortality in each effluent
concentration for mortality in the control group using Abbott's formula
(Finney, 1971).  The adjustment takes the form:
Where:
        Pi
                  Pi =  (P/-P03) /(l-PoS)


=  the smoothed observed proportion mortality for the control

=  the smoothed observed proportion mortality for effluent
   concentration i.
5.1  Because the smoothed observed proportion mortality for the control  group
is greater than zero, the responses must be adjusted using Abbott's  formula,
as follows:
                                     326

-------
       a_   a_   a _   a _  Pi ~Pv  _  0 . 0125 - 0 . 0125  =   0.0
      Po  -Pi  -Pa  -Ps -       s       1 T 0.0125      "0.9875
              a =   a _ P/~Pog  _ 1.00 -0.0125  _  0.9875 _ ฑ  00
            P4     5     i _ns      1-0.0125      0.9875
A table of the smoothed,  adjusted response proportions  for  the  effluent
concentrations are shown  in Table L.2.
                             ^
5.2  Plot the smoothed,  adjusted data on 2-cycle semi-log graph paper with the
logarithmic axis (the y axis) used for percent effluent concentration and the
linear axis (the x axis)  used for observed percent mortality.   A plot of the
smoothed, adjusted data is shown in Figure L.I.

6.  Locate the two points on the graph which bracket 50% mortality  and connect
them with a straight line.

7.  On the scale for percent effluent concentration, read the value for the
point where the plotted line and the 50% mortality line intersect.   This value
is the estimated LC50 expressed as a percent effluent concentration.

7.1  For this example, the two points on the graph which bracket the 50%
mortality line (0% mortality at 25% effluent, and 100% mortality at 50%
effluent) are connected with a straight line.  The point at which the plotted
line intersects the 50% mortality line is the estimated LC50.   The  estimated
LC50 = 35% effluent.
                                      327

-------
   LLJ
   H

   LU
   ID

   LL
   LJL
   LU

   H

   LLI
   O
   DC
   LU
   Q.
      100
       50
10
         1
           0   10 20  30  40  50  60  70  80  90 100

                  PERCENT MORTALITY
Figure L.I   Plot of the smoothed adjusted response proportions  for fathead
          minnow, PimephaTes promeTas, survival data.
                           328

-------
                                  APPENDIX M

                         LINEAR  INTERPOLATION METHOD     j

1.  GENERAL PROCEDURE

1.1  The Linear Interpolation Method is used to calculate a point estimate of
the effluent or other toxicant concentration that causes a given percent
reduction (e.g., 25%, 50%,  etc.)  in the reproduction or growth of the test
organisms (Inhibition Concentration, or 1C).  The procedure was designed for
general applicability in the analysis of data from short-term chronic toxicity
tests, and the generation of an endpoint from a continuous model that allows a
traditional quantitative assessment of the precision of the endpoint, such as
confidence limits for the endpoint of a single test, and a mean and
coefficient of variation for the endpoints of multiple tests.
1.2  The Linear Interpolation Method assumes that the responses (1)
monotonically non-increasing, where the mean response for each higher
concentration is less than or equal to the mean response for the previous
concentration, (2) follow a piecewise linear response function, and  (3) are
from a random, independent, and representative sample of test data.  If the
data are not monotonically nonincreasing, they are adjusted by smoothing
(averaging).  In cases where the responses at the low toxicant concentrations
are much higher than in the controls, the smoothing process may result in a
large upward adjustment in the control mean.  Also, no assumption  is made
about the distribution of the data except that the data within a group being
resampled are independent and identically distributed.  ,
                                                        i
2.  DATA SUMMARY AND PLOTS

2.1  Calculate the mean responses for the control and each toxicant
concentration, construct a summary table, and plot the data.

3.  MONOTONICITY

3.1   If the assumption of monotonicity of test results  is met, the observed
response means  (Y,-) should stay the same or decrease as the toxicant
concentration  increases.  If  the means do not decrease monotonically,  the
responses are  "smoothed" by  averaging  (pooling)  adjacent means.

3.2   Observed means  at each  concentration are considered  in  order  of
 increasing  concentration,  starting with  the  control mean  (Y,).  If the mean
observed  response  at the lowest  toxicant concentration  (Y2) is equal to or
 smaller than the  control mean (Y.,), it is used as the response.  If  it is
larger than the  control mean,  it  is  averaged with  the  control,  and this
 average  is  used  for  both the  control  response  (M.,) and the lowest  toxicant  .
 concentration  response  (M2) .  This mean  is then  compared to the mean observed
 response  for the  next  higher toxicant concentration  (Y3) .  Again,  if the mean
 observed  response for  the  next  higher toxicant  concentration is  smaller  than
 the  mean  of the control  and  the  lowest toxicant  concentration,  it  is used  as
 the  response.   If it is  higher  than  the  mean of the  first  two,  it  is averaged
 with the  first two,  and  the  mean is  used as  the. response for the  control  and

                                      329

-------
two  lowest  concentrations  of toxicant.   This  process  is  continued  for  data
from the  remaining  toxicant  concentrations.   A numerical  example of smoothing
the  data  is provided  below.   (Note:   Unusual  patterns  in  the  deviations  from
monotonicity may  require an  additional  step of smoothing).  Where  Yf decrease
monotonically,  the  Y,- become M,- without smoothing.

4.   LINEAR  INTERPOLATION METHOD

4.1  The  method assumes a  linear  response  from one  concentration to the  next.
Thus, the ICp is  estimated by  linear  interpolation  between  two  concentrations
whose responses bracket the  response  of interest, the  (p) percent  reduction
from the  control.

4.2  To obtain  the  estimate, determine  the concentrations Cj and CJ4.- which
bracket the response  M, (1 - p/100), where M.,  is the smoothed  control mean
response  and p  is the percent  reduction in response relative  to the control
response.   These  calculations  can easily be done  by hand  or with a computer
program as  described  below.    The linear interpolation estimate is calculated
as follows:
              ICp = Cj + [  Mฑ (l - p/100)  -
(g.7 + i  -

UfcH-l  -
Where:  Cd       =   tested concentration whose observed mean response is
                     greater than M.,(l - p/100).

        C, + .    =   tested concentration whose observed mean response is less
                     than M,(l - p/100).

        M.,       =   smoothed mean response for the control.

        Mj       =   smoothed mean response for concentration J.

        Mj+ 1    =   smoothed mean response for concentration J + 1.

        p        =   percent reduction in response relative to the control
                     response.

        ICp      =   estimated concentration at which there is a percent
                     reduction from the smoothed mean control response.
                     The ICp is reported for the test, together with the 95%
                     confidence interval calculated by the  ICPIN.EXE program
                     described below.

4.3  If the Cj is the highest  concentration tested,  the ICp would be specified
as greater than C,.   If the response at the lowest concentration tested is
used to extrapolate the ICp value, the ICp should be expressed as a less than
the lowest test concentration.
                                      330

-------
5.  CONFIDENCE INTERVALS

5.1  Due. to the use of a linear interpolation technique to calculate an
estimate of the ICp, standard statistical methods for calculating confidence
intervals are not applicable for the ICp.  This limitation is avoided by use a
technique known as the bootstrap method as proposed by Efron (1982) for
deriving point estimates and confidence intervals.

5.2  In the Linear Interpolation Method, the smoothed response means are used
to obtain the ICp estimate reported for the test.  The bootstrap method is
used to obtain the 95% confidence interval for the true mean.  In the
bootstrap method, the test data Yj,-  is  randomly resampled with replacement to
produce a new set of data Yj,-*,  that  is  statistically equivalent  to  the
original data, but a new and slightly different estimate of the ICp (ICp*) is
obtained.  This process is repeated at least 80 times (Marcus and Holtzman,
1988) resulting in multiple "data" sets, each with an associate ICp* estimate.
The distribution of the ICp* estimates derived from the sets of resampled data
approximates the sampling distribution of the ICp estimate.  The standard
error of the ICp is estimated by the standard deviation of the individual ICp*
estimates.  Empirical confidence intervals are derived from the quantiles of
the ICp* empirical distribution.  For example, if the test data are resampled
a minimum of 80 time, the empirical 2.5% and the 97.5% confidence limits are
approximately the second smallest and second largest ICp* estimates (Marcus
and Holtzman, 1988).

5.3  The width of the confidence intervals calculated by the bootstrap method
is related to the variability of the data.  When confidence intervals are
wide, the reliability of the 1C estimate is in question. !However, narrow
intervals do not necessarily indicate that the estimate is highly reliable,
because of undetected violations of assumptions and the fact that the
confidence limits based on the empirical quantiles of a bootstrap distribution
of 80 samples may be unstable.

5.4  The bootstrapping method of calculating confidence intervals is
computationally intensive.  For this reason, all  of the calculations
associated with determining the confidence intervals for the ICp estimate have
been incorporated into a computer program.  Computations are most easily done
with a computer program such as the revision of the BOOTSTRP program (USEPA,
1988; USEPA, 1989) which is now called "ICPIN" which is described below in
subsection 7.

6.  MANUAL CALCULATIONS

6.1  DATA SUMMARY AND PLOTS                              ;

6.1.1.  The data used in this example are the Ceriodaphm'a dubia reproduction
data used in the example in Section 13.  Table M.I includes the raw data and
the mean reproduction for each concentration.  Data are included for all
animals tested regardless of death of the organism.  If an animal died during
the test without producing young, a zero is entered.  If death occurred after
producing young, the number of young produced prior to death is entered.  A
plot of the data is provided in Figure M.I.

                                     331

-------
             TABLE M.I.  CERIODAPHNIA DUBIA REPRODUCTION DATA

Replicate
1
2
3
4
5
6
7
8
9
10
Mean (Y,)
1

Control
27
30
29
31
16
15
18
17
14
27
22.4
1

1.56
32
35
32
26
18
29
27
16
35
13
26.3
2 .
Effluent
3.12
39
30
33
33
36
33
33
27
38
44
34.6
3
Concentration (%)
6.25
27
34
36
34
31
27
33
31
33
31
31.7
4
12.5
10
13
7
7
7
10 .
10
16
12
2
9.4
5
25.0
0
0
0
0
0
0
0
0
0
0
0
6
6.2  MONOTONICITY

6.2.1  As can be seen from the plot in Figure M.I, the observed means are not
monotonically non-increasing with respect to concentration.  Therefore, the
means must be smoothed prior to calculating the 1C.
6.2.2  Starting with the control mean T, = 22.4 and T2  =  26.3, we  see that
                         othed means:

                         L = M2 =  ( TI  + T) 2  /2 = 24.35
< T2 .   Calculate the smoothed means:
6.2.3  Since T3 = 34.6 is larger than M2,  average T3 with the previous
concentrations:
Mi = M2 = Af3 = (
                                        M
                                                ) /3  =  27 .7
6.2.4  Additionally, T4 =31.7 is larger than M3,  and is pooled with  the  first
three means.  Thus,

             M1 = M2 = M3 = M4 =• ( ML + M2 + M3 + T^ ) /4  = 28 .7

6.2.5  Since M4 > T5 =  9.4,  set M5 = 9.4.   Likewise,  M= > T6 = 0 and M6 becomes
0.  Table M.2 contains the smoothed means  and Figure M.I gives a plot of the
smoothed response curve.

                                      332

-------
                                                             O)
                                                             S-
                                                             o
                                                             CD
                                                             E

                                                             -o
                                                             O
                                                             O
                                                             to  a)
                                                             c:  >
                                                             O) +->
                                                             E  O
                                                             O)  O
                                                             >  S-
                                                             S-  Q.
                                                             O)  O)
                                                             00  S-
                                                             JO
                                                             o   ••
                                                                 10
                                                                 "3
                                                              0-0
                                                                 o
                                                             •fj •'~
                                                              O  i.
                                                             i—  Ol
                                                             Q- O
                                                              O>
333

-------
                 TABLE M.2.   CERIODAPHNIA  DUBIA  REPRODUCTION MEAN
                             RESPONSE AFTER SMOOTHING
             Effluent
          Concentration
  Response
  Mean (Y,-)
(Young/female)
  Smoothed
  Mean  (Mf)
(Young/female)
Control
1.56
3.12
6.25
12.5
25.0
1
2
3
4
5
6
22.4
26.3
34.6
31.7
9.4
0.0
28.75
.28.75
28.75
28.75
9.40
0.00
6.3  LINEAR INTERPOLATION

6.3.1  Estimates of the IC25 and IC50 are calculated using the Linear
Interpolation Method.  A 25% reduction in reproduction,  compared to the
controls, would result in a mean reproduction of 21.56 young per adult, where
MJl-p/100)  = 28.75(1-25/100).   A 50% reduction  in  reproduction,  compared to
the controls, would result in a mean reproduction of 14.38 young per adult,
where M^l-p/lOO)  = 28.75(1-50/100).   Examining  the smoothed means  and  their
associated concentrations (Table M.2), the two effluent concentrations
bracketing the reproduction of 21.56 young per adult are C4 = 6.25% effluent
and C5 - 12.5% effluent.   The two effluent concentrations  bracketing a
response of 14.38 young per adult are also C4
effluent.
             6.25% effluent  and  C5 = 12.5%
6.3.2  Using Equation 1 from 4.2, the estimate of the IC25 is calculated as
f ol 1 ows :
             IC25 =  6.25 +  [28.75  (1 - 25/100)  - 28.75]

                    = 8.57% effluent
                                      334

-------
 6.3.3   Using  the  equation  from  section  4.2,  the  estimate  of  the  IC50  is
 calculated  as follows:
               ICp = Cj +  [ ML  (1  - p/100)  -
                                                      (Mr
              IC50  = 6.25  + [28.75 (1 - 50/100)  - 28.75]   (12.5  - 6.25
                                                                (9.40 - 28.7f
                     =  10.89% effluent

 6.4  CONFIDENCE  INTERVALS
                                                         i
 6.4.1  Confidence intervals  for the  ICp are derived  using the bootstrap
 method.  As described  above, this method involves randomly resampling the
 individual observations and  recalculating the  ICp at least 80 times, and
 determining the  mean ICp, standard deviation,  and empirical 95% confidence
 intervals.  For  this reason, the confidence intervals are calculated using a
 computer program called ICPIN.  This program is described below and is
 available to carry out all the calculations of both  the  Interpolation estimate
 (ICp) and the confidence intervals.

 7.  COMPUTER CALCULATIONS
                                                         I
 7.1  The computer program, ICPIN, prepared for the Linear Interpolation Method
 was written in TURBO PASCAL  for IBM compatible PCs.  The program (version 2.0)
 has been modified by Computer Science Corporation, Duluth, MN with funding
 provided by the  Environmental Research Laboratory, Duluth, MN (Norberg-King,
 1993).  The program was originally developed by Battelle Laboratories,
 Columbus, OH through a government contract supported by the Environmental
 Research Laboratory, Duluth, MN (USEPA, 1988).   To obtain the program and
 supporting documentation,  send a written request to EMSL-Cincinnati at 3411
 Church Street, Cincinnati, OH  45244.

 7.2  The ICPIN.EXE program performs the following functions::   1) it calculates
 the observed response means  (Y,)  (response  means);   2)  it calculates  the
 standard deviations;  3) checks the responses for monotoriicity;  4)  calculates
 smoothed means (M,)  (pooled  response  means)  if  necessary; 5)  uses the means,
 M,-,  to calculate  the initial  ICp  of choice  by linear interpolation; 6)
 performs a user-specified number of bootstrap resamples between  80  and  1000
 (as multiples of 40); 7) calculates the mean and standard deviation of the
 bootstrapped ICp estimates;  and 8)  provides an  original  95% confidence
 intervals to be used with  the initial ICp when  the number of  replicates per
 concentration is over six and provides both original and expanded  confidence
 intervals when the number  of replicates per concentration are less  than seven
 (Norberg-King, 1993).

 7.3  For the ICp calculation, up  to twelve  treatments can be  used  (which
 includes the control).   There can be  up to  40 replicates per  concentration,
 and the program does not require  an equal  number of replicates per
concentration.  The  value  of p can  range from  1% to 99%.
                                                         i
                                     335

-------
7.4  DATA INPUT.

7.4.1  Data is entered directly into the program onscreen.  A sample data
entry screen in shown in Figure M.2.  The program documentation provides
guidance on the entering and analysis of data for the Linear Interpolation
Method (Norberg-King, 1993).

7.4.2  The user selects the ICp estimate desired (e.g., IC25 or IC50) and the
number of resamples to be taken for the bootstrap method of calculating the
confidence intervals.  The program has the capability of performing any number
of resamples from 80 to 1000 as multiples of 40.  However, Marcus and Holtzman
(1988) recommend a minimum of 80 resamples for the bootstrap method be used
and at least 250 resamples are better (Norberg-King, 1993).

7.5  DATA OUTPUT.

7.5.1  The program output includes the following (Figures M.3 and M.4):

   1. A table of the concentration identification, the concentration tested
      and raw data response for each replicate and concentration.
   2. A table of test concentrations, number of replicates, concentration
      (units),  response means  (?,), standard deviations for each response
      mean, and the pooled response means  (smoothed means; M,-).
   3. The linear interpolation estimate of the ICp using  the means  (M,-).  Use
      this value for the ICp estimate.
   4. The mean  ICp and standard deviation  from the bootstrap resampling.
   5. The confidence intervals calculated  by the bootstrap method for the  ICp.
      Provides  an original 95% confidence  intervals to be used with  the
      initial  ICp when the number  of replicates per concentration is over  six
      and provides both original and expanded confidence  intervals when  the
      number  of replicates per concentration are less  than seven.

 7.6   ICPIN program output for  the  analysis of the Ceriodaphnia dubia
 reproduction  data  in Table M.I  is  provided in Figures  M.3 and  M.4.

 7.6.1  When the ICPIN  program  was  used  to  analyze this set of  data,  requesting
 80 resamples,  the  estimate  of  the  IC25  was 8.57% effluent.  The  empirical  95%
 confidence  intervals for the true  mean  were  8.30% to 8.85% effluent.

 7.6.2  When the ICPIN  program  was  used  to  analyze this set of  data,  requesting
 80 resamples,  the  estimate  of  the  IC50  was 10.89% effluent.  The empirical  95%
 confidence  intervals for the true  mean  were  10.36%  to  11.62% effluent.
                                       336

-------
      ICp  Data  Entry/Edit Screen
                 Current  File:
Cone. ID
Cone. Tested
Response 1
Response 2
Response 3
Response 4
Response 5
Response 6
Response 7
Response 8
Response 9
Response 10
Response 11
Response 12
Response 13
Response 14
Response 15
Response 16
Response 17
Response 18
Response 19
Response 20
1





















2





















3





















4



















.

• 'trm • w • ••<**•
5











































F10 for Command Menu
Use Arrow Keys to Switch Fields
Figure M.2.  ICp data entry/edit screen.   Twelve concentration identifications
            can be used.   Data for concentrations are entered in columns 1
            through 6.   For concentrations 7 through 12 and responses 21-40
            the data is entered in additional  fields of the same screen.

                                     337

-------
Cone. ID
Cone. Tested
Response 1
Response 2
Response 3
Response 4
Response 5
Response 6
Response 7
Response 8
Response 9
Response 10
1
0
27
30
29
31
16
15
18
17
14
27
2
1.56
32
35
32
26
18
29
27
16
35
13
3
3.12
39
30
33
33
36
33
33
27
38
44
4
6.25
27
34
36
34
31
27
33
31
33
31
5
12.5
10
13
7
7
7
10
10
16
12
2
6
25.0
0
0
0
0
0
0
0
0
0
0
*** Inhibition Concentration Percentage  Estimate ***
Toxicant/Effluent:
Test Start Date: app M   Test Ending Date:
Test Species: Ceriodaphnia dubia
Test Duration:             7-d
DATA FILE: cerioman.icp
OUTPUT FILE: cerioman.i25
Cone.
ID
1
2
3
4
5
6
Number
Replicates
10
10
10
10
10
10
Concentration
%
0.000
1.560
3.120
6.250
12.500
25.000
Response
Means
22.400
26.300
34.600
31.700
9.400
0.000
Std.
Dev. F
6.931
8.001
4.835
2.946
3.893
0.000
Pooled
Response Means
28.750
28.750
28.750
28.750
9.400
0.000
The Linear Interpolation Estimate:     8.5715   Entered P Value:  25
		
Number of Resamplings:   80
The Bootstrap Estimates Mean:   8.6014 Standard Deviation:     0.1467
Original Confidence Limits:   Lower:     8.3040 Upper:     8.8496
Resampling time in Seconds:     2.53  Random Seed: -1652543090
           Figure  M.S.   Example of  ICPIN program output for the IC25.
                                      338

-------
Cone. ID
Cone. Tested
Response 1
Response 2
Response 3
Response 4
Response 5
Response 6
Response 7
Response 8
Response 9
Response 10
1
0
27
30
29
31
16
15
18
17
14
27
2
1.56
32
35
32
26
18
29
27
16
35
13
3
3.12
39
30
33
33
36
33
33
27
38
44
4
6.25
27
34
36
34
31
27
33
31
33
31
5
12.5
10
13
7
7
i 7
10
10
16
12
2
6
25.0
0
0
0
0
0
0
0
0
0
0
*** Inhibition Concentration Percentage Estimate ***
Toxicant/Effluent:
Test Start Date: app M   Test Ending Date:
Test Species: Ceriodaphnia dubia
Test Duration:             7-d
DATA FILE: cerioman.icp
OUTPUT FILE: cerioman.i50
Cone.
ID
1
2
3
4
5
6
Number
Replicates
10
10
10
10
10
10
Concentration
%
0.000
1.560
3.120
6.250
12.500
25.000
Response
Means
22.400
26.300
34.600
31.700
9.400
0.000
Std.
Dev. 1
6.931
8.001
4.835
2.9415
3.893
0.000
Pooled
Response Means
28.750
28.750
28.750
28.750
9.400
0.000
The Linear Interpolation Estimate:    10.8931   Entered P Value: 50

Number of Resamplings:   80
The Bootstrap Estimates Mean:  10.9108 Standard Deviation:     0.3267
Original Confidence Limits:   Lower:    10.3618 Upper:    11.6201
Resampling time in Seconds:     2.58  Random Seed: 340510286
          Figure M.4.  Example of ICPIN program output for
                                     339
the IC50.

-------
                                  CITED REFERENCES


Bartlett, M.S.  1937.  Some examples of statistical methods of research in
      agriculture and applied biology.  J. Royal Statist. Soc. Suppl.
      4:137-183.

Conover, W.J.  1980.  Practical Nonparametric Statistics.  Second edition.
      John Wiley and Sons, New York, NY. pp. 466-467.

Dixon, W.J. and F.J. Massey, Jr.  1983.  Introduction to Statistical Analysis.
      Fourth edition.  McGraw Hill, New York, NY.

Draper, N.R. and J.A. John.  1981.  Influential observations and outliers in
      regression.  Technometrics 23:21-26.

Dunnett, C.W.  1955.  Multiple comparison procedure for  comparing several
      treatments with a control.  J. Amer. Statist. Assoc. 50:1096-1121.

Dunnett, C.W.  1964.  New table for multiple comparisons with a control.
      Biometrics 20:482.

Efron,  B.   1982.  The Jackknife, the Bootstrap, and other  resampling plans.
      CBMS  38, Soc.  Industr. Appl. Math., Philadelphia,  PA.

Finney,  D.J.   1948.  The  Fisher-Yates  test of  significance in 2X2 contingency
      tables.  Biometrika 35:145-156.

Finney,  D.J.   1971.  Probit Analysis.  Third  Edition.   Cambridge Press,  New
      York,  NY.  668 pp.

Finney,  D.J.   1978.  Statistical methods  in  biological  assay.  3rd  ed.  Charles
      Griffin  &  Co.  Ltd.,  London.   508 pp.

Hamilton,  M.A.,  R.C. Russo,  and  R.V.  Thurston.   1977.   Trimmed Spearman-Karber
      method for estimating  median  lethal concentrations.  Environ.  Sci.  Tech.
       11(7):714-719.

Marcus, A.H.  and A.P.  Holtzman.   1988.  A robust  statistical  method for
       estimating effects  concentrations in  short-term fathead minnow toxicity
       tests.   Manuscript  submitted  to the Criteria and Standards  Division,  U.
       S. Environmental  Protection  Agency, by Battelle Washington  Environmental
       Program Office,  Washington,  DC,  June  1988,  under EPA Contract No.
       69-03-3534.   39  pp.

Miller, R.G.  1981.   Simultaneous statistical  inference.  Springer-Verlag,  New
       York, NY.   299 pp.

Norberg-King,  T.J.  1993.   A Linear Interpolation Method for  Sublethal
       Toxicity:   The Inhibition Concentration (ICp) Approach. Version 2.0.
       National Effluent Toxicity Assessment Center Technical  Report 03-93,
       Environmental  Research Laboratory, Duluth,  MN 55804.  June 1993.

                                          340

-------
Pearson,  E.S.  and  T.O.  Hartley.   1962.   Biometrika tables for statisticians.
      Vol.  1.   Cambridge  Univ.  Press,  England,   pp.  65-70.

Scheffe,  H.   1959.   The Analysis  of Variance.   John  Wiley and Sons, New York,
      NY.   477 pp.                                                ,

Snedecor, G.W.  and W.G. Cochran.   1980.   Statistical  Methods.   Seventh
      Edition.   Iowa State  University  Press, Ames,  Iowa.   593 pp.

Steel, R.G.D.   1959.  A multiple  comparison  rank sum test: treatments versus
      control.   Biometrics  15:560-572.

Stephens, M.A.   1974.   EOF  statistics  for goodness of fet and some
      comparisons.   J.  Amer. Stat.  Assoc.   (JASA)  69:730-373.

USEPA.  1988.   An  interpolation estimate  for chronic  toxicity:  The ICp
      approach.  Norberg-King, T.J.  Technical  Report 05-88,  National  Effluent
      Toxicity  Assessment Center,  Environmental  Research  Laboratory,  U.  S.
      Environmental  Protection Agency, Duluth,  MN  55804.
                                                   •
USEPA.  1989.   Short-term methods  for estimating the  chronic  toxicity of
      effluents  and  receiving waters to freshwater organisms.   Second edition.
      Weber, C.I., W.H. Peltier, T.J. Norberg-King, W.B.  Horning,  II,  F.A.
      Kessler,  J.R. Menkedick, T.W. Neiheisel,  P.A.  Lewis,  D.J.  Klemm, Q.H.
      Pickering, E.L. Robinson, J.M. Lazorchak,  L.J.  Wymer, and  R.W.  Freyberg
      (eds.).   Second Edition.  Environmental Monitoring  Systems Laboratory,
      U.S. Environmental Protection Agency, Cincinnati, OH 45268.   EPA/600/4-
      89/001.
 '
USEPA. 1993. Methods for measuring the acute toxicity of  effluents  to
      freshwater and marine organisms.  Weber,  C.I.  (ed.).  Environmental
      Monitoring Systems Laboratory, U.S.  Environmental Protection  Agency,
      Cincinnati, OH  45268.  EPA/600/4-90/027F.
                                        341
   & U.S. GOVERNMENT PRINTING OFFICE: 1995 615-003/21017

-------

-------

-------

-------

-------

-------

-------

-------