United States
             Environmental Protection
             Off ice of Water
             Regulations and Standards (WH-553)
             Washington DC 20460
November 1980
An  Exposure
and Risk Assessment
for Trihalomethanes


This is a contractor's final report, which has been reviewed by the Monitoring and Data Support
Division, U.S. EPA. The contents do not necessarily reflect the views and policies of the U.S.
Environmental  Protection Agency,  nor does  mention of trade names or commercial products
constitute endorsement or recommendation for use.

PAGE j EPA-440/4-31-018 j
4. Till* td SuMKto
An Exposure and Risk Assessment for Trihalomechanes :
Chloroform, Bromoform, Bromodichloromethane, Dibromochloromethar
7. Autnortl)
Perwak, J.; Goyer, M. ; Harris, J.; Schimke,; Scow, K.; Wallace.D
9. torfermlni OmnluHm Nun* Mid Addiw
Arthur D. Little, Inc.
20 Acorn Park
Cambridge, MA 02140
12. Sponsoring Orgmuitlan Nun* and Addrms
Monitoring and Data Support Division
Office of Water Regulations and Standards
U.S. Environmental Protection Agency
Washington. D.C. 20460
1. Roclennrs Accaulon No.
5. Rtport Oito
November 1980
a, Performing Oranliillen Ropt. No.

10. Pfaioct/TMk/Work Unit No.
11. CantroetfQ or Grant(6) No.

                                         November 1980



                Joanne Perwak
Muriel Goyer, Judith Harris, Gerald Schimke,
       Kate Scow, and Douglas Wallace
            Arthur D. Little, Inc.
               Michael Slimak
    U.S. Environmental Protection Agency
           EPA Contract 68-01-3857
Monitoring and Data Support Division (WH-553)
  Office of Water Regulations and Standards
        Washington, D.C.  20460
           WASHINGTON, D.C.  20460

     Effective  regulatory  action  for  toxic  chemicals  requires  an
understanding of the human and environmental risks associated with the
manufacture, use,  and disposal of- the chemical.  Assessment  of risk
requires a  scientific judgment about  the  probability of harm to the
environment resulting from known or potential environmental concentra-
tions.   The risk  assessment  process  integrates health  effects data
(e.g., carcinogenieity,  teratogenicity)  with  information  on exposure.
The components of exposure include an evaluation of the sources of the
chemical, exposure pathways, ambient  levels,  and an identification of
exposed .populations including humans and aquatic life.

     This assessment  was performed as part of a program  to determine
the  environmental  risks associated  with  current use  and  disposal
patterns for  65 chemicals and  classes of chemicals  (expanded to 129
"priority pollutants") named in the 1977 Clean Water Act.   It includes
an assessment of  risk for humans  and aquatic life and  is  intended to
serve  as a technical  basis  for  developing  the  most  appropriate and
effective strategy for mitigating these risks.

     This  document  is  a contractors'  final  report.    It  has  been
extensively reviewed  by  the  Individual contractors ?nd by  the EPA at
several  stages  of  completion.   Each chapter of  the draft  was reviewed
by members of the authoring contractor's senior  technical staff  (e.g.,
toxlcologlsts,  environmental  scientists) who had  not previously been
directly involved in  the work.  These  individuals  were selected by
management  to  be  the technical  peers of  the   chapter authors.   The
chapters were  comprehensively checked  for  uniformity in quality and
content by  the contractor's editorial team, which also was responsible
for  the production  of  the  final  report.   The  contractor's  senior
project  management  subsequently  reviewed  the  final  report   in its

     At  EPA a  senior staff member was  responsible  for guiding the
contractors, reviewing the manuscripts, and soliciting comments, where
appropriate, from related programs within EPA (e.g.,  Office of Toxic
Substances,  Research  and   Development,  Air   Programs,   Solid  and
Hazardous  Waste,  etc.).  A  complete  draft was  summarized   by the
assigned  EPA  staff member  and  reviewed  for   technical  and  policy
implications with the Office  Director  (formerly the  Deputy Assistant
Administrator)  of Water  Regulations and  Standards.   Subsequent revi-
sions were  included in the final report.
                         Michael W. Slimak, Chief
                         Exposure Assessment Section
                         Monitoring & Data Support Division (WH-553)
                         Office of Water Regulations and Standards

                            TABLE OF CONTENTS

LIST OF TABLES                                                     vii

I.    EXECUTIVE SUMMARY                                              1

II.   INTRODUCTION                                                   7

III.  MATERIALS BALANCE                                              9

A.  Introduction                                                     9
B.  Chloroform                                                      10
    1.  Commercial Production and Use                               10
        a.  Commercial Production                                   10
        b.  Uses                                                    10
        c.  Production as a Contaminant                             10
    2.  Indirect Production                                         14
        a.  Chlorination of Water          '                         14
        b.  Industrial Processes                                    14
        c.  Automobile Exhaust                                      16
        d.  Atmospheric Formation                                   16
        e.  Marine Sources                                          16
    3.  Environmental Releases of Chloroform                        16
        a.  Releases During Commercial Production and Use           16
        b.  Releases from Indirect Production                       17
        c.  Releases of Chloroform That are Unaccounted For         18
C.  Bromoform, Bromodichloromethane, Dibromochloromethane           18
    1.  Commercial Production and Use of Bromoform                  18
    2.  Commercial Production and Use of Bromodichloromethane
        and Dibromochloromethane                                    19
    3.  Indirect Production                                         19
    4.  Environmental Releases                                      19
D.  Summary                                                         20
References                                                          24

IV.   DISTRIBUTION IN THE ENVIRONMENT                               27

A.  Introduction                                                    27
3.  Physical and Chemical Properties                                27
C.  Monitoring Data                                                 31
    1.  Introduction                                                31
    2.  Watar                                                       31
        a.  Seawatar                                                31
        b.  Freshwater                                              31
    3.  Sediment                                                    36
    i.  Aquatic Organises                                           36
    5.  Air                                                         36

D.  Environmental Pathways and Fate                              '   38
    1.  Overview                                                    38
    2.  Sources to the Environment                                  38
    3.  Inadvertent Formation of Trihalomethanes                    41
    4.  Fate in the Atmospheric Environment                         42
    5.  Fate in the Aquatic Environment                             42
    6.  Fate in Soil and Sediment                                   44
    7.  Fate .in Water Distribution Systems                          45
E.  Modeling of Chloroform in the Environment                       48
    1.  Estimated Environmental Distribution                        48
    2.  EXAMS Model Results                                         48
F.  Summary                                                         54
References                                                          55
V.    EFFECTS AND EXPOSURE  HUMANS                                59

A.  Human Toxicity                                                  59
    1.  Chloroform                                                  59
        a.  Introduction                                            59
        b.  Metabolism and Bioaccumulation                          59
        c.  Animal Studies                                          61
            i.    Carcinogenesis                                    61
            ii.   Mutagenesis                                       64
            iii.  Teratogenesis                                     64
            iv.   Other Toxicological Effects                       65
        d.  Human Studies                                           68
    2.  Bromoform                                                   69
        a.  Metabolism                                              69
        b.  Animal Studies                                          70
            i.    Carcinogenesis                                    70
            ii.   Mutagenesis                                       70
            iii.  Teratogenesis                                     70
            iv.   Other Toxicologic Effects                         70
        c.  Human Studies                                           71
    3.  Dibromochloromethane                                        71
        a.  Metabolism                                              71
        b.  Animal Studies                                          71
            i.    Carcinogenesis                                    71
            ii.   Mutagenesis                                       71
            iii.  Teratogenesis                                     71
            iv.   Other Toxicologic Effects                         71
        c.  Human Studies                                           72
    4.  Dichlorobromomethane                                        72
        a.  Metabolism                                              72
        b.  Animal Studies                                          72
            i.    Carcinogenesis                                    72
            ii.   Mutagenesis                                       72
            iii.  Teratogenesis                              .       "2
            iv.   Other Toxicologic Effaces                         ~_-
        c.  Human Studies                                           73
    5.  Overview                                                     3


3.  Human Exposure                                                  74
    1.  Introduction                                                74
    2.  Ingestion                                                   74
        a.  Drinking Water                                          74
        b.  Food                                                    75
    3.  Air                                                         78
    4.  Dermal Contact                                              3^
    5.  Conclusions                                                 34
References                                                          35
VI.   EFFECTS AND EXPOSURE  BIOTA                                91

A.  Effects of Trihalomethanes on Aquatic Organisms                9^
    1.  Introduction                                               9^
    2.  Chloroform                                                 9^
    3.  Bromoform                                                  92
    4.  Summary                                                    92
B.  Exposure of Aquatic Biota to Trihalomethanes                   95
References                                                         95
VII.  RISK CONSIDERATIONS                                          97

A.  Introduction                                                   97
B.  Risk .to Humans                                                 97
    1.  Effects of Trihalomethanes                                 97
    2.  Carcinogenicity of Chloroform                              97
    3.  Literature Review of Chloroform Risks                     104
    4.  Human Exposure Scenarios
    5.  Other Trihalomethanes
C.  Aquatic Biota                                                 2.09

                            LIST OF FIGURES
  No.                                                             Page

   1     Geographic Location of Trihalomethane                     13
         Producers in the U.S., 1978

   2     Materials Balance for Chloroform                          22

   3     Vapor Pressures of Trihalomethanes as a                   29
         Function of Temperature

   4     Major Pathways of Chloroform in the Environment           40

   5     Concentration of Chloroform in Water Column     ,          53
         Following Cessation of Discharge of 1.0 kg Hour

   6     Distribution of Concentrations of Total Trihalo-          76
         methanes in Major U.S. Drinking Water Supplies

                            LIST OF TABLES
 No.                                                             Page

  1.     Production and Uses/Releases of Chloroform, 1978       11,12

  2.     Production and Uses/Releases of Bromoform,                21
         Bromodichloromethane, and Dibrotnochloromethane, 1978

  3.     Basic Physical and Chemical Properties of                 28

  4.     Concentrations of Total Chloroform Detected in Surface    32
         Waters of the U.S., 1970-1979

  5.     Concentrations of Total Bromoform Detected in Surface     33
         Waters of the U.S., 1970-1979

  6.     Concentrations of Total Bromodichloromethane Detected     34
         in Surface Waters of the U.S., 1970-1979

  7.     Concentrations of Total Dibromochloromethane Detected     35
         in Surface Waters of the U.S., 1970-1979

  8.     Chloroform Residues Detected in Marine Organisms          37

  9.     Chloroform Concentrations Detected in Ambient Air of      39
         Industrial Areas

 10.     Rate Constants for Hydrolysis of Trihalomethanes          43

 11.     Trihalomethane Concentrations at Various Locations        46
         in POTW Systems (averaged for four cities)

 12.     Mass Flow Analysis of POTW Data for Trihalomethanes       47

 13.     Rate Constants Used in the EXAMS Analysis of the          49
         Aquatic Fate of Chloroform

 14.     The Fate of Chloroform in Various Generalized             51
         Aquatic Systems

 15.     Steady-State Concentrations of Chloroform in Various      52
         Generalized Aquatic Systems Resulting from Continuous
         Discharge at a Rate of 1.0 kg/hr

 16.     Incidence of Hepatocellular Carcinoma in S6C3F, Mica      62
         Exoosed to Chlorofora

                      LIST OF TABLES  (Continued)

 No.                                                             Page

 17.     Distribution of Modules and Fatty Cysts in the             62
         Livers of Chloroform-Treated Beagle Dogs

 18.     Acute Toxicity of Chloroform in Laboratory Animals         66

 19.     Concentrations of Chloroform, Bromoform,                   75
         Bromodichloromethane, and Dibromochloromethane,
         and Total Trihalomethanes in Water Supplies  HORS
         and MOMS

 20.     Sampling Locations for MOMS with Concentrations of         77
         Total Trihalomethanes Greater Than 200 ug/1

 21.     Estimated Human Exposures to Chloroform Via Drinking       79
         Water in the U.S.

 22.     Estimated Human Exposures to Trihalomethanes Via           80
         Drinking Water

 23.     Estimated Chloroform Exposures via Inhalation              82

 24.     Exposure of a Child to Trihalomethanes in Swimming         83

 25.     Effects of Chloroform on Aquatic Organisms                 93

 26.     Effects of Bromoform on Aquatic Organisms                  94

 27.     Adverse Effects of Chloroform on Mammals                   98

 28.     Adverse Effects of Bromoform on Mammals                    99

 29.     Adverse Effects of Dibromochloromethane on Mammals       100

 30.     Adverse Effects of Bromodichloromethane on Mammals       101

 31.     Carcinogenic Effects of Chloroform in Rodents            102

 32.     Probable Upper Bounds on Expected Excess Tumors per      105
         Million Population Due to Chloroform Exposure

 33.     Estimated Exposure of Man co Chloroform                  107

 3-.     Estimated Human Exposure co Bromoform,                   108
         Dibrotaochioromechane, and Bromodichloromenhane

     The Arthur D. Lictle, Inc. cask manager for this study was Joanne
Perwak.  Other major contributors were Muriel Goyer (human effects),
Judith Harris (Environmental Fate), Susan Coons (Environmental Fate), 
Douglas Wallace (biotic effects and exposure), Kate Scow (biological
fate), .Melba Wood (monitoring data), Joseph Fiksel (risk considerations),
Jane Metzger (editor), and Alfred Wechsler (technical review).  Irene
Rickabaugh was responsible for typing and preparation of the final draft

                             CHAPTER I

                          EXECUTIVE SUMMARY

     The Monitoring and Data Support Division, Office of Water Regulations
and Standards, U.S. Environmental Protection Agency is conducting an
ongoing program to identify the sources of and evaluate the exposure to
the 129 priority pollutants.  This report assesses the exposure to and
risk associated with trihalomethanes (chloroform, bromodichloromethane,
dibromochloromethane, and bromoform).  The summary is organized somewhat
differently than the report, focusing on the risk considerations first
since this section presents the major conclusions of the study.


     Chloroform has been shown.to be carcinogenic in rats and mice;
however, no evidence indicates mutagenicity in these species and evi-
dence for teratogenicity is contradictory.  Based on the multistage model
for risk extrapolation, estimates of per capita lifetime risks of cancer
ranged from 0.3 x  10-6 in a rural area with no water chlorination to
5.2 x 10-6 in an industrial area.  An estimated lifetime risk of cancer
for a child is 47  x 10~& due to swimming in chlorinated pools, inhalation,
and ingestion of chlorinated drinking water.

      Risks  due  to  other  trihalomethanes  cannot be quantitatively
 estimated.   They have  not  been shown to  be carcinogenic,  although they
 are all considered potential mutagens.   In addition,  exposure levels
 due to  dibromochloromethane,  bromodichloromethane,  and bromoform,  are
 generally considerably lower  than exposure levels for  chloroform.


      The information available regarding the  effects and  exposure  of
 aquatic biota to trihalomethanes  is  insufficient  to determine risk.
 However,  it appears  that levels observed in the  environment are  gener-
 ally well below levels at which effects  are observed.


      Studies regarding trihalomethanes have concentrated  on chloroform,
 since it is the most prevalent of the  four chemicals being  considered

      Ingestion  of  90-133 ag/kg of chloroform  has  been  shown to  induce
 boch hepacic and renal tumors  in  experimental animals,  although  some
 experiments ac  lower dosages  have yielded  contradictory results.   In
 addition, several  epidemiological studies  have suggested  an association
 between ingestion  of chlorinated  drinking  water  and a  higher incidence
 of cancer.

     No indications of nutagenic activity of chloroform  have  been  reported,
and evidence of teratogenicity is conflicting.  However,  inhalation of
chloroform by rats during days 6-15 of gestation resulted  in  a  high inci-
dence of fetal resorption and a few'cases of acudate  fetuses.

     Species-, strain-, and sex-related differences exist  with  respect  to
the acute lethal effects of chloroform, although exposure  to  chloroform
generally results in liver necrosis and kidney damage.   In man,  serious
illness has been reported following ingestion of 5 ml, although ingestion
of 120 ml of chloroform has been survived.  Dermal contact produces bum-
ing oain within a few minutes of exposure and, depending upon dose,
erythema, hyperamia, and vesication may also result.

     Limited toxicologic information is available for bromoform, dibromo-
chloromethane, and bromodichloromethane.   However, because of structural
similarities to chloroform, all three compounds' are cause  for concern with
regard to carcinogenic effects.  Preliminary results in  lung adenoma bio-
assays with bromoform and dichlorobromomethane support this concern.  In
addition, all three trihalomethanes are mutagenic in the Ames test.  To
date, other adverse health effects have not been demonstrated and  there-
fore, cannot be quantified.


     The chlorinacion of drinking water represents the largest source of
exposure of chloroform to'most humans in the United States; generally
ranging from 0.02-0.2 mg/day.  The maximum exposure due  to food was
estimated to be 0.04 mg/day.  However, very scant data are available
regarding levels of chloroform in food, and the 0.04 mg/dayestimate
is thought to be an upper range or maximum exposure.  Inhalation exposure
to chloroform is generally low; however in urban and highly industrialized
areas exposures have been estimated to be 0.02 mg/day and  0.5 mg/day,
respectively, and thus comparable with drinking water exposures in  areas
with chlorinated water supplies.  Dermal exposure can also be comparable,
primarily due to the rapid absorption of chloroform through the skin.
Swimmers, especially children, may receive up to 1.1 mg/day via this

     Exposure routes for the other trihalomethanes are believed to be
limited to drinking water, though no data are available regarding  levels
in other media.  Median exposures through drinking water are less  than
0.01 mg/day for bromoform, dibromochloromethane,  and bromodichloromethane.
However, maximum exposures of up to 0.6 mg/day have been estimated.


     Acute lethality to aquatic organisms due to chloroform and bromoform
generally occurs at concentrations greater than 10 mg/1,  as aeasured by
LC50 values.  A chronic value and a 96-hr LCjQ value for bromoform  in the
sheepshead minnow were reported :o be 0.2 mg/1 and 17.9 ag/1, respectively.
No data are available regarding the toxicity of dibroniochloromethane and
bronodichloromechane :o aquatic organisms.

"^               X

     Ambient concentrations of ,'chloroform in surface vater generally
fall between 0.1 ug/1 and 10 ug/.l.  Concencracions of 100 ug/1 were
exceeded only rarely, although7monitoring data are limited.  Concentra-
tions of the other trihalomethanes were lower than chloroform, generally
in the range of the detection limit.


     Approximately 158,500 kkg of chloroform were commercially produced
in the U.S. in 1978.  The predominant use (about 90%) is as a feedstock
for the manufacture of chlorodifluoromethane (Fluorocarbon-22).  The
pharmaceutical industry consumed about 3700 kkg (0.3%) and 42 kkg were
used for pesticides (0.02%).  Of the commercially produced chloroform,
about 11,600 kkg (7.3%) are unaccounted for.  A large part of this may
be used in laboratories, and some represents chloroform in stockpiles.
The remainder probably reflects uncertainties in estimates of the amounts
produced and used for other purposes.

     Chloroform is also inadvertently produced as a contaminant during
commercial production of other products such as vinyl chloride monomer
(VCM).  About 2700 kkg were produced in this process in 197.8.  Small
smounts of chloroform ( 54 kkg) are produced in the production of methyl
chloride, methylene chloride, and carbon tetrachloride.
     Chloroform is also produced inadvertently upon the reaction.of
chlorine with numerous organic precursors.  It is produced in chlorinated
drinking water, in the effluents of municipal and industrial wastewater;
and in chlorinated industrial cooling water.  The bleaching process in
the pulp and paper industry accounts for the single largest source of
this type, producing about 12,500 kkg of chloroform in 1978.  The chlori-
nation of drinking water, municipal wastewater, and industrial cooling
water produced 912 kkg, 91 kkg, and 3460 kkg, respectively.

     Chloroform may be produced as combustion product of leaded gasoline
(970) kkg  and as a decomposition product of trichloroethylene in the
lower troposphere (450 kkg).  In addition, it may be produced by marine
algae, although such a production has not been quantified.

     Of the identified environmental releases (20,600 kkg) of chlorofom,
93% (or 19,200 kkg) were released to air.  Of this, pulp and paper bleach-
ing is the largest contributor, accounting for 63% of the air releases.
About 17% was released as a result of chlorination of drinking water,
municipal wastewaters and cooling tower waters.  Losses from pharmaceuti-
cal uses accounted for about 3% of the releases to air.  Releases from
chloroform production and Fluorocarbon-22 production are relatively
small, .accounting for 27, and 1% of che total air releases, respectively.
The remaining  9 % is attributed co automobile exhaust, crichloroethylene
manufacture, VCM production, pesticide use, and transportation and
storage losses.  It should be pointed out that there is a great deal of
uncertainty  in these  estimates of releases due to  lack of data.

     Releases to water are relatively small (4% of the total).  Again,
pulp and paper bleaching appears to represent the major source to water
(44%).  The chlorination of water and the use of chloroform in the
pharmaceutical industry represent 21% and 30% of the total, respectively.
The remaining 5% is attributed to chloroform production and 7CM produc-
tion.  No water discharges have been identified from Fluorocarbon-22

     The disposal of chloroform on land represents only 3% of the total
known releases to the environment.  About 59% of the amount placed in
landfills is attributed to pharmaceutical operations, and 40% is attri-
buted to disposal of solid waste from VCM manufacture.  The remaining
1% is a result of solid waste from production of chloroform.

     Information regarding the production and uses of bromoform, bromo-
dichloromethane, and dibromochloromethane is very limited.  The produc-
tion of bromoform is thought to be less than 500 kkg, and its major use
is as an intermediate in chemical synthesis.  The other two trihalometh-
anes are thought to be produced only in laboratory quantities.

     The major releases of these chemicals result from their production
in water chlorination.  Total releases of 17 kkg, 832 kkg, and 204 kkg,
have been roughly estimated for bromoform, bromodichloromethane, and
dibromochloromethane, respectively.   It is expected that the, greater
portion of these amounts reach the atmosphere, as is the case for


     Trihalomethanes, especially chloroform, are commonly found in the
natural waters of the United States.  Levels in the open ocean appear
to be in the range of 1-10 ug/1.  The other trihalomethanes are not
commonly detected, and when found are at lower concentrations than

     Chloroform is also found in ambient air,  both indoors and outdoors.
Continental background levels range  from 0.04-0.13 ug/m3, while marine
background levels are somewhat higher.  Levels in urban areas are highly
variable, ranging from <0.05 ug/m3 to 90 ug/m3, with the maximum occur-
ring in a highly industrialized area.

     The fate of trihalomethanes is  largely controlled by their volatil-
ity.  While much of the chloroform is originally released to the air,
that reaching water is rapidly volatilized.   The half-life from a
stirred aqueous solution is 20 minutes.   The importance of hydrolysis.
adsorption, bioaccumulation, and biodegradation appear low compared to

     Once reaching Che atmosphere, chloroform, and other trihalomethanes
probably travel considerable distances before degradation occurs.   The
lifetime in the troposphere has been estimated to be 2-3 months.

attributable primarily to reaction with hydroxyl radicals.  Photochemical
degradation of trihalomethane, as well as rainout are not expected to be
important pathways.

     Volatilization is also likely from soil surfaces.  Biodegradation
and adsorption are not likely to be important fate processes.  Thus,
trihalomethanes which are not volatilized are subject to rapid movement

     A partitioning model suggests that at equilibrium, the trihalometh-
anes would be primarily (>99%) in the air compartment.  EXAMS (Exposure
Analysis Modeling System,  U.S. EFA, Athens, Ga.) also showed that
volatilization was the dominant fate process in lakes and ponds.  In
rivers, physical transport dominated volatilization as a removal mecha-
nism (over the short stretch of river modelled).

                              CHAPTER II

     The Office of Water Regulations and Standards, Monitoring and Data
Support Division of the Environmental Protection Agency is conducting
a program to evaluate the exposure to and risk of 129 priority pollutants
in the nation's environment.  The risks to be evaluated include poten-
tial harm to human beings and deleterious effects on fish and other
biota.  The goal of the task under which this report has been prepared
is to integrate information on cultural and environmental flows of
specific priority pollutants and estimate the risk based on receptor
exposure to these substances.  The results are intended to serve as a
basis for developing suitable regulatory strategy for reducing the risk,
if such action is indicated.

     This document is an assessment of risks associated with exposure to
crihalomethanes, including chloroform, bromodichloromethane, dibromo-
chloromethane and bromoform.  In an attempt to associate exposure to
specific sources, the report also provides information on the production
use, distribution, and fate of these chemicals.  There are several
difficulties in such an'assessment which should be mentioned.

     First, a significant route of human exposure is the inadvertent
production of trihalomethanes that results during the chlorination of
drinking water.   This production is difficult to quantify from a materials
balance point of view, however, the monitoring data for these-chemicals is
fairly extensive.  Thus,  while the exposure levels are fairly well docu-
mented, the total production of trihalomethanes in chlorination is not.

     Secondly, at least some data on most aspects considered here can be
found for chloroform.  However, data on the other trihalomethanes is
extremely sparse.  The lack of toxicological data makes the consideration
of risk of these chemicals inconclusive.

     Finally, the carcinogenic risk of chloroform has been extensively
considered by other authors.  The ranges in the values of estimated risks
give a good indication of the uncertainties involved in such estimations.
We have included a discussion of the ranges involved and the various
methods used in estimation of risk.

     The following sections of the report are as follows:

       Chapter III contains information on the production
        (commercial and inadvertent), releases, and disposal
        of triha1omethanes.

       Chapter IV describes available monitoring data and
        considers the fate of trihalomethanes in various

  Chapter V considers reported effect levels  (for
   laboratory animals) and estimated exposure  levels
   for humans.
                    ; l
  Chapter VI discusses reported effect levels and
   estimated exposurejilevels for biota.

  Chapter VII discusses risk to various subpopulations
   of humans and other biota.

                              CHAPTER III

                           MATERIALS BALANCE

     This chapter presents a materials balance for trihalomethanes,
considering the processes by which these chemicals are produced, their
various commercial uses, and the modes of disposition or disposal.  At
each stage in the life cycle of these chemicals, the amounts and forms
of releases to the environment are characterized and the environmental
compartments (air and water) initially 'receiving the pollutant release
are identified.

     A number of problems are encountered in attempting to develop a
comprehensive materials balance for the trihalomethanes.  Mot all
current production volumes, uses, and disposal practices are known with
any degree of certainty, and in many cases only sparse information is
available concerning environmental releases.  Nevertheless sufficient
information exists to characterize in general terms the environmental
releases of these chemicals.

     Of the trihalomethanes, chloroform is produced and consumed in a
far greater volume than all of the other trihalomethanes combined.
Consequently far more information is available concerning chloroform.
A detailed analysis of chloroform production and use was recently
published as a Level I Materials Balance by JRB Associates (1980) for
the U.S.  EPA Office of Toxic Substances.   This report served as the major
source document for the chloroform materials balance in this chapter and
was supplemented by other literature and some new analysis in specific

     No comparable analysis has been done for bromoform, bromodichloro-
methane, or dibromochloromethane.  Furthermore, the use of these
chemicals is so limited that information concerning them is considered
proprietary by the manufacturers and/or distributors.  This situation
severely limits the level of detail that could be achieved in the
materials balance developed for these substances.

     This chapter presents first a materials balance for chloroform by
considering its commercial production and uses, its production as a
contaminant, indirect production, and the environmental releases at
various stages of its production and use.  Then a materials balance is
developed for Che remaining trihalomechanes.


 1.   Commercial Production  and Use

 a.   Commercial Production

     Chloroform is  produced commercially  by  Che  successive  chlorinacion
 of  the methyl radical by one of  two  principal  chlorination  processes
 (using methane, or  methanol as feedstocks),  which  can be  controlled to
 produce methyl chloride, methylene chloride, and carbon tetrachloride,
 as  well as chloroform.   [Details of  these processes  are described  by
 JRB Associates (1980).]  In 1978 the methanol  process accounted  for
 122,500 kkg  (77%) of the chloroform  produced,  while  the methane  process
 accounted for 36,000 kkg (23%) (Table  1).  Because of the flexibility  of
 the production facilities to switch  processes, there is an  uncertainty
 of  greater than -10% (greater than -15,850 kkg)  associated  with  the
 total commercial production of 158,500 kkg of  chloroform  in 1978
 (JRB Associates 1980).

     Imports amounted to 7,670 kkg of  chloroform in  1978  and 7,900 kkg
 were exported (JRB  Associates 1980).   The locations  of chloroform
 production facilities and their capacities are shown in Figure 1.

 b.   Uses

     The predominant consumptive use of chloroform is as  feedstock for the
manufacture of chlorodifluoromethane (Fluorocarbon-22  or F-22),  though  there
is  some uncertainty as to the exact amount of chloroform consumed  in this
manner.   JRB Associates (1980) assumed that in 1978  90% of  the chloroform
produced was consumed as F-22 feedstock.   Other estimates indicate  that as
much as 93% may have been consumed for this use in 1973 (NAS 1978).

     The pharmaceutical industry is  a  consumer of approximately 3,700 kkg
 of  chloroform for use as a  solvent in  extraction processes  (JRB  Associates
 1980).  An estimated 42 kkg of chloroform is used  as pesticide.  The use
 of  chloroform as an industrial solvent, and  in the textile  and dye
 industries, was investigated by JRB  Associates (1980)  because of previous
 references to these categories of use  (NAS 1978) and found  to be insig-
 nificant and declining.

     Of the commercially produced chloroform,  about  11,600  kkg are
 unaccounted for.  Some of this is used in laboratories and  some  represents
 chloroform in stock piles.   The remainder may  reflect uncertainty  in the
 amounts used for other purposes.

 c.   Production as  a Contaminant

     Chloroform is  produced as a contaminant during  commercial production
 of  other products.  The same process that  produces chloroform is used  for
 Che commercial production of methyl  chloride, methylene chloride and
 carbon tetrachloride.  The  amount of chloroform  included  as impurities

                           Production (kkg)
Commercial Production                                   159,000
   Methyl Chloride Process                  122,500
   Methane Process                           36,000
   Loss during Production                       500

Imports                                                   7,670

Production as Contaminant                                 2,733
   Vinyl Chloride Monomer                     2,679
   CH3C1, CH2C12, and CC14                       54

Chlorination of Water                                     3,466
   Cooling Water                              2,460
   Potable Water                                912
   POTW1                                         91
   Swimming Pools                                 32
Bleaching of Paper Pulp                                  12,500

Automobile Exhaust

Photodecomposition of Trichloroethylene

Marine Algae


                          Uses/Releases (kkg)

Feedstock for F-22 Production                                    142,700

Exports                                                            7.900

Incinerated/Retained In Products/Storage                           3,968
   VCM Products                                          2,290
   Pharmaceutical Production                             1,610
   F-ll/F-12 Production (and others)                        47
   CHC13 Production                                         17
   Pesticide Production                                      4

Unaccounted for (including laboratory
   use and stockpiles)                                            11,600
                                      Air    Water  Land
Released to Environment             19.207    912   496           20,615
   CHC13 Production                    3704^  14     6
   'Pulp and Paper Bleaching         12,1002^ 400   	
   Chlorination of Water             3,2453' 221   	
   Pharmaceutical Extractions        1,523 "" 275   290
   Automobile Exhaust                  965 *  	   	
   Trichloroethylene Decomposition     450    	   	
   VCM Production                      187'     2   200
   Transportation and Storage Loss     177    	   	
   F-22 Production                     150'  	
   Pesticides                           38"  
TOTAL                                                            186.783

 Publicly Owned Treatment Works
"Arthur D. Little, Inc., estimate
 Arthur D. Little, Inc., after JRB Associates (1980)
 Average of JRB estimates of controlled (248 kkg) and
 uncontrolled (491 kkg) releases.

Source:  JRB Associates (1980). except as otherwise noced

    r.liluiulunu Pruilucuik
    1  Alliuil Uliuiniciil
       Miitmihvilli!. W.V/J.
    'I  Oidiiiniiil Slkiiniiick
       Uu-llu. W Vd
    3  IJllVU CllUIIIIUdl
    4  Uxvu Chuiiiicul
    b  iildiillur ChuiiiiCdl
       l.iiui&ville. Ky.
    G  Vulcan Miiuimli
       Gribiiui. La
    7  Viilc.ni MuluiiuU
       Wll.llllu. Kdll*

    Uiuinulurin 1'railin.ark
    U  Hum Clivniical
       Mnlluiiil. Mi.
    !)  Niiiiniidl Iliucliuiiiical
in chese products is estimated to be  54 kkg each year  (JRB Associates

     Chloroform is also produced during the manufacture of vinyl  chloride
monomer.  JRB Associates  (1980) estimated  that  this process contributed
2,679 kkg in 1978.  Most  of this is contained in the product, but some
is lost during the process.

2.   Indirect Production

a.   Chlorination of Water

     Chlorine reacts with organic precursors in vater  to  form chloroform
and other halomethanes  (NAS 1978; Trussel  and Umphres  1978; Norwood
t l. 1980).  The nature and amount  of haloforms produced depends upon
a number of factors, including water  source (surface,  groundwater),
organic content, temperature, form and amount of chlorine added,  etc.
Annual chloroform production associated with municipal water supplies,
municipal waste treatment facilities, and  treatment of industrial cooling
water has been estimated  to be 912 kkg, 91 kkg, and 2,460 kkg, respec-
tively (JRB Associates 1980).

b.   Industrial Processes

     Chloroform and associated halocarbons have been observed in  the
effluents of many industries.  In some cases, the presence of these
chemicals is believed to  result from  the direct use of the observed
chemical in the industrial process (NAS 1978).  However,  in at least
.some cases, relatively large amounts  of chloroform may be produced and
released during the manufacture of another product.

     The pulp and paper industry is recognized as a significant producer
of chloroform released to the environment  (NAS 1978).   Estimates  of the
total annual release from this source have ranged from 1.4 kkg/year (based
on concentrations in treated wastewate$ to a global estimate of 300,000 kkg/
year.  If one assumes that about 50% of worldwide pulp and paper  production
occurs in the U.S. (Arthur D.  Little,  Inc., estimate), the chloroform
releases in the U.S.  amount to about 150,000 kkg each year.

     The approach yielding the 1.4 kkg value is based on an assumed average
chloroform concentration  of 1,000 ug/1 in  finished effluent from  the waste-
water treatment plants (Lowenbach and Schlesinger Associates 1979).  This
approach does not account for the chloroform produced during the bleaching
process and lost directly to the atmosphere during the rinsing stages, or
the amount removed during treatment of the raw wastewater from the mill.

     The global estimate  is based upon an  industry-wide average chlorine-to-
chlorofora conversion efficiency of 6% (Yung eic ai.  1975).  JRB Associates
Inc.  (1980) report that 4.52 aay be a more appropriate conversion efficiency,
but also aoce that che validity of assuming a fixed chlorine-to-chloroforn

conversion race is suspect.  Unanalyzed factors that contribute to uncer-
tainty in this regard include process and equipment variations from one
mill to another, differences in processing related to types of wood chips
used, and lack of detailed knowledge of the functional groups within the
lignins that contribute to formation of chloroform.

     Though no comprehensive study of chloroform formation by the pulp
and paper industry has been conducted to date, some portions of the
problem have been investigated.  In 1977, the National Council of the
Paper Industry for Air and Stream Improvement (NCASI) monitored effluents
at 11 pulp mills with bleach plants (NCASI 1977).  The plants were
selected with a view toward picking bleaching sequences representative
of the industry as a whole (Andre Caron, NCASI, personal  communication,
May 1980).  An average of  0.37 kg of chloroform  (per kkg  of pulp bleached)
was found in the untreated mill effluent.  Ninety-five percent of the
chloroform was removed during treatment (aerated lagoons) of the mill
wastes prior to discharge  to natural waterways.  Most of  the reduction
was due to vaporization of chloroform to the atmosphere during treatment.
On the basis of NCASI (1977), JRB Associates (1980) calculated that
6,700 kkg of chloroform are produced in the untreated mill effluent.

     The NCASI (1977) study did not directly address the  question of
overall conversion efficiency of Cl to chloroform within the mill.
However, preliminary laboratory experiments conducted in  conjunction
with the study indicated that bleaching with hypochlorite would result
in a 1.0% conversion of chlorine to chloroform while bleaching with
chlorine water resulted in a 0.17% conversion factor.  These experiments
were reportedly set up _to  simulate the bleaching process  within various
pulp mills (Robert Claeys, NCASI, personal communication, May 1980).

     Thus, the above information suggests that previous estimates of
conversion rates have erred on the high side.  JRB Associates (1980)
reported the total amount  of chlorine (both Cl2 and hypochlorite, NaOCl)
used in the paper products industry in 1977 as 1.25 million kkg, but gave
no estimate of the quantity used for purposes other than  bleaching of
pulpwood.  Because of the  uncertainties regarding the rate of chlorine
conversion to chloroform in cooling water applications, as well as in
the bleaching process, and uncertainties regarding the chemical form of
the chlorine consumed, it  is assumed for this materials balance that 1%
of the chlorine consumed annually is converted in pulp mills to yield
12,500 kkg of chloroform.  However, it must be noted chat these estimates
are based upon limited laboratory testing,  When this is  compared with
the 6,700 kkg of chloroform production accounted for by JRB Associates
in the raw wastewater from the mills, the 5,800 kkg difference is believed
to represent an estimate of the amount of chloroform produced within the
mills chat escapes co the  atmosphere during pulp rinsing  and processing
stages, but prior to the discharge of the mill effluents  to the waste
treatment plane.  For purposes of che materials balance,  12,500 kkg is
estimated co be che cocal  amounc of chloroiora produced by pulp aills.

c.   Automobile Exhaust

     The combustion of leaded gasoline produces small quantities of
chloroform.  The details of the formation process are not known, but
JRB Associates (1930) calculated an assumed upper limit on this produc-
tion of 965 kkg per year, assuming that ethylene dichloride is the sole
source of chloroform in automobile gas, and that yields from this source
are as much as 1% of the ethylene dichloride combusted.

d.   Atmospheric Formation

     Laboratory experiments have shown that chloroform can be formed from
the degradation of trichloroethylene under conditions simulating those
found in the lower troposphere (Lowenbach and Schlesinger Associates
1979).  Based on laboratory work and their own estimates of trichloro-
ethylene released to the atmosphere, JRB Associates (1980) calculated
that 450 kkg of chloroform are produced annually by this mechanism,
although a high degree of uncertainty is associated with this number.

e.   Marine Sources

     Chloroform has been observed in marine waters and in marine algae
(NAS 1978).  However, the mechanisms for its production in marine algae
are not thoroughly understood, and the distribution of chloroform in
marine waters suggests possible anthropogenic rather than natural
sources.  Though the National Academy of Sciences (NAS 1978) concluded
that marine sources are negligible, the issue remains unresolved.  No
estimate of marine production can be made at present.

3.   Environmental Releases of Chloroform

a.   Releases During Commercial Production and Use

     Environmental releases of chloroform occur during its commercial
production and use and during transportation and storage.  Between 248 kkg
(controlled) and 492 kkg (uncontrolled)  of chloroform are believed to be
emitted to the atmosphere during the commercial production of chloroform.
Some 17 kkg of chloroform are incinerated in sludge, 14 kkg are discharged
to water, and 6 kkg are deposited on land during commercial production
(JRB Associates 1980).

     Only 150 kkg of chloroform are emitted into the atmosphere by the
process that accounts for the largest consumptive use of chloroform,
142,700 kkg consumed as a feedstock for F-22.  No releases to land or
water from this process have been identified (JRB Associates 1980).

     The pharmaceutical industry accounts for the largest release of
commercially produced chloroform.   Some 1,525 kkg are estimated to be
released co air, 275 kkg to water, 290 kkg to land,  and some 1,610 kkg
are incinerated vith waste products.  The uncertainties associated vi:h
all of these estimates are high (JRB Associates 1980).

     The use of chloroform as a pesticide  (fumigant) results in release
of 38 kkg (-11%) to the atmosphere and the incineration of 4 kkg
(JRB Associates 1980).

     A detailed description of the release of chloroform present as a
contaminant in methyl chloride, methylene chloride, and carbon tetra-
chloride, and their endproducts is given by JRB Associates (1980).  These
releases are dispersed and account for only 10 kkg of chloroform annually.

     The chloroform produced as a byproduct of vinyl chloride monomer
(VCM) manufacture is also described by JRB Associates (1980).  Their
calculations indicate that about 190 kkg  more than 100%) are emitted
to air, and 200 kkg (+130%, -71%) are deposited in landfill.  Water
discharges were not calculated, and 2,290 kkg (*5%) were assumed to be

b.   Releases from Indirect Production

     The largest environmental release of chloroform from a single
category arises from the pulp and paper bleaching industry (see Table 1).
Some 72% (12,500 kkg,) of the total estimated environmental loading
from indirect production (17,381 kkg) is attributed to this source.
Approximately 400 kkg (5% of the 6,700 estimated in untreated effluent)
are estimated to be discharged to surface waters and the remainder-
(12,100 kkg) are estimated to be released to the atmosphere.

     The next largest release is from chlorination of water, which
includes cooling water,  municipal water supply,  and municipal wastewater,
each of which are discussed by JRB Associates (1980) and shown in Table  1.
The amounts produced in these processes in 1978 are estimated to be
2460 kkg, 912 kkg,  and 91 kkg, respectively, most of which is released
to the atmosphere.

     Swimming pools have been found to contain high concentrations of
crihalomethanes.  Freshwater pools tend to have higher concentrations of
chloroform, while saltwater pools have higher levels of bromoform (Beech
e jl.  1980).  If the total volume of water in U.S. swimming pools is
assumed to be 2.6 x 10^- liters,  about equally distributed between the North
and the South (3-month and 5-month season, respectively), a release rate
to the atmosphere of 3.3 kkg/year can be calculated for chloroform
(Arthur D.  Little,  Inc., estimate).  Although this constitutes a rela-
tively small contribution to the materials balance, it may present a
significant exposure to humans because of the direct full-body contact
of those who swim.   The total environmental release of chloroform from
the chlorination of water is estimated to be 3,466 kkg,  of which 3,245
kkg are emitted to the atmosphere and 221 kkg are discharged to water
(JRB Associates 1980; Arthur D. Little, Inc., estimates).

     All of the estimated chloroform production from automobile exhausts
(965 kkg) and.atmospheric photodecomposicion of crichloroethylene (450
kkg) is released to the atmosphere.

c.   Releases of Chloroform That are Unaccounted For

     Some 11,600 kkg of commercially produced chloroform are unaccounted
for in che materials balance.  Although there is considerable uncertainty
regarding this number, a large portion is probably used in laboratories
and the rest is in stock piles.  It is quite likely that the predominant
release will be to the atmosphere, although it is unquantified.  A ma.ior
deviation from the conclusions of this section would occur if an
undetected, significant use resulted in significant releases to other


1.   Commercial Production and Use of Bromoform

     In contrast to chloroform, which is commercially produced in large
quantities and has been the subject of considerable study, bromoform,
bromodichloromethane, and dibromochloromethane are neither produced nor
used in large quantities, and have not been extensively studied.  Bromo-
form is commercially produced by the Dow Chemical Company and by National
Biochemical Corporation in plants located in Midland, Michigan, and
Chicago, Illinois (see Figure 1).  It is only one of many chemicals known
to be produced at these plants, and information is not available on either
the capacity or the actual volume of bromoform production at these plants.
The total production of bromoform in 1973 was estimated to be less than
500 kkg (MAS 1978).

     Information on bromoform use is also sparse.  The major use of the
chemical is as an intermediate in the synthesis of other chemicals, for
example, as a chain transfer agent in the preparation of certain polymers
(HAS 1978).  Other uses cited by Standen (1963) include use as a heavy
fluid in solids separation, and as a gage fluid.

     Bromoform has pharmacologic action as a sedative and antitussive
agent, but the extent of its use in these applications is undetermined.
There are indications that bromoform was formerly used as a sedative
(Strecher _e_al. 1968).  Bromoform was not cited in U.S. Pharmacopeia
(1978), a compendium of pharmacological agents.  A fraction of the
bromoform produced by the Dow Chemical Company may be used in pharma-
ceutical applications (SRI 1979).

     Bromoform has also been formulated into flame-retardant composi-
tions for cellulosic textile fabrics.  Use of bromoform is believed to
be limited in this application because after bromoform-based flame-
retardant fabric finishes have been applied, the fiber may become stiff
or discolored (Standen 1964).  As a result, other formulations that are
flame retardant and retain fiber character are considered preferable.

2.   Commercial Production and Use of Bromodichloromethane
     and Dibromochloromethane

     Bromodichloromethane and dibromochloromethane are synthesized by
the treatment of chloroform with a more than stoichiometric quantity of
anhydrous aluminum bromide or hydrogen bromide in the presence of a
catalyst.  The final steps in the process include a cool water wash to
remove inorganic material and distillation of the final product (Standen

     Data on the amounts of dibromochloromethane and bromodichloromethane
produced are unavailable, but it has been estimated that these chemicals
are commercially produced in laboratory quantities only (NAS 1978).  They
are distributed by Freeman Industries, Inc.; PRC, Inc.; and Aldrich
Chemical Company.  A major fraction of the Freeman Industries supply is
imported (Sales representative, Freeman Industries, personal communica-
tion, 1980).  Some portion may also be obtained domestically.  These
chemicals are sold for laboratory use, and no commercial use has been
identified  (NAS 1978).

3.   Indirect Production

     Bromoform, dibromochloromethane, and bromodichloromethane are all
found in chlorinated waters in association with chloroform, and are
believed to be produced as a result of the chlorinacion process (Rook
1976; Bellar e al. 1974).  In order to estimate the quantity of these
lesser trihalomethanes that is produced in association with the chlorina-
tion process, the generating mechanism is assumed to be similar to the
one that produces chloroform.  The generation rates are also assumed to
be proportional to the relative concentrations of these chemicals found
in the water supply as reported by U.S. EPA (1978):. bromoform concentra-
tion is equal to 0.005 of the chloroform concentration; bromodichloro-
methane equals 0.24 times chloroform concentration; and dibromochloro-
methane equals 0.059 times chloroform concentration.  When these ratios
are applied to the 3,466 kkg of chloroform .generated by water chlorina-
tion (see Table 1), the following estimates for 1978 have been calculated:

                  Bromoform               17 kkg

                  Bromodichloromethane   832 kkg

                  dibromochloromethane   204 kkg

4.   Environmental Releases

     The environmental releases of bromoform, bromodichloromethane and
dibromochloromethane are not defined any better than are their production
and uses.  Most commercially produced bromoform is used as a chemical
intermediate and either decomposed or recycled.  The 17 kkg of bromofora
produced in association vich chlorinacion of water could constitute the
                                   i o

major environmental release.  The commercially produced "laboratory
quantities" of bromodichloromethane and dibromochloromethane are small
in comparison with the amounts of these compounds that are believed  to
be generated as a result of chlorination of water and released directly
into the environment (832 kkg and 204 kkg, respectively).  Table 2
summarizes the available information concerning the production and uses/
releases of these chemicals.


     Trihalomethanes are released into the environment during commercial
production and use, their production as contaminants in other commercially
produced products, and their indirect production as a result of chlorina-
tion, and other processes, and possibly through natural production by
marine algae.  The chloroform, budget is more than 100 times larger than
the budget of the next most abundant trihalomethane, and has, therefore,
received much more attention in the literature than the others.  There
are still major areas of uncertainty concerning the magnitude of produc-
tion, use, and release rates so that it is necessary to estimate many
values for use in a materials balance for chloroform.  In recent work on
chloroform, JRB Associates (1980) have summarized the existing knowledge
and reduced the uncertainty associated with many of these estimates.  No
comparable work exists for bromoform, bromodichloromethane, and dibromo-
chloromethane.  These substances are produced in relatively small quanti-
ties, which are not Important commercially so that routine statistics on
production and use are unavailable.  The information that does exist is
scanty and yields only the most rudimentary materials balance.

     Figure 2 summarizes the materials balance for chloroform.  The  total
chloroform budget is estimated to be on the order of 187,000 kkg/year.
Imports and exports nearly balance each other at 4% of the total chloro-
form budget.  Commercial production accounts for 85% of production, and
1.5% of production occurs as a contaminant in other products.  Indirect
sources account for 9.3% of total production.  The largest uncertainty
with apparent environmental significance on the production side of the
budget is associated with the indirect sources (chlorination of water,
bleaching of paper pulp, automobile exhaust, photodecomposition of
trichloroethylene in the atmosphere, and marine algae).  No estimate is
available for the contribution of marine algae.

     The major use of commercially produced chloroform is the production
of chlorodifluoromethane (F-22), accounting for 76% of the total budget
(90% of commercial production).   Some 2% of the total budget is destroyed
(by incineration) or retained in storage,  and 11% of the total budget is
released to the environment.  JRB Associates (1980) estimated that
11,600 kkg chloroform (more than 6% of the total budget) was channeled
to "laboratory use and stockpiles," and no information is available
regarding the'amount released or destroyed from this use.

(]ii icKory/Source

   Conmiorcial Production  of  CliBr_
   Aqueous Production
   Marine Algae Production
   Chemical Intermediates
   He I cased Directly to Water  and Air
      from Aqueous Production
Hiomod Icliloromethane
   Commercial ProductJon
   A|iicous Production
   Ke I cased Directly to Water  and  Air
      from Aqueous Production
   Commercial Production
   Aqueous Production
                                                                     Annual  Budget (kkfi)
   Ke I eased Directly  to Water  and  Air
      from Aqueous Production
Laboratory Quantities
Laboratory Quantities


                                Laboratory Quantities


                                Laboratory Quantities


Source:  MAS,  1978, Artluir  I).  Little,  Inc., estimate, 1980.

I >
I i
                                  F-22 Fndtlock
                                   142.70O bky
                                                                                                                                        Imliietl Production
                                                                                                                                            I7.3S1 kkg
                                                                           CumiMicUl Piuilucllon
                                                                                159.0OO kk
Ptoducad H
  2733 kk(
                                                                                                                                                    ul Malm   211%
                                                                                                                                                 (licl.loiuclliyk.-iiu - 3%
                                                                                                                                              Aulo Exhaust -  6%
                                                                fUluied to               /^ All 03.2%
                                                               fciwifoinn.nl > 20.615 kk  
                      Suuiu: Auliui O I idle, luc
                                                         FIGURE 2   MATERIALS BALANCE FOR CHLOROFORM

     Of the 20,600 kkg of chloroform identified as being released  to  the
environment each year, 93%  (or 19,200 kkg) is released  to  the air, 3% to
the water, and 2% to the land.  The single largest contribution  to known
environmental releases is 12,500 kkg from pulp and paper bleaching,
accounting for 60% of the environmental releases tabulated.  The largest
unquantified source to the  environment is the contribution from  the
unaccounted for "laboratory use and stockpile" category.

     The largest source of  atmospheric and aquatic releases of chloroform
is pulp and paper bleaching, accounting for approximately 12,000 kkg
chloroform in air emissions each year or nearly two-thirds of known
atmospheric releases and 400 kkg in aquatic discharges or 44% of known
releases to water.  Chlorination of water is believed to contribute
17% (about 3200 kkg) of atmospheric emissions of chloroform, and 24%
(about 220 kkg) of total known aqueous discharges.  Major sources to
land include the pharmaceutical industry (about 290 kkg) and production
of vinyl chloride monomer (200 kkg).

     Such analyses contain considerable uncertainties and these releases
have been estimated as 'carefully as possible, often on the basis of very
limited data.   Therefore, the estimated releases should be considered in
that light.

    The limited information available concerning the other three trihalo-
methanes indicates that the predominant environmental release results
from their production in association with water chlorination.  No
environmental releases due  to their commercial production and use have
been identified.

Beech, J.A.; Diaz, R.; Ordaz, C.; Palomeque, B.  Nitrates, chlorates,
and trihalomethanes in swimming pool water.  Accepted for publication.
Am. J. Pub. Health; 1980.

Bellar, T.A.; Lichtenberg, J.J.; Kroner, R.C.  The occurrence of organo-
halides in chlorinated drinking waters.  J. Am. Water Works Assoc.
66:703-706;  1974.

Jolley, R.L.; Jones, G.; Pitt, W.W.; Thompson, J.E.  Chlorination of
organics in  cooling waters and process effluents.  Jolley, R.L. ed.
Proceedings  of the conference on the environmental impact of water
chlorination; 1975 October 22-24,  Oak Ridge, TN:  Oak Ridge National
Laboratory; 1976.  115-152.   Available from:  NTIS, Springfield, VA:

JRB Associates.-  Level I materials balance.  Chloroform.  Draft report.
Contract No. 68-01-5793, Task Order 18.  Washington, DC:  Office of
Pesticides and Toxic Substances, U.S. Environmental Protection Agency;

Lowenbach and Schlesinger Associates, Inc.  Chloroform:  A preliminary
material balance.  Contract W-1434-NNSX.  Washington, DC:  Office of
Toxic Substances, U.S. Environmental Protection Agency; 1979.

National Academy of Sciences (NAS).  Chloroform, carbon tetrachloride,
and other halomethanes. -An environmental assessment.  Washington, DC:
NAS;  1978.   294p.

National Council for Air and Stream Improvement (NCASI).  Analyses of
volatile halogenated organic compounds in bleached pulp mill effluent.
Stream Improvement Technical Bulletin No. 298.  New York, NY:  NCASI;

Norwood, D.L.; Johnson, J.D.; Christman, R.F.; Hass, J.R.; Bobenneich,
M.J.  Reactions of chlorine with selected aromatic models of aquatic
humic material.  Environ. Sci. Technol. 14(2): 187-189;  1980.

Pharmacopoeia of the United States.  12th Rev.  Washington, DC:  U.S.
Government Printing Office; 1978.

Rook, J.  Haloforms in drinking water.  J. Am. Water Works Assoc.  68(3):
168;  1976.

Standen, A.  ed.   Kirk-Othmer encyclopedia of chemical technology.
Mew York, MY:  Wiley/'Intarscience; 1963-

 Standeti,  A.   ed.   Kirk-Othmer  encyclopedia of  chemical technology.
 New York,  NY:   Wiley/Interscience;  1964.

 Standen,  A.   ed.   Kirk-Othmer  encyclopedia of  chemical technology.
 New York,  NY:   Wiley/Interscience;  1978.

 Stanford  Research Institute (SRI).   Directory  of  chemical manufacturers.
 Menlo  Park,  CA:   SRI  International;  1979.

 Stecher,  P.G. t  al.  ed.   The  Merck index.   8th ed.   Rahway,  NJ:  Merck
 and Co. Inc.; 1968.

 Trussel,  R.R.;  Umphres, M.D.   The  formation of trihalomethanes.   J.  Am.
 Water  Works  Assoc.  70:604-612;  1978.

 U.S. Environmental Protection  Agency  (U.S.EPA).   The national organic
 monitoring survey.  Unpubl.  Washington, DC:   Technical Support Division,
 Office of Water Supply, U.S. EPA;  1978.

 Yung,  V.L.;  McElroy,  M.B.;  Wofsy,  S.C.  Atmospheric  halocarbons:  a
discussion with emphasis  on chloroform.  Geophys.  Res.  Lett.  2(9):  397-
 399; 1975.   (As cited in  Lowenbach  and Schlesinger Associates 1979).

                              CHAPTER IV


     This chapter describes the important physicochemical properties of
Che trihalomethanes (Section B), provides detailed information on
observed ambient levels of the chemicals (Section C), describes the
important pathways and degradation routes in the environment  (Section D),
and gives the results of two different modeling efforts that were designed
to assist in the assessment of the major transport pathways (Section E).
A summary statement covering all aspects is provided in Section F.

     Modeling of the fate of trihalomethanes, specifically chloroform,
in environmental media was undertaken in order to illustrate important
aspects of the chemical's fate and transport in selected environmental
scenarios.  One model predicts the expected distribution of chloroform
in each environmental compartment (air, water, soil, sediment, biota)
under the assumption that all phases are in equilibrium.  This was done
in order to determine the tendencies of this chemical to be transported
in the environment.  The second model used is EPA's EXAMS (Exposure
Analysis Modeling System; U.S. EPA SERL, Athens, Ga.), which analyzes'
aquatic fate for various environmental scenarios.


     Table 3 summarizes the available basic chemical and physical
property data that are directly relevant to the environmental fate of
the trihalomethanes.  Of particular significance in terms of partitioning
within the environment are the vapor pressure, the water solubility, and
the octanol-water partition coefficient.  Complete or nearly complete
data were found for only chloroform and bromoform.  For the other two
trihalomethanes, the actual available data were minimal.

     However, for the environmentally important properties, the differ-
ence between the values for chloroform and those for bromoform
large, and it is reasonable to expect that values for bromodichloro-
aethane and dibromochloromethane will fall in the range defined by those
for CHC13 and CHBr3.  Thus, the important properties of the mixed trihalo-
methanes can be estimated with a considerable degree of confidence.

     Because of the importance of volatility in determining environmental
fate, the vapor pressure data were reviewed for bromoform and chloroform.
Figure 3 shows the temperature dependence of the vapor pressure for these
two soecies.  The corresponding regression equations are:

 Molecular Weight

 Melting Point (C)

 Boiling Point (C)

 tfaoor Pressure at 25 "C
-63.5 l
61.2 l
190 mm Hg
25% (v/v)
89-90 8 742 mm Hg 1
ca 50 mm Hg .
7% (v/v)
me thane
118-120 @ 742 mm Hg
ca 20 mm Hg
3% (v/v)
1 147.5 L
6.3 mm Hg
0.8% (v/v)
 ffacer Solubility at 25'C  7800 mg/1
                           65 x 10-3
1250 aig/1
4.9 x 10"3 J
 Qceanol-Water Partition       90
  ca 370
 Henry's Law Constant        2900
    igman (1962).

 ^Interpolated from Figure 3,  by use of the data of Hodgman (1962).

  Arcliur D.  Little, Inc.  estimate from b.p. data, assuming aH A of 8.7 kcal/mola.
  oifnfch is average of AHVa_ for CHCLj (7.9 kcal/mole) and CHBrj (9.6 ^cal/mole).

 "\yman (1978).

 xssam&d solubility of 30 x 10   M. average of value for CHC1. and CHBr..

 Sacfcer et al.  (1968).

 'Esttsch and Lao  (1979).
                          .00300          25C          .00400

                                    1/T K

      Source: Arthur 0. Little, Inc., based on Hodgman (1962).



          .   .   -2090  + 7.83, r  0.9999
          log P a -

where P equals vapor pressure (mm Hg) and T equals temperature  (K).

     The vapor pressures of chloroform and bromoform at 25C were cal-
culated from these equations.  The slopes of these lines correspond to
latent heats of vaporization of 7.9 fecal/mole for chloroform and 9.6
kcal/moie .for bromoform.  In order to estimate the 25*C vapor pressures
of the mixed trihalomethanes, the latent heat of vaporization was
estimated to be 8.7 kcal/mole, the average of the corresponding values
for CHC13 and CHBr^.  The estimated latent heat of vaporization (AHvap)
was then combined with the reported boiling point data (Hodgman 1962)
in order to estimate vapor pressures of 50 mm Hg for bromodichloromethane
and 20 mm Hg for dibromochloromethane at 25C.

     The probable range of solubilities for the mixed bromo/chloro
trihalomethanes may be defined by the 65 mM saturation concentration
for-chloroform in water (Lyman 1978) and the 4.9 mM value for bromoform
(Stecher et._!. 1968).  An average solubility of about 30 mM was esti-
mated for the mixed trihalomethanes.  This would correspond to about
5000 mg/1 of bromodichloromethane and about 6000 mg/1 of dibromochloro-
methane at saturation.

     The octanol-water partition coefficient provides a useful indica-
tion of tendencies for water-sediment and water-lipid partitioning.
Literature data on this property were available only for chloroform (90,
Hansch and Leo 1979).  The value shown in Table 3 for bromoform was
estimated from that compound's water solubility, according.to the linear
free-energy relationship approach of Hansch_et_al. (1968).  For alkyl
halides, the Hansch .et al. (1968) correlation between S (molar) and
partition coefficient (dimensionless) is:

          log P = 0.681 - 0.819 log S.

The estimated value of P for bromoform is, therefore, 370.  The parti-
tion coefficient, like the vapor pressure and solubility, appears to
vary by no more than an order of magnitude for the chemicals under
consideration.  A value of 250 was selected as an estimate for CHBrCl2
and CHBr2Cl.

     The Henry's Law constant listed in Table 3 is simply the ratio of
che vapor pressure at 25C co Che water solubility (with no corrections

for activity coefficients).  This parameter is fundamentally related  to
the tendency of a solute to escape from water into  the atmosphere.  Once
again, the range in values for chloroform and bromofora  is not laree.  An
average value of 2000 mm Hg - L/mole was assumed for CHBr?Cl and CHBrC^-


1.   Introduction

     This section summarizes data regarding the presence of trihalometh-
ane in the environmental media air, water, and biota.  In general, limited
data are available, especially for bromoform, bromodichloromethane, and
dlbromochloromethane, and no data were found regarding levels of trihalo-
methanes in soil.  Levels of chloroform in POTW influent and effluent are
discussed in Section IV.D.7.

2.   Water
     The only monitoring data available for trihalomethanes in seawater
were for chloroform.  Pearson and McConnell (1975) found a maximum
chloroform concentration of 1.0 ug/1 in Liverpool Bay, U.K.  Su- and
Goldberg (1976) found mean levels of 0.05 ug/1 in open waters of  the
East Pacific, and average concentrations of 0.009-0.012 ug/1 closer to
shore.  Murray and Riley (1973) reported mean chloroform concentrations
of 0.0008 ug/1 in the Northeast Atlantic.

b.   Freshwater

     STORET data provided the most complete survey of ambient concentra-
tions of trihalomethanes in surface waters of the U.S.  However,  the
data are far from complete:  only 168 unremarked^ measurements have been
reported for chloroform, the most commonly monitored trihalomethane.

     Summaries of the STORET data concerning trihalomethanes are  given
in Tables 4-7.  Most of the chloroform concentrations were between 1 ug/1
and 10 ug/1, and only rarely did they exceed 100 ug/1.  Most of the data
for bromoform were remarked; almost all of these concentrations were
between 1 ug/1 and 10 ug/1 as well.  For bromodichloromethane, the
majority of the measured concentrations fell between 0.1 ug/1 and 1.0
ug/1, with the exception of the Pacific Northwest, for which the  reported
levels were approximately one order of magnitude higher.  Two-thirds of
 A large number of data entries for  trihalomethanes in the STORET  system
 are "remarked."  This means  that some note regarding the data has been
 entered.   In general, the notation  refers to a value less than  the
 detection  liait.  Thus,  there is a  large amount of uncertainty  associ-
 ated with  "remarked" data.

                   IN SURFACE WATERS OF THE U.S., 1970-1979
River Basin
New England
Mid Atlantic
Great Lakes
Upper Mississippi
Lower Mississippi
Souris and Red of North
Arkansas and Red
Western Gulf
Rio Grande and Pecos
Lower Colorado
Great Basin
Pacific Northwest
United States           168  100        39        49         12           <1

 Unremarked data.
 Remarked data.



sles (
R2 :

3.1-1 us/1
1-10 us/1 10-100 uc/1
100-1.000 ut., i
Source:  U.S. Environmental Protection Agency, STORET (1979)

                   IN SURFACE WATERS OF THE U.S., 1970-1979
River Basin

New England
Mid Atlantic
Great Lakes
Upper Mississippi
Lower Mississippi
Souris and Red of North
Arkansas and Red
Western Gulf*
Rio Grande and Pecos
Upper Colorado
Lower Colorado
Great Basin
Pacific Northwest
  Slo    Percentage of Observations at Concentration
Ul R*
4 12
1 1
' 2
12 12
0.1-1 us/1
1-10 ug/1
10-100 us/1
United States
17  114
"TJnremarked data.
 remarked data.

Source:  U.S. Environmental Procection Agency, 5TORET  (1979)

                  DETECTED IN SURFACE WATERS OF THE U.S., 1970-1979
River Basin
New England
Mid Atlantic
Great Lakes
Upper Mississippi
Lower Mississippi
Souris and Red of North
Arkansas and Red
Western Gulf
Rio Grande and Pecos
Upper Colorado
Lower Colorado
Great Basin
Pacific Northwest
No    Percentage of Observations at Concentration



0.1-1 ug/1
1-10 us/1
.10-100 ug/1
United States

Unremarked data.
'Remarked data.
25   118
Source:  U.S.  Environmental Protection Agency,  STORET (1979)

                  DETECTED IN SURFACE WATERS OF THE U.S., 1970-1979
River Basin
New England
Mid Atlantic
Great Lakes
Upper Mississippi
Lower Mississippi
Souris and Red of North
Arkansas and Red
Western Gulf
Rio Grande and Pecos
Upper Colorado
Lower Colorado
Great Basin
Pacific Northwest
   No    Percentage of Observations at Concentration
Ul R2
2 12
1 7
h 1
12 12
0.1-1.0 us/1
1-10 us/1
United States
15   115
  hremarked Data.
"Remarked Data.
Source:  U.S. Environmental Protection Agency, STORET (1979)

Che dibromochloromethane concentrations also fell between 0.1 ug/1 and
1.0 ug/1, although for California and the Pacific Northwest concentra-
tions were roughly one order of magnitude greater.  The apparent higher
levels of these two chemicals in the Pacific Northwest and California are
probably due to a higher detection limit for data from these areas.

     Chloroform is the only trihalomethane for which' data were  found on
concentrations in precipitation.  Pearson and McConnell (1975)  reported
maximum concentrations of 0.2 ug/1 in rainfall near Runcorn, U.K.  Su
and Goldberg (1976) found a mean level of 17 ug/1 in rainwater  collected
in Southern California.  The same authors reported chloroform levels in
snow collected in different parts of North America, ranging from 3 ug/1
to 90 ug/1.

3.   Sediment

     The only information available on trihalomethane concentrations in
sediment was from Pearson and McConnell (1975) who found maximum chloro-
form residues of 4 mg/kg in Liverpool Bay, U.K.

4.   Aquatic Organisms

     STORE! data on trihalomethane residues in fish were limited.  Only
one unremarked measurement was reported for each of the four compounds.
All four measurements were taken in the Pacific Northwest basin, and
the concentrations reported all fell between 1 mg/kg and 10 mg/kg.  The
remarked data,  however, indicated concentrations from 100 mg/kg up to
100 mg/kg for each of the compounds.

     Chloroform is the only trihalomethane compound for which.any
detailed residue data were found.  The data of Pearson and McConnell
(1975) as shown in Table 8 include residue analyses for 21 species of
marine finfish' and shellfish, as well as plankton.  The plankton had the
lowest overall residue levels (0.02-5 ug/kg, wet weight); shellfish
residues ranged from 2 ug/kg to 180 ug/kg.  Residues in finfish were
analyzed in flesh, liver, and gastrointestinal tract.  Residues of up to
50 ug/kg were found in mackerel flesh; concentrations in liver and
gastrointestinal tract were generally one-third to one-tenth of those
found in flesh.

5.   Air

     Chloroform is the only trihalomethane compound for which data on
concentrations in air were found.  Harsch .et_al.  (1977) have reported
that continental background levels of chloroform in the atmosphere
range between 0.04 ug/m3 and 0.13 ug/m^.   Oceanic background levels,  as
reported in NAS (1978), are somewhat higher,  with mean concentrations
ranging from 0.13 ug/m^ to 0.20 ug/m^.

     In the urban environment, ambient atmospheric levels of chlorofom
may be considerably higher as a result of anthropogenic sources.  The
data of Harsch et_ al.  (1977) suggesc that concentrations in city air
are highly dependent upon automobile activity.   Urban air (Pullman,

Chloroform Concentration
   (ug/kg wee weight)
Plankton                                               0.02-5
Mussel (Mytilus edulis)                                 3-10
Cockle (Cerastodenna, edule)                             4-150
Oyster (Ostrea edulis)                                     3
Whelk (Buceinum undatum)                                 10
Slipper limpet (Crepidula fornicata)           '            6
Crab (Cancer pagurus)                                   3-180
Shore crab (Carcinus maenas)                             15
Hermit crab (Eupagurus bernhardus)                     20-73
Sand shrimp (Crangon crangon)                            45
Starfish (Asterias rubens)                               13
Sunstar (Solaster sp.)                                     3
Sea urchin (Echinus esculentus)                            2
Flounder (Platychthys flesus). flesh/liver               21/6
Mackerel (Scomber scomb'rus), flesh/liver                 50/18
Dab (Limanda limanda), flesh                             23
Plaice (Pleuronectes platessa). flesh                    17
Sole (Solea solea). flesh/gastrointestinal tract         26/9
Red gurnard (Aspitrigla cuculus). flesh/gastrointestinal 21/2
Scad (Trachurus trachurus), flesh                        48
Pout (Trisopterus luscus), flesh                         15
Spurdog (Squalus acanthias), flesh                      110
Source:  Pearson and McConnell  (1975)

Washington) in the early morning contained 0.04 ug/m^ chloroform;  in
light traffic the level was 0.10 ug/m3; and in heavy traffic with  no
breeze, the concentration rose to 0.44 ug/m^.  Chloroform levels in
automobile exhaust varied between 0.32 ug/m3 and 33 ug/m^, depending on
car model and year.  (The car equipped with a catalytic converter  emitted
fumes with chloroform levels approximately two orders of magnitude lower
than a car that lacked this device.)  Lillian est, al. (1975) found  a
wider range of urban ambient levels, ranging from <0.05 ug/m^ to 74 ug/m^;
the maximum value was reported for Bayonne, N.J.

     Pellizzari et al. (1979) have sampled four highly industrial  areas
for halogenated organic compounds.  Chloroform was commonly found  in
these samples, although none of the other trihalomethanes was reported.
Table 9 summarizes the results.  It can be seen that chloroform levels
in air were highly variable, with a maximum of 139 ug/m3 reported  in the
New Jersey area.
     Harsch and Rasmussen (1977) have conducted extensive monitoring in
various indoor environments,.particularly in commercial establishments.
Concentrations ranged from 0.07 ug/m3 in the dining area of a restaurant
to 3.6 ug/m3 in the aisle of a food store containing household cleaners.
However, chloroform levels rarely exceeded 0.5 ug/m3, and most concen-
trations were similar to tropospheric background levels.

     However, Pellizzari e,t al. (1979) sampled basements of houses in
the Old Love Canal area, and found levels of chloroform ranging from
0.46 ug/m3 to 13.48 ug/m3 in the seven samples taken.  Upstairs in the
same homes concentrations from trace levels to 15.3 ug/nn chloroform were


1.   Overview

     This section examines the sources of trihalomethanes released to
the environment, as discussed in Chapter III, the chemistry of their
inadvertent production, and the fate of these chemicals in various
environmental compartments.  Most of this discussion is centered around
chloroform, but the general fate of the other trihalomethanes is expected
to be similar.

2.   Sources to the Environment

     The largest of the identified sources of chloroform released  to
the environment are pulp and paper bleaching, the chlorination of
drinking water and wastewater, and the use of this chemical in pharma-
ceutical extractions.  The greater portion of these releases goes  to
air (-^17,000 kkg/yr), while about 900 kkg/yr are released to water.
figure 4 shows a general representation of the releases of chloroform,
as well as their ultimate distribution in the environment.

City                              Mean        Range    No. Detected/No.  Samples

Niagara Falls and Buffalo, N.Y.   89,000   1050-105,461           9/9
Rahway/Woodbridge, Boundbrook    47,000   T  - 98.625'
   and Fassaic, N.J.
Baton Rouge, Geismar, and
   Plaquemine, LA
5,500   181-11,742
Houston, Deer Park and
   Pasadena, TX
1,000   T - 53,846
Source:  Pellizzari et al.  (1979)

                   Ti*nt|MMl 11 mil
                        l Souitci
                                                                                                                       Reaction wviih hydfoxy rtdicnl
                                                                                                                            (lu~2 3 months)
                   Aic< Sourc*
Wei and Dry
                                   SulljCt Wilui 4111!
                                                                                                                                             Phoiochnmical UvyitilHiou in kVaiot
                                                                                                                                          Traiitpoii 10 Deep Waioci wd SoJiincnls sli
                                                                                                                                                   (lor.l n:s.iloi:u limit)
                                I e jthliuj 10 0

3.   Inadvertent Formation of Trlhalome thanes

     Since Che formation of trihalomethanes upon chlorination has been
shown to be an important source of these chemicals in the environment,
it is important to consider the chemistry of these reactions.

     Fulvic and humic acids constitute the major portion of the organic
matter in surface or ground waters and have been demonstrated to be
precursors for halo forms when treated with chlorine  (Rook 1974) .  The
reactions cited are direct analogs to the haloform reaction; these
reactions are initiated by the hypochlorite Ion present in chlorinated
waters .

     One suggested mechanism for trihalome thane formation during
industrial bleaching and water purification processes is given in
reaction 1 (Hendrickson .et al. 1970) .  This reaction is a direct  P,
analog of the haloform reaction and is seen with methyl ketones (-C-CB^) or
compounds that can be readily oxidized to produce methyl ketones.  The
reaction is base catalyzed; since each halogen introduced increases the
acidity of the remaining hydrogens, the intermediates are unstable and
trihalomethanes are produced.
            0                                   0
          R-C-CH3 + 3X2 + 40H  - CHX3 + 3X  + R-C-0  + 3H20         (1)

where X = Cl and/ or Br (Hendrickson t al. 1970)

     The presence of bromide ion in aqueous systems has been shown  to
reduce substantially the yield of chloroform, while increasing levels of
dichlorobromomethane, bromodichlorome thane, and bromoform  (Bunn et  al.
1975).  Available hypochlorite ion oxidizes bromide to give bromine,
which then undergoes the "haloform" type reaction to form tne appropri-
ate bromochloromethanes as shown in reaction 1.

     The above mechanism is certainly applicable to industrial chlorina-
tion processes; however, the pH range of municipal water chlorination
is much closer to neutral.  For this reason, and the observation that a
variety of degradation products other than trihalomethanes were produced
in nonbasic aqueous solutions, a mechanism other than the base-catalyzed
halogenation was suspected.  For chlorination in aqueous solutions, Rook
(1977) proposes a slightly different pathway and has determined that the
most reactive site  for the haloform reaction with fulvic acids at
neutral pH is at the carbon between two meta-OH groups.  His mechanism
is outlined in reaction 2.  The incorporation of bromine is according
to the pathway outlined above.
       OH                   UOH
              - 4 HOC:
                                                       '(Rook 1977)   (2)


4.   Fate in the Atmospheric Environment
     Versar (1979a,b) has concluded on Che basis of a review of Che licer-
aCure that;"once chloroform is in Che troposphere, reaction with hydroxyl
radicals is;1, Che primary mechanism for removal.  They reported, based upon
Che work o$ Spence at 1. (1976), Hansc (1978), and Che U.S. EPA  (1975),
chac che produce of chis reaccion is a CC13 radical, which is oxidized
Co produce phosgene (COC12) and chlorine oxide (CIO).  The estimated
lifetime is reported Co be 2-3 months, based on reported lifetimes of
0.19 years and 0.32 years reported by Cox et al. (1976) and Yung  et al.
(1975).  Bromo-alkanes generally have rate conscants for reaction with
hydroxyl radical that are the same or slightly higher than those  for
the chlorinated analogs.  Thus, the atmospheric residence times of
CHBrCl2 CHBr2Cl and CHBr3 are also expected Co be on Che order of 2 co
3 months.

     Direct photochemical degradation of trihalomethanes in the tropo-
sphere is not likely to be important because these species do not absorb
UV radiation at wavelengths that penetrate the ozone layer.  The  esti-
mated atmospheric residence time is sufficiently short that extensive
transport to the stratosphere, where direct photochemical process could
occur, may not be a major pathway.  Export from the atmosphere by rainout
is a ..possible, but not probable fate.  The magnitude of the vapor pres-
sure in comparison with the solubility suggests that an insignificant
fraction of chloroform in the atmosphere would be associated with water
droplets or dust particles.  Since the other trihalomethanes have similar
Henry's Law constants, they also are expected to be resistant to  rainout.

     Thus, trihalomethanes reaching the atmosphere may travel consider-
able distances before reaction with the hydroxyl radical results  in

5.   Fate in the Aquatic Environment

     Of the numerous modes by which the trihalomethanes might be  removed
from the aquatic environment, volatilization is expected to be the most
significant.  A laboratory study by Billing et. al. (1975) indicated that
the half-life for evaporation of chloroform from a stirred aqueous
solution was about 20 minutes, compared with a 1-minute prediction based
on vapor pressure and solubility.  In fact, the results obtained  strongly
suggest that the mass transport phenomena were limiting the evaporation
rate, since the same half-life (20 minutes) was observed for several
alkylhalides with vapor pressures from 19 mm to 426 mm.

     The rate conscants for hydrolysis of the bromo/chloro trihalometh-
anes, as compiled by Habey and Mill (1977), are summarized in Table 10.
The reactions are base catalyzed, but the race constants for che basic
hydrolysis are aoc high enough to decrease significantly the predicted
hydrolytic half-life within the normal environmental ?H range (?H 5-3
for most natural waters).  Chloroform, for example, has a predicted half-
life of_>500 days at pH 7 due to neutral hydrolysis, while the basic

                                              Rate Constant
                         Temp.        Neutral              Basic
Compound                 (C)        lO^sec"1       104kOHM~1sec"L
Chloroform                 25           1.621

                          100           7.29

                           25                              0.60

Bromodichloromethane       25                             16.0

Dibromochloromethane       25                              8.01

Bromoform       '25                              3.20
 Considered to be "suspect" value by Mabey and Mill (1977)
 because of poor temperature control during measurement.
 If the 100'C value is considered to be reliable, a 25 value
 much lower than 1.62 (ca 0.07 perhaps) would be more likely
 for a reaction with a typical activation energy.
Source:  Mabey and Mill (1977)

hydrolysis  reaction at  pH 8  vould give  a 500-year half-Life.   The base-
catalyzed reaction  rate constants of  Table 10  are included because they-;
provide  one indication  of the  relative  reactivity of the trihalomethanes.
With  regard to basic hydrolysis,  Che  relative  reactivities are:   chloror-
form  (1), bromoform (5),  dibromochloromethane  (13),  bromodichloromethane
(27).  It is not  clear  that  the same  order of  magnitude vould be observed
for the  neutral reactions, because the  reaction mechanism may be signifiV
cantly different.   However,  it appears  probable that the mixed trihalo-
methanes have considerably shorter hydrolytic  half-lives than bromoform
or chloroform.  It  also appears likely  that hydrolysis  is not fast
enough under any  conditions  to compete  with volatilization as a  pathway
for loss from the aquatic compartment.

     The octanol/water  partition  coefficient of 90 suggests that a
significant portion of  chloroform in  the hydrosphere is associated with
suspended solids  and sediments (Lyman 1978).   However,  there  is  no
experimental evidence to  support  this as an important pathway, and no
monitoring  data exist 'for chloroform  in sediment.  Thus,  it is not clear
that adsorption competes  with  volatilization as an important  fate path-

     Similarly,  no  information is available regarding the biodegradation
of trihalomethane in aqueous systems.   However, again it  is unlikely that
this mechanism could compete with volatilization.

     Bioaccumulation does not appear  to represent an important fate
pathway,  although the octanol-water partition coefficient of 90 suggests
that accumulation may occur.   In a summary of bioaccumulation studies on
various aquatic animals, U.S. EPA  (1978) found a mean bioconcentration
factor of 6, following 14 days of exposure to a 110 ug/1 solution of
chloroform.   The half-life of chloroform in the tissues was found to be
less than 24 hours.   Preliminary accumulation tests on channel catfish
by Anderson _et_al.  (1979) indicated a bioconcentration factor of approxi-
mately 10 over ambient water levels, with the chloroform levels in fat,
eggs,  and gastrointestinal system at  least twice those of levels in
other tissues.

     Anderson et al. (1979) also conducted a 28-day uptake/28-day
depuration study of bromoform, using menhaden,  oysters, shrimp, and two
species of clams  (all marine species).  After 24 hours or exposure to
bromoform,  all species were found to have bromoform residues in the
tissues,  and within a 48-hour depuration period bromoform residues
disappeared.  In the highest bromoform concentrations (0.21 mg/1 to
19 mg/1,  depending on the species), tissue levels were usually lower
than those  in the water; in low concentrations  (0.03 mg/1)  the reverse
was true.

6.   Fate in Soil and Sediment

     Little  is known regarding the fate of trihalomethanes  in soils and
sediment.  Volatilization is likely from surface soils, and is probably

 Che  dominant  pathway.   However,  some movement may occur, as evidenced by
 some detectable levels  of  chloroform in ground waters  (Coniglio e al.

      Lyman (1978)  suggests that  some chemical degradation may occur.
 Adsorption is also not  likely  to be  important because  of the high solu-
 bility and vapor pressure  of the trihalomethanes.  The importance of
 biodegradation  is  unknown.

      One  recent study by Wilson  et al.  (1980) investigated the fate of
 chloroform and  dichlorobromomethane  in  a sandy soil.   In a soil column
 receiving solutions  of  these chemicals,  most  was  volatilized (66-122%)
 and  31-48% was  recovered in the  column  effluent.   Actually,  more was
 recovered than  applied.  This  excess was attributed  to an overestimation
 of the amounts  volatilized.  Degradation in the column was thought to
 be insignificant.  These authors calculated retardation factors, which
 represent the interstitial water velocity/velocity of  the pollutant.
 The  finding that these  factors were  less than 1.5 for  both chemicals
 suggests  that little adsorption was  occurring.

      Thus,  trihalomethanes are likely to be volatilized from surface
 soils.  Otherwise, they are subject  to  migration  into  aquifers.

 7.    Fate in Water Distribution  Systems

     Because of the importance of the chlorination process in trihalo-
methane formation,  some consideration was given to the  fate of trihalo-
methanes in water distribution systems.   An EPA study of the sources of
priority pollutants found in publicly owned treatment works provides a
means of estimating the extent of loss and regeneration of trihalomethanes
in aqueous solutions (Levins it al.  1979a).  Samples were collected  at  the
POTW, as well as at various source locations  (residential, commercial, and
industrial).  The summary data from four separate studies are presented in
Table 11.   The concentrations  of the  trihalomethanes were consistently
higher in the samples of tap water than in the POTW influent samples, indi-
cating either dilution or volatilization is occurring.  The relative
proportions of the concentrations of  the four trihalomethanes are  fairly
consistent from the tap to the sewers to the influent.   Data for bromo-
form are inconclusive because  of the very low concentrations reported.
The data indicate that the rates of dilution or volatilization are similar
for the four compounds.   They  also suggest that levels of these  compounds
are not being augmented by mechanisms other than chlorination.  The  sources
sampled in this study did not  include any paper and pulp industries.

     Table 12 presents the mass flows for one city (Levins e_t al_.  1979b).
It can be seen that significant percentages of the trihalomethanes pro-
duced on chlorination are lost between the tap and the  influent, either
through volatilization or consumptive use of tap  water.  Furthermore, a
comparison of the sum of the mass flows  for each  of the scaled-up source
categories with the mass flow  at the influent  shows that very little loss
takes place within  the sewer system.

                                     IN POTW SYSTEMS (averaged for four cities)
                               Influent to
 Det.    Mean    Det.   Mean    Det.   Mean    Det.    Mean    Det.   Mean
Freq.    Cone.  Freq.   Cone. ' Freq.   Cone.   Freq.    Cone.  Preq.   Cone.
 (%)    (UR/D   (%)    (iie/l)   (%)    (ug/l)    (Z)   dig/1)   (%)    (iiB/l)
Itr01110J iclilorometliane
I)J liromoclilorometliane
                       0.7     57
    Source:   Levins et al.  (1979a)

                       TABLE 12.  MASS FLOW ANALYSIS OF POTW
                                  DATA FOR TRIHALOMETHANES
                      Tap .
       Compound      Water'
                                            Mass Flow (kg/dav)
Sources to   Influent Co
   POTW3        POTW4
            Influent/Sum   Tap/Influent
Oibromochloromethane  0. 71
        Calculated by use of the assumption that tap flow equals influent flow.

        flow of trihalomethanes at the tap.

        Sum of all mass flows to POTW.

        Measure of mass flow at influent to POTW.
       Source:   Levins t al.  (1979b)


1.   Estimated Environmental Distribution

     A preliminary estimate of the environmental distribution of the
bromo/chloro trihalomethanes was attempted by application of the simple
model of Neely (1978).  Neely's model consists of regression equations
that correlate the percentage distribution of a chemical among air,
water, and soil compartments with the Henry's Law constant and the
aqueous solubility.  These equations predict that, at equilibrium, all
of the halomethanes under consideration would be almost entirely in the
air compartment, with 1% or less in either the water or soil compart-
ments.  It is important to note, however, that the equilibrium distribu-
tion among phases may not be attained in an actual situation.  For
volatile species such as trihalomethanes, the intercompartmental
distribution is affected by kinetic, as well as thermodynamic, factors.

     The probable distribution of the residual amounts of trihalomethanes
between water and soil/sediment within the aqueous compartment can be
predicted from the soil adsorption models developed at the U.S. EPA
Athens laboratory (Means  al. 1978, Karickhoff ej: al. 1979).  This
suggests that the ratio of sediment concentration to water concentration
is related to the water-octanol partition coefficient:
           sediment  _P^ x organic content of sediment
          c           i. o*

     For a typical river sediment containing 2% organic carbon or less,
the sediment to water concentration ratio for the halomethanes ranges
from about 1 (chloroform) to about 5 (bromoform).  Trace amounts of
trihalomethanes that have not (yet) volatilized may thus be expected to
be found in both soil and aqueous phases.

2.   EXAMS Model Results

     For the purpose of examining the probable fate of chloroform in
various aquatic environments under conditions of continuous discharge,
the EXAMS (Exposure Assessment Modelling System; U.S.  EPA-SERL, Athens,
Ga.) model AETOX 1 was implemented.

     Rate constants for the primary processes thought to influence the
fate of chloroform in aqueous environments were estimated (SRI 1980)
and are presented in Table 13.

     A loading rate for trlchloromethane was estimated using the follow-
ing assumptions for a pulping plant:   0.1 Ib trichloromethane discharged
per ton of pulp production,  500 tons of production per day,  therefore,
2 Ibs or 1 kg trichloromethane released per hour (JRB  Associates 1930
and Andrew Caren,  NCASI, personal connaunieation 1980).   This astizrate
is not meant to be representative of the industry but  to provide a
reasonable example of a discharge rata.)


Race Constant
Volatilization                     0.583
Hydrolysis (base-catalyzed)        0.23
                     mole [OH"]-1 hr"1
Hydrolysis (neutral)               5.40xlO"9
Oxidation (water column)           0.70
                  mole (oxidants)   hr
Oxidation (sediment)             360.0
                  mole (oxidants)'  hr
Source:  SRI (1980)

     Six prototype systems were simulated in order to provide a range
of environmental conditions:  pond, eutrophic and oligotrophic lakes,
and three river systems (average, turbid, and coastal plain).

     As would be expected, in relatively static systems  (pond and lakes)
in which physical transport processes did not dominate, volatilization
was the most important removal mechanism, responsible for a 93-96% loss
of the equilibrium chloroform mass.  Table 14 presents information on
the distribution and transformation of chloroform in the different
systems.  In all the river systems (1-km segments), transport downstream
alone accounted for at least 85% of the removal, and at a much faster
rate than volatilization (on the order of hours rather than days).  The
overall time for self-purification (following cessation of discharge)
was thus over 2 months for the lakes, approximately 1 month for the
pond, and less than 2 days for the rivers.

     Table 15 presents the simulated chloroform concentrations in differ-
ent environmental compartments (water column, sediment, plankton and
benthos) at steady-state conditions.  Water concentrations were approxi-
mately 0.1-3 mg/1 in the pond and lake systems and considerably lower
(due to dilution and flow rates), 1-10 ug/1, in the river systems.
Sediment concentrations were variable, ranging from -U3.8 ug/kg up to
about 8* mg/kg.  The highest levels were reached in the pond, where the
rates of both volatilization and physical transport were not fast enough
to prevent accumulation slightly above water column concentrations.
Concentrations in biota were generally one to two orders of magnitude
above levels in water.

     Based on the results of the EXAMS run and dependent on the assump-
tions of the model and the rate constants used as input, some conclusions
can be drawn about the potential environmental behavior of chloroform.
In relatively static systems with slow flow rates (e.g. lakes), persis-
tence is a function of volatilization.  In more dynamic river systems,
physical transport processes are much more competitive for chloroform,
removing it before volatilization can reduce the mass significantly.
Transformation processes and bioaccumulation will not significantly
reduce water concentrations.  The sediment layer does not appear to
absorb chloroform at levels much above water concentrations; in fact,
these levels are sometimes lower than water levels.   Since volatilization
is such an important parameter, then the environmental factors that
Increase volatilization rates  conditions of high temperature, high
wind speeds, turbulence of the water, and others  would increase the
rate of loss of chloroform.

     Figure 5 further illustrates the difference between a static sys-
tem's (eutrophic lake) and a dynamic system's (river) response following
the termination of a chloroform discharge (after a system equilibrium has
been reached).  Within the 12 hours following the last discharge, there
is little change in chloroform concentration that is primarily controlled
by volatilization.  In Che river system, however, physical transport
processes reduce the concentration significantly by five orders of

                                     TABLE 14.   Tilt PATE OF CIILOKOFOKM IN VARIOUS CliNKKAUZEU  A()UATIC SYSTEMS
                             Percent  Difitribiitlan
                                                                         Percent Lost by Various Processes
till iii|>lili. l.ilku
III IKIII mplili: l.lliiLint'a Initial illstrlhutIon.
 the results
cleans In'g

                                                AQUATIC SYSTEMS RESULTING  FROM CONTINUOUS DISCHARGE AT  KATE OP 1.0 kg/ltr
                                                              Maximum Concentrations
Sjalciu        	 l.ujdlilB	
                1 .O kg  hr
                                  0. IJ
                                                              5.1 X 10~3     1.2 X  10-2
0. U          0.13          1.7 X ID"3     4.1  X  10~3

9.9 X IO-*    9.9 X 10'*    3.0 X 10~4     1.1  X  10'3
                                                                                                                   7.9  x  10-2
                                                                                                               2.6 X l(l~2
                                  9.9 X
                                                    9.9 X 10-*    6.3 X 10~*     8.6  X  UT4
                                                                                                 1.5 X 10~2    4.7  X  Ifl"1     0.90
                                                               1.5 X 10~2    9.6 X 10~3     0.1)9
                                                                                                                                                   l-oad (kg/day)

Co.iiHl  I- 1.1 ill
                                  9.3 X 10-3    9.3 x 10-3    3.0 x 10~3     1.6  X 10"2
                                                                             4.6 X 10~2     8.0.
                                                                                                                                                 24  '
 Ml .l.i u  ^ I inn 1.1 luil by EXAMS (U.S.  EPA-KKHL. ALliunb,  tta.)  model (sec text for further  information).




f 10-4

a   e
5 10'5
S  10'



I  10'8

u  10'9

                                           Eutro'phic Lake Ecosystem
                                           River Ecosystem

                          (Mrs following last discharge)
                                                     8    9
10   11


 magnitude,  in the  first 5 hours.   Although the rate of removal by physi-
 cal transport is almost entirely  system dependent (e.g.  dependent on
 flow rate,  etc.),  one can assume  that through many systems most of the
 chloroform  would be removed from  the vicinity of discharge before
 significant volatilization occurs.

      Trihalomethanes,  especially  chloroform,  are  commonly  found in the
 natural waters  of  the  United  States.   Levels  in the  open ocean appear
 to  be in  the  range of  1-10  ug/1.   The  other trihalomethanes  are not
 commonly  detected,  and when found are  at  lower  concentrations  than

      Chloroform is  also  found in  ambient  air, both indoors and outdoors.
 Continental background levels range  from  0.04 ug/m3  to  0.13  ug/m3,  while
 marine background  levels are  somewhat  higher.   Levels in urban areas  are
 highly variable, ranging from <0.05  ug/m3 to  90 ug/m3,  with  the maximum
 occurring in  a  highly  industrialized area.

      The  fate of trihalomethanes  is  largely controlled  by  their volatil-
 ity.   While   much  of the chloroform  is originally released to  the  air,
 that  reaching water is rapidly volatilized.   The  half-life of  volatili-
 zation from a stirred  aqueous  solution is  20 minutes.   The importance of
 hydrolysis, adsorption and biodegradation  appear  low compared  with  vola-

     Once reaching  the atmosphere, chloroform and  the other trihalometh-
anes as well,  probably travel  considerable distances before degradation
 occurs.  The  lifetime  in the  troposphere has been  estimated to be 2-3
months, attributable primarily to reaction with hydroxyl radicals.
Direct photochemical degradation of trihalomethane,  as well as rainout,
are not expected to be important pathways.

     Volatilization is also likely from soil surfaces.  Biodegradation
 and adsorption are not likely  to be important fate processes.  Thus,
 trihalomethanes  that  are not volatilized are subject to rapid movement

     A partitioning model suggests that at equilibrium,  all of the
 trihalomethanes would  be primarily (>99%) in the air compartment.
 EXAMS  also showed that volatilization was the dominant  fate process
 in  lakes and  ponds.  In rivers, physical transport over the short
 stretch of river modelled dominated volatilization.

Anderson, D.R. ^et_al.  Quarterly Progress Report Covering Period January 1
through March 31, 1979:  Biocide by-products in aquatic environments.
PNL-2988.  U.S. Nuclear Regulatory Commission; 1979.

Bunn, W.W.; Hass, B.B.; Deane, E.R.; Kloepfer, A.D.  Formation of tri-
halomethanes by chlorination of surface water.  Environ. Lett. 10(3):205-
213; 1975.

Coniglio, W.A.; Miller, K.; MacKeever, D.  The occurrence of volatile
organics in drinking water.  Briefing.  Washington,D.C.:  Criteria and
Standards-Division, U.S. Environmental Protection Agency; March 6, 1980.

Cox, R.A.; Derwent, R.G.; Eggleton, A.E.J.; Lovelock, J.E.  Photochemical
oxidation of halocarbons in the troposphere.  Atmospheric Environ. 10:305-
308; 1976.  (As cited in Versar 1979b).

Dilling, W.L.; Tefertiller, N.B.; Kallos, 6.J.  Evaporation rates of
methylene chloride, chloroform, 1,1,1-trichloroethane, trichloroethylene,
tetrachloroethylene, and other chlorinated compounds in dilute aqueous
solutions.  Environ. Sci. Techno1. 9(9):833-838; 1975.
Hansch, C.; Quinlan, J.E.; Lawrence, G.L.  The linear free energy rela-
tionship between partition coefficients and the aqueous solubility of
organic liquids.  J. Org. Chem. 33:347-350; 1968.

Hansch, C.; Leo, A.  Substituent constants for correlation analysis in
chemistry and biology.  New York, NY:  John Wiley and Sons; 1979.

Hanst, P.L.  Noxious trace gases in the air, Part II:  halogenated
pollutants.  Chemistry 51(2):6-12; 1978.  (As cited in Versar 1979b).

Harsch, D.; Rasmussen, R.A.  Identification of methyl bromide in urban
air.  Unpubl. Contract No. WA 6-99-2922-J.  Washington, DC:  Office of
Research and Development, U.S. Environmental Protection Agency; 1977.

Harsch, D.E.; Rasmussen, R.A.; Pierotti, D.  Identification of a poten-
tial source of chloroform in urban air.  Chemosphere 11:769-775; 1977.

Hendrickson, J.B.; Cram, D.J.; Hammond, G.S.  Organic chemistry.  3rd ed.
New York, SY:  McGraw Hill; 1970.

Hodgman, C.O. ed in chief.  Handbook of chemistry and physics.  43rd ed.
Cleveland, OH:  Chemical Rubber Publishing Company; 1962.

JRB Associates.  Level I materials balance.  Chloroform.  Draft report.
Contract Mo. 68-01-5793, Task Order 13.  Washingcon, DC:  Office of
Pesticides and Toxic Substances, U.S. Environmental Protection Agency;


Karickhoff, S.W.; Brown, D.S.; Scott, T.A.  Sorption of hydrophobia
pollutants on natural sediments.  Water Res. 13:241-248; 1979.

Levins, P.; Adams, J.; Brenner, P.; Coons, S.; Thrun, K.; Harris, G.;
Wechsler, A.  Sources of toxic pollutants found in influents to sewage
treatment plants.  VI.  Integrated interpretation Part I.  Contract
No. 68-01-3857.  Washington, DC:  U.S. Environmental Protection Agency;

Levins, P.; Adams, J.; Brenner, P.; Coons, S.; Thrun, K.; Varone, J.
Sources of toxic pollutants in influents to sewage treatment plants. III.
Coldwater Creek Drainage Basin, St. Louis, Missouri.  Contract No. 68-01-
3857.  Washington, DC:  U.S. Environmental Protection Agency; 1979b.
Lillian, D.; Singh, H.B.; Appelby, A.; Lobban, L.; Arnts, R.; Gumpert, R.;
Hagne, R.; Toomey, J.; Kazazis,'J.; Antell, M.; Hansen, D.; Scott, B.
Atmospheric fates of halogenated compounds.  Environ. Sci. Technol. 9(12):
1042-1048; 1975.

Lyman, W.S.  Literature review  problem definition studies on 13
selected chemicals.  Chloroform.  Contract DAMD17-77-C-7037.  Washington,
DC:  U.S. Army Medical Research and Development Command; 1978.

Mabey, W.; Mill, T.  Critical review of hydrolysis of organic compounds
in water under environmental conditions.  J. Phys. Chem. Ref. Data 7:383-
415; 1978.

Means, J.C.; Hassett, J.J.; Wood, S.B.; Banwart, W.L.  Sorption properties
of energy-related pollutants and sediments.  Jones, P.M.; Leber, P. eds.
Polynuclear aromatic hydrocarbons.  Ann Arbor, MI:  Ann Arbor Science;
1978:  p. 327-340.

Murray, A.J.; Riley, J.P.  Occurrence of some chlorinated aliphatic
hydrocarbons in the environment.  Nature 242:37-38; 1973.  (As cited in
NAS 1978).

National Academy of Sciences (NAS).  Chloroform, carbon tetrachloride,
and other halomethanes.  An environmental assessment.  Washington, DC:
NAS; 1978.  294 p.

Neely, W.B.  An integrated approach to assessing the potential impact of
organic chemicals in the environment.  Proceedings of the workshop on
philosophy and implementation of hazard assessment procedures  for chemical
sub. in the aqu. environment,  Waterville Valley, NH, August 14-18, 1978.

Pearson, C.R.; McConnell, G.  Chlorinated GI and C2 hydrocarbons in the
marine environment.  Proc. Roy. Soc. London B 189:305-322; 1975.

Pellizzari, E.D.; Erickson, M.C.; Zweidinger, R.A.  Formulation of a
preliminary assessment of halogenated organic compounds in man and
environmental media.  Washington, DC:  U.S. Environmental Protection .
Agency; 1979.  Available from:  NTIS, Springfield, VA; PB 80 112170.

Rook, J.J.  Formation of haloforms during chlorination of natural waters.
Water Treatment Exam. 23(2):234-243; 1974.

Rook, J.J.  Chlorination reactions of fulvic acids in natural waters.
Environ. Sci. Technol.  11(5):478-482; 1977.

Spence, J.W.; Hanst, P.L; Gay, B.W. Jr.  Atmospheric oxidation of methyl
chloride, methylene chloride, and chloroform.  Air Poll. Control Assoc.
26(10):994-996; 1976.  (As cited in Versar 1979b).

Stanford Research Institute (SRI).  Estimates of physical-chemical
properties of organic priority pollutants.  Preliminary draft.
Washington, DC:  Monitoring and Data Support Division, U.S. Environ-
mental Protection Agency; 1980.

Stecher, P.G. jst al. eds.  The Merck index.  8th ed.  Rahway, NJ:  Merck
and Co., Inc.; 1968.

Su, C.; Goldberg, E.D.  Environmental concentrations and fluxes of some
halocarbons.  Windom, H.L,.; Duce, R.A. eds.  Marine pollutant transfer.
Lexington, MA:  D.C. Heath and Co.  (As cited in MAS 1978).

U.S. Environmental Protection Agency (U.S. EPA).  Report on the problem
of halogenated air pollutants and stratospheric ozone.  Report EPA 600/
9-75-008.  Research Triangle Park, NC:  Office of Research and Develop-
ment, U.S. EPA; 1975.

U.S. Environmental Protection Agency (U.S. EPA).  In-depth studies on
health and environmental impacts of selected water pollutants.  Contract
No. 68-01-4646.  Washington, DC:  Office of Water Planning and Standards,
U.S. EPA; 1978.

U.S. Environmental Protection Agency (U.S. EPA).  STORET.  Washington,
DC:  Monitoring and Data Support Division, U.S. EPA; 1979.

Versar, Inc.  Non-aquatic fate.  Trihalomethane.  Draft report for
Monitoring and Data Support Division, U.S. Environmental Protection
Agency, Washington, D.C.; 1979a.

Versar, Inc.  Trihalomethane:  statement of probable fate.  Callahan,
M.A.; Slimak, M.W.; Gabel, N.W.; May, I.P.; Fowler, C.F. ; Freed, J.R.;
Jennings, P.; Durfee, R.L.; Whitmore, F.C.; Maestri, B.; Mabey, W.R.;
Holt, 3.R.; Gould, C.  Water-related environmental fate of 129 priority
pollutants. I.  Report No. EPA-440/4-79-029a.  Washington, DC:  Moni-
toring and Data Support Division, U.S. Environmental Protection Agency;

Wilson, J.T.; Enfield, C.G.; Dunlap, W.J.; Cosby, R.L.; Foster, D.A.;
Baskin, L.B.  Transport and fate of selected organic pollutants in a
sandy soil.  Ada, OK:  Robert S. Kerr Environmental Research Laboratory,
U.S. Environmental Protection Agency; 1980.

Yung, Y.L.; McElroy, M.B.; Wofsy, S.C.  Atmospheric halocarbons:  a
discussion with emphasis on chloroform.  Geophysical Research Letters
2(9):397-399; 1975.  (As cited in Versar 1979b).

                                CHAPTER V

                     EFFECTS AND EXPOSURE  HUMANS

1.   Chloroform .

a.   Introduction

     Prior  to  the mid-1940s,  chloroform was  widely used as an anesthetic
and, until  recently, was  utilized  in many pharmaceutical preparations
(e.g.,  cough medicine, mouthwashes,  dentifrices,  linaments).   The possi-
bility  of chloroform carcinogenicity in rodents has raised concern about
long-term,  low level exposure in man,  particularly in view of recent
findings of chloroform in concentrations up  to  300 ug/1 in the drinking
water of municipalities throughout the United States.   This section of*
the report  examines the extensive  data base  on  the toxic effects of
chloroform  in  man and laboratory animals.

b.   Metabolism and Bioaccumulation

     Chloroform (CHC13) is rapidly absorbed  through the lungs, through
the gastrointestinal tract if ingested, and,  to a lesser extent, through
intact  skin (Brown et .al.  1974, Van  Dyke .et  al. 1964).   With  some
species variation, chloroform is partially excreted unchanged and par-
tially  metabolized to carbon  dioxide (C02) and  unidentified urinary
metabolites in mice, rats, monkeys,  and humans  (NAS 1977). Pohl et al.
(1977)  recently reported  that chloroform was metabolically activated to
phosgene  (COC12) by liver microsomes of phenobarbital-pretreated rats.

     Mice exhibit the highest conversion rate to  CC^,  metabolizing
approximately  80% of a 60-tng/kg oral dose of ^Q^ (Brown e_t  al. 1974).
No differences in the conversion rate due to strain (Brown e_t al. 1974)
or sex  (Taylor &t_ al. 1974) were noted.  Tissue distribution  of the
radio-label, however, showed  striking sex differences:   males had higher
levels  of radioactivity (^ threefold)  bound  to  the renal cortex and
medulla, while females accumulated the label in the liver to  a greater
extent  than males (Taylor _et  al. 1974).

     In rats administered 60  mg/kg 1^CHC13 by gavage,  66% of  the dose
was metabolized to l^cc^,  with an  additional 20%  of the dose  eliminated
unchanged in expired air  (Brown et al.  1974).  Administration of a much
larger  dose (1,484 mg/kg)  incraduodenally resulted in a much  smaller
conversion  to  C02 (4%); the major  portion of the  dose (70%) was elimi-
nated unchanged in expired air (Van  Dyke ec  al. 1964).

     Only 17% of an orally administered dose of 14CHCl3  (60 mg/kg) was
metabolized to ^ C02 in the squirrel monkey; 78% of the  dose was excreted
unchanged in expired air (Brown _e al. 1974).

     In eight healthy human volunteers, between 17.8% and 66.6% of
unchanged chloroform was recovered in expired air within an 8-hour-
period following oral administration of a gelatin capsule containing
500 mg of chloroform (^ 7 mg/kg) in olive oil.  The wide range of values
is partially attributable to body weight variations (58-80 kg); obese
subjects excreted a smaller proportion of the dose through the lungs as
unchanged chloroform.  The maximum pulmonary excretion occurred between
40 and 120 minutes post-dosing.  Less than 1% of the dose was detected
in urine (Fry ej: al. 1972).  In a separate experiment, two subjects, one
male and one female, swallowed a capsule containing 500  mg of l^c-labelle
chloroform.  A total of 50.6% and 48.5% of the dose was  accounted for as
exhaled 13C02 ^or the male an<* female subject, respectively (Pry et al.

     Poobalsingham and Payne (1978) studied the rate of  alveolar uptake
of chloroform in 16 patients undergoing elective surgery.  Eight patients
spontaneously breathed known concentrations (2-2.5%) of  chloroform; the
second group were artificially ventilated with a fixed inspired concen-
tration of 1% chloroform.  Initial uptake of chloroform  was rapid in
both groups, approaching a plateau after 40-45 minutes.  Expressed as
a percentage .of the equilibrium value, the arterial concentration of
chloroform increased more steeply with controlled ventilation than with
spontaneous breathing.   In spontaneously breathing patients, the arterial
concentration reached approximately 25% of the inspired  concentration
after about 1 hour of exposure compared with an equilibration of approxi-
mately 41% in the same time period under controlled ventilation.
Increasing concentrations of chloroform in the blood of  spontaneously
breathing patients causes respiratory depression sufficient to reduce
alveolar ventilation.  Elimination following chloroform  withdrawal was
rapid and exponential in nature.

     An average chloroform concentration of 51 ug/kg (range:  5-68
ug/kg wet tissue) was detected in postmortem samples of  body fat taken
from eight subjects between 48 and 82 years of age.  Concentrations of
1.0 ug/kg to 10 ug/kg were present in kidney, liver, and brain tissues
(McConnell .et al. 1975).
     In summation, chloroform is rapidly absorbed from the lungs and
gastrointestinal tract, and to a lesser extent, through intact skin.
With some species differences,'chloroform is excreted partially unchanged
and partially metabolized to CO2 and unidentified urinary metabolites.
Body organs contain 1-10 ug/kg, with 5-68 ug/kg chloroform detected in
bodv fat.

c.   Animal Studies

i.   Carcinogenesis

     Chloroform was implicated as a possible carcinogen more  than  30
years ago by Eschenbrenner and Miller (1945).  They administered 30 oral
doses of 0, 150, 300, 600, 1200 or 2400 ing/kg chloroform  to groups of
five male and female strain A mice at 4-day intervals.  Males in the
three highest treatment groups and females receiving 2400 mg/kg chloro-
form died within 48 hours.  When the mice were 8 months of age, non-
metastasizing hepatomas and cirrhosis were found in all surviving
females given 600 mg/kg or  1200 mg/kg.  No hepatomas were seen in
either male or female mice at the lower dosage levels.

     An abstract of another study of limited'duration reports that
Rudall (1967) gave 24 mice (strain unspecified) 0.1 ml or a 40% solution
of chloroform orally twice weekly for 6 months.  Three hepatomas were
found in the five survivors.  No additional information was given.

     USF-grade chloroform in corn oil was administered by gavage to
groups of 50 male and female mice (B6C3F1 strain) at a predetermined
maximally tolerated dose and one-half this amount 5 days per week for
78 weeks.  The dose of chloroform was changed during the course of the
experiment; time-weighted-average doses were 138 mg/kg and 277 mg/kg for
males and 238 mg/kg and 477 mg/kg for females.  The vehicle control
group consisted of 20 mice of each sex.

     Frank hepatocellular carcinoma was found in 36% to 98% of treated
mice (p<0.001) compared with 0% to 6% incidence in controls (see
Table 16). Nodular hyperplasia of the liver was observed in several
low-dose males that had not developed hepatocellular carcinoma
(NCI 1976, Powers and Voelker 1976, Renne et al. 1976).

     A concurrent study was conducted with groups of 50, 52-day-old
Osborne-Mendel rats given time-weighted oral doses of 90 mg/kg and
180 mg/kg chloroform for males and 100 mg/kg and 200 mg/kg for females
for 78 weeks.  A significant (p=0.0016) increase in the incidence of
kidney epithelial tumors was seen in males:  8% at 90 mg/kg;  24% at
180 mg/kg compared with none in controls.  An increase in thyroid tumors
was observed in female rats, but was not considered as resulting from
the chloroform exposure (NCI 1976, Renne e a. 1976).

     Another recent series of studies (Roe jat jil. 1979, Palmer et al.
1979, Heywood jt_al. 1979) examined the carcinogenic effects  of feeding
chloroform to Sprague-Dawley rats, beagles, and four strains of mice at
dosage levels ranging from 15 mg/kg to 60 mg/kg.  Except  for  an excess of
renal tumors in male ICI/CFLP mice at the 60-mg/kg treatment  level, ao
carcinogenic effects './era reported.

     In  the first study,  four strains of mice  (male and female ICI/CFL?,
aale C5736, CF/1 and C3A) vera administered chlorofonn in a coothpasca

                      IN B6C3F1 MICE EXPOSED TO CHLOROFORM
                               No.  Mice with Carcinoma
                               	(% incidence)	
High CHC1
Low CHC13
- Dose

 Source:   NCI (1976),  Powers and Voelker (1976),  Renne et a_l. (1976)
30 mg CHC13 /kg/day
15 mg CHOLj /kg/day
vehicle control
untreated control
No. Dogs with Nodules
(% incidence)
0/7 (0)
1/7 (14)
0/15 (0)
1/7 (14)
4/8 (50)
1/8 (13)
3/12 (25)
1/5 (20)
No. Dogs with Fatty Cysts
(% incidence)
7/7 (100)
6/7 (86)
8/15 (53)
2/7 (29)
7/8 (88)
5/8 (63
3/12 (25)
1/5 (20)
Source:  Heywood t al. (1979)

vehicle by gavage 6 days per week for 80 weeks, followed by 16-24 weeks
of observation.  No overall increase in neoplastic changes was observed,
except for an increase in tumors of the renal cortex in male ICI/CFLF
mice (9/52) at the 60-mg/kg level compared with vehicle (6/260) and
untreated (1/52) controls.  No overall increase in tumors was observed
in males of this strain given 17 mg/kg/day of chloroform nor in female
ICI/CFLP or male C57BL, CBA or CF/1 strain mice given up to 60 mg/kg/day
of chloroform for 1.5 years (Roe e al. 1979).

     In a separate experiment, male ICI/CFLP mice were given 60 mg/kg
chloroform in arachis oil, as well as in the toothpaste vehicle.  There
were 12/52 renal tumors in the CHCl3/arachis-oil-treated group compared
with 1/52 in arachis oil controls; mice given 60 mg/kg chloroform in the
'toothpaste vehicle had 5/52 renal tumors (Roe e al. 1979).

     In the second study, caesarian-derived SPF Sprague-Dawley rats
(50 males, 50 females) were intubated 6 days per week with 60 mg/kg
chloroform in a toothpaste vehicle for 80 weeks, followed by 15 weeks
of observation.  The only noteworthy changes were a depression of plasma
(but not erythrocyte) cholinesterase in females, which reversed itself
upon cessation of treatment, and a decrease in absolute and relative
liver weights of females killed at termination of the study.  Females,
but not males, possess plasma butyrlcholinesterase activity, which was
inhibited by chloroform treatment.  No clear histologic evidence of
toxic effects on liver or kidney were seen in either sex.   A wide spec-
trum of-different types of neoplasms was seen but did not appear to be
related to treatment (Palmer t al. 1979).

     In the third investigation of this series, Heywood and co-workers
(1979) gave beagles 0, 15 or 30 mg/kg chloroform in a toothpaste vehicle
(in a gelatin capsule) orally 6 days per week for 7.2 years; the dogs
were maintained without treatment for an additional 20-24 weeks.  There
were 8 dogs of each sex in untreated controls and each treatment level;
16 dogs of each sex were in the vehicle control group.  The only signifi-
cant toxic response was a moderate rise in serum glutamic pyruvic trans-
aminase (SGPT) levels, which peaked during the sixth year of the study.
Most of the dogs reverted to normal SGPT values during the post-treatment
period.  No palpable tumors were noted, but a few subcutaneous nodules
thought to be of mammary gland origin were detected.  Possible chloro-
form-related aggregations of vacuolated histiocytes ("fatty cysts") were
seen in the livers of several dogs at postmortem; the distribution of
nodules of altered hepatocytes and fatty cysts in the liver is presented
in Table 17.  The distribution of fatty cysts appears to be dose-related.
A small number of macroscopically and microscopically visible neoplasms
was  seen but were predominantly age-related.  No neoplasia of liver or
kidney was found.

     Capel and Williams  (1978) investigated  the effect of  chloroform on
 the  growth of  some murine  tumors  (Lewis lung  carcinoma, B16 melanoma,
and  Ehrlich ascites).  Groups of male TO strain mice were  given 0, 0.15
or 15  mg/kg of chloroform  in  their drinking water for varying  periods  of
time either prior to and post-inoculation or  only post-inoculation with
1 x  10^ cunor  cells.   Exposure co 0.15 tng/kg  chloroform did noc enhance
 the  growth or  aetastasis of Lewis  Lung carcinoma or significantly

increase the number of Ehrlich ascices cumor cells.  Ac  Che  15-mg/kg
chloroform treatment level, however, a significanc increase  in  foci of
Lewis lung and Ehrlich Cumor cells was noted.  Similarly, both  chloroform
exposure levels resulted in invasion of a greater percentage of organs
by B16 melanoma Chan Chac seen in controls.

     Roe .ec al. (1968) gave an unspecified number of newborn (C57xDBA
mice either a single subcutaneous 200-ug dose of chloroform  before they
were 24 hours old or eight daily 200-ug injections during the first week
of life.  No evidence of carcinogenicity was observed after  77  to 80 weeks.

     Thus, ingestion of 90-138 mg/kg of chloroform has been  shown to
induce both hepatic and renal tumors in experimental animals, although
other experiments have yielded contradictory results.  The negative
results may reflect different vehicles employed to administer chloroform,
lack of sensitivity-of various test species, populations too small to
measure statistically significant and/or the lower dosage levels admini-
stered in the Roe eit al. (1979) and Palmer _e_t al. (1979) studies compared
with the NCI bioassay (1976).

ii.  Mutagenesis

     No indications of mutagenic activity have been reported in the
scientific literature for chloroform.  No significant mutations were
found at .the 8-azaguanine locus in Chinese hamster lung fibroblast cells
following 24-hour exposure to 3% chloroform in culture (Sturrock 1977).
     Negative results have also been reported for chloroform vapors in
microbiol assays conducted with Salmonella typhimurium strains  TA100,
TA1535 (base-pair mutants) and TA1538 (frameshift mutant), both in the
presence and absence of liver microsomal activation (Simmon  and Tardiff
1978, Uehleke et al. 1977).

     Chloroform also did not induce DNA damage (single-strand breaks)
in liver DNA of rats given 200-400 mg/kg of chloroform orally.  DNA
damage was measured by alkaline elution assay (Petzold and Swenberg

iii. Teratogenesis

     Schwetz_et_al. (1974) exposed pregnant Sprague-Dawley rats to 30,
100 or 300 mg/kg chloroform in the diet, 7 hours a day on days  6 through
15 of gestation.  At 30 mg/kg (diet), an increase in wavy ribs  and
delayed skull ossification were observed.  Ac 100 mg/kg, a significant
incidence of fetal abnormalities such as acuadia, imperforate anus,
subcutaneous edema, missing ribs, and delayed ossification of both skull
and sternebrae were noted.  At the 300 mg/kg exposure level, maternal
weight gain and food consumption were markedly reduced.  The conception
race in Chis group was only 15% (3/20) compared with 88% (63/77) in the
control group.  The aon-pregnant bred females in Chis treatment group
exhibited vascular evidence of implantation in the aesometrium, buc no
intrauterine evidence of implantation.  The authors suggest  that chloro-
form aay interfere vich Che process of iaplancacion.  Litters from Che
chrea dams exposed co 300 tag/kg chloroform had a reduced number of live

healing  (Torkelson ec al.  1976).  A single occluded  24-hour  application of
fetuses and decreased fetal body measurements.  The  number of fetal
resorptions was also increased.  These effects do not appear to be
related to the anorexia seen in the dams at this exposure level, in  that
the same degree of starvation  (3.7 g food per rat per day on days 6
through 15 of gestation) without exposure to chloroform was neither
embryo- nor feto-toxic in a special control group maintained on starva-
tion rations.

     In another study, rats administered oral dosages up to 126 mg/kg
chloroform on days 6-15 of gestation and rabbits given up to 50 mg/kg
orally on days 6-18 of gestation showed no evidence of teratogenicity.
Offspring of  both species  had  reduced birth weight at  these  maternally
toxic levels, but not at 50 mg/kg and 35 mg/kg treatment levels for rats and
rabbits, respectively, when compared with the short  daily oral exposure.

iv.  Other lexicological Effects

     Available data indicate species-, strain-, and  sex-dependent variations
in  the acute  toxicity of chloroform in laboratory animals (see Table 18).
Oral LD50 (50% lethal dose) values range from 119 mg/kg to 1400 mg/kg in
the mouse to  2000 mg/kg in the rat.

     In mice, chloroform toxicity varies according to  strain and sex.
The oral 1.050 for chloroform was four times higher in  C57BL/6J males
than in DBA/2J males with  first generation offspring of these two strains
(B6D2F1/J),  exhibiting a value midway between the parental strains.   In
addition, the histopathological response to acute exposure to chloroform
varied with  sex.  At doses > 252 mg/kg, males of all three genotypes
exhibited necrosis of the  proximal convoluted renal  tubules  and hepatic
centrilobular necrosis; only renal tubular necrosis  was seen at doses
less than 252 mg/kg of chloroform.  Females had a similar threshold  to
hepatic damage without developing renal lesions.  Immature or castrated
adult males were also resistant to chloroform-induced  renal  toxicity,  but
both castrated males and females treated with testosterone became sensi-
tive to chloroform-induced renal toxicity (Hill e al. 1975, Vessell
et  al. 1976,  Hill 1978).

     Acute oral exposure to 250 mg/kg chloroform produced fatty infiltra-
tion and necrosis of the liver and kidney damage in  rats (Torkelson  et al.
1976).  Jones and co-workers (1958) noted similar findings in mice given
350 mg/kg chloroform orally.   The hepatotoxic effects  of chloroform  have
been attributed to the interaction of a reactive metabolite  of chloroform
with tissue  proteins and appears to be related to two  factors:  (1)  the
rate of biotransformation  and  (2) the availability of  endogenous hepatic
antioxidant,  reduced glutathione (GSH).  In rodents, hepatic necrosis can
be  enhanced by pretreatment with hepatic microsomal  enxyme inducers  and
conversely reduced by inhibiting biotransformation (Cresteil_et_al.  1979,
Ilett _et _al.  1973).  Supplying antioxidants such as  GSH also reduces the
severity of  liver necrosis associated with chloroform  treatment; the
actual mechanism by which  GSH  reduces tissue damage  is not known (Stevens
and Anders 1978).

     One or  cwo 2&-hour applications of chloroform to  shaved rabbit
abdomen produced hypersraia, necrosis, scab formation,  and delayed

LPsn (me/kg)
Mmisi! TCI Swiss F
fCl Swiss M
H6D2F1/J M
Swiss Webster M
Itat M
300 to 470 g M
80 to 160 g M
16 to 50 g M-i-F
Dog M+F
1400 (1120-1680)
1120 ( 789-1590)
489 ( 386- 594)
297 ( 237- 356)
119 ( 103- 148)
1,781 (1,484-1,929)
134 mg/m
1,182 (1,038-1,336)
1,336 (1,182-1,632)
445 ( 297- 742)
490 mg/m
289 mg/m3
                                                  Bowman  et  al.  (1978)

                                                  Bowman  et  al.  (1978)

                                                  Hill  e al.  (1975)

                                                  Hill  t a\_.  (1975)

                                                  Hill  t al.  (1975)

                                                  Klassen and  Plaa (1966)

                                                  RTECS (1977)

                                                  Torkelson  et^ al. (1976)

                                                  Torkelson  et^ _al. (1976)

                                                  Torkelson  e al. (1976)

                                                  Kimura  ^it  al.  (1971)

                                                  Klmura  e  al.  (1971)

                                                  RTliCS (1977)

                                                  Brownlee eit  _al. (1953)

                                                  RTECS (1977)

                                                  RTECS (1977)

1,000 mg/kg to 3,980 aig/kg of chloroform to rabbit abdomen produced
extensive necrosis and body weight loss.  After 2 weeks, necropsy revealed
degeneration of the kidney tubules; the livers were not grossly affected
(Torkelson t al. 1976).

      Instillation  into  the eyes  of  rabbits  resulted  in slight conjuncti-
val  irritation and corneal injury,  which was  not  reduced  by  washing the
eye  30  seconds later  (Torkelson   at%  1975).

      Torkelson and co-workers  (1976)  exposed  rats to 123,  245 or  417
mg/m3 chloroform by inhalation 7 hours  per  day  for 138 and 144 days
during  a  195- to 203-day  period.  All males displayed an  increased
relative  kidney  weight, cloudy swelling of  the  renal tubular epithelium,
and  centrilobular  granular degeneration with  focal areas  of  necrosis
throughout  the liver.   Decreased body weight  was  observed in males  at the
50 mg/kg  and 85  mg/kg levels, with pneumonia and an increase  in the  rela-
tive liver  weight  present in the 85 mg/kg male  rats.   Only an increase
in relative kidney weight was  noted in  females  at the lowest treatment
level;  at the two  higher  exposure levels, liver and  kidney pathology was
similar to  that  seen in male rats.  Male rats allowed to  recover  for 6
weeks after their  last  exposure  to  25 mg/kg chloroform, exhibited no
pathological findings.  Even at  the 85  mg/kg  treatment^level, terminal
blood counts and urine  analyses  were  unaffected and  serum levels  of urea
nitrogen, glutamic-pyruvic transaminase, and  alkaline phosphatase were
within  normal limits, despite  liver damage, observed histologically.

      Dogs exposed  to 123  mg/nr* chloroform,  and  rabbits and guinea pigs
exposed by  inhalation to  123 mg/m^  or  417  mg/m^  under the same condi-
tions showed inconsistent liver, kidney, and  lung changes  (Torkelson
.et al.  1976).

      In another  study,  Miklashevskii  (1966) administered  0.4 mg/kg  or
35 mg/kg  chloroform orally to  male  guinea pigs  for 5 months  (daily
administration implied, but  not  stated).  No  effects were noted at  the
0.4-mg/kg dose.  At 35  mg/kg,  only  two  of the six guinea  pigs survived
beyond  3  months.   The blood  albumin-globulin  ratio and blood catalase
activity  decreased by the end  of the  first  and  second months, respec-
tively.   Animals that died exhibited  fatty  infiltration,  necrosis,  and
cirrhosis of liver parenchyma, lipoid degeneration,  proliferation of
interstitial cells in the myocardium, and acute edema of  the submucosal
and  muscular layers of  the stomach.

      No mortality  was seen in  dogs  given 30 mg/kg to 120  mg/kg chloro-
form orally for  12-18 weeks.   Vomiting  occurred at times  in  dogs  given
30,  60  or 120 mg/kg of  chloroform and one of  two  dogs given  120 nig/kg/day
became  jaundiced after  4-5 weeks.   Marked body  weight losses and
increased levels of serum glutamic  pyruvic  transaminase were observed
in dogs given 60,  90 or 120  mg/kg/day.  At  necropsy,  liver discoloration
and  increased liver weight were  noted in dogs receiving 45 aig/kg/day and
above,  but  not at  the 30-sig/kg treatment level.   At  50, 90 and 120  mg/kg,
hepatocyta  enlargement  and vacuolation  were seen,  together with deposition
of fat  vithin the  heoatocytes  (Heywood  ac al. 1979).

     Thus, the acute toxicity of chloroform  is species-, strain-,  and      '
sex-dependent.  Oral LD$Q values range from  119 tag/kg  to 2000 nig/kg,      .'.
with indications of renal and hepatic necrosis.  Dermal  contact with      /' ,
chloroform for 24 hours resulted in tissue necrosis in rabbits and        '\i
instillation into eyes resulted in ocular injury.  Subchronic inhalation  ''
exposure to chloroform produced liver and kidney pathology  in rats, but
findings were inconsistent in dogs, rabbits,  and guinea pigs.

d.   Human Studies

     Although there are many documented fatalities from  chloroform-
induced anesthesia, as well as cases of accidental or  intentional
ingestion, limited information is available  on controlled human exposure
to chloroform.  Ingestion of 120 ml of chloroform has  been  survived
(Schroeder 1965), while serious illness occurred in another individual
after ingestion of only 5 ml (Winslow and Gerstner 1978).

     Signs of chloroform poisoning include a characteristic sweetish
odor on the breath, dilated pupils, cold and clammy skin, initial
excitation alternating with apathy, loss of  sensation, abolition of
motor functions, prostration, unconsciousness, and eventual death.
Central lobular necrosis of the liver and renal damage are  the most
outstanding pathological findings (Winslow and Gerstner  1978).

     litmus and Moser (1975) reported patchy  pulmonary  infiltrates and
acute respiratory distress in a 21-year old  man who intentionally
injected himself intravenously with a 5-tnl bolus of reagent-grade
chloroform.  Evidence of acute hemolysis was also noted.  No hepatic
or renal toxicity was seen, probably because of the route of administra-
tion and the probable rapid pulmonary clearance of the injected chloro-
form.  The subject returned to normal pulmonary functional  status
within 3 weeks.

     Desalva ejt al. (1975) found no indications of hepatotoxicity in
chronic users of a dentifrice and mouthwash  containing 3.4% and 0.43%
chloroform, respectively, over a one- to five-year period.  Estimated
daily ingestion was 0.3-0.96 mg/kg per day.  Some reversible hepatotox-
icity was seen, however, in a ten-year clinical study with  patients who
ingested a cough suppressant containing 1.6 g to 2.6 g of chloroform daily
(^ 23-37 mg/kg per day) (Wallace 1959).

     Dermal exposure to chloroform may cause irritation, erythema,
hyper emia, and destruction of the epithelium (Ma It en _et_  al. 1968).  Eye
contact produces burning, redness of conjunctival tissue, and possible
damage to the corneal epithelium.  Recovery  generally  occurs in one to
three days (Grant 1974).

     The carcinogenic effects of chloroform  in rodents,  and the detec-
tion of chloroform-in concentrations up to 300 ug/1 in the  drinking

water of municipalities throughout the United States have raised concern
on the human health impact of long-term, low-level exposure to chloro-
form in public water supplies.  Several epidemiologic studies have
suggested an association between cancer mortality rates and levels of
chloroform in drinking water  (Page t, al. 1976, De Rouen and Diem 1977,
Cantor jet al. 1978, Alavanga j a!L. 1978).  Many of these studies,
however, are deficient in a number of areas such as:  lack of individual
exposure data, variations in concentrations in water data, and use of
current rather than historical exposure data.  Furthermore, cancer rates
can be confounded by population mobility patterns; regional trends in
diet, tobacco, and alcohol consumption; as well as possible contributions
of other sources such as occupation, local extent of industrialization,
air pollution, etc. (Kraybill 1978).

     In a more direct attempt to assess relative cancer risk associated 
with exposure to chlorinated drinking water, Alavanga and co-workers
(1978) conducted the only case-paired study cited in the scientific
literature.  These investigators compared individual death certificate
data for all females in seven New York State counties for the years
1968 to 1970 that died of gastrointestinal and urinary tract cancer
with a corresponding set of matched controls.  In three of seven coun-
ties studied, the combined gastrointestinal and urinary tract cancers
were significantly higher in areas supplied by chlorinated water, but
there was no significant difference in the other four counties.  Further-
more, the results were complicated by the overriding finding that both
cancer and chlorination correlated with urban living and this common
relationship would result in the apparent correlation between cancer
and chlorination.

     Thus, although positive correlations between chloroform in drinking
water and increased cartcfcr incidence may be apparent, causal relation-
ships have not been proven on the basis of results from these indirect

2.   Bromoform

a.   Metabolism
     Absorption of bromoform may occur by inhalation, from the gastro-
intestinal tract, and, to a certain extent, through the skin.  Lucas
(1928) reported that following administration of bromoform to rabbits
either rectally or by inhalation, inorganic bromides could be detected
in tissues and urine.  Anders and associates (1978) found substantial,
dose-dependent elevations in blood carbon monoxide levels in male rats
given 1 tnmol/kg bromoform intraperitoneally.  Enzyme induction by
pretreatment with phenobarbital (but not 3-methylcholanthrene) markedly
increased the blood carbon monoxide levels seen after bromoforn admini-
stration compared to saline-treated controls.  The administration of
SKF 525-A, a known inhibitor of drug metabolism, significantly inhibited
in vivo aetaboLisin of bromofora to carbon monoxide.

b.   Animal Studies

1.   Carcinogenesis

     The is s _e_t al. (1977) reported Chat bromoform produced a positive
response in a pulmonary adenoma bioassay.  The administration  to  strain  A
male mice of 48 mg/kg bromoform intraperitoneally three  times  per week
for 23 weeks resulted in 1.13 0.36 lung tumors per mouse p   0.041)
compared with 0.27 t 0.15 tumors for vehicle controls.   A higher  bromo-
form dosage (100 mg/kg), however, failed to produce a significant
pulmonary response (0.67 t 0.21 lung tumors; p ** 0.136) under the  same
treatment conditions.

     Bromoform is currently being tested for carcinogenicity by the
standard bioassay protocol of the National Cancer Institute.

ii.  Mutagenesis

     Bromoform vapor tests positive in the Ames system (Simmon and
Tardiff 1978).   Exposure to the vapor phase of 30 ul of  bromoform was
mutagenic in Salmonella typhimurium TA 100 with or without metabolic
activation.  A weak response was also reported for strain TA 1535. exposed
to the vapor phase of 100-ul of bromoform.

iii. Teratogenesis

     No data are available regarding teratogenic effects due to bromo-

iv.  Other Toxicologic Effects

     Little is known of the toxicity of bromoform.  Bowman et  al. (1978)
recently reported oral LD50 values of 1400 mg/kg (1205-1595) and  1550
mg/kg (1165-2065) in male and female ICR Swiss mice.  Males appeared to
be more sensitive than females to the acute lethal effects of  bromoform.
Ataxia, sedation, and anesthesia occurred within 60 minutes of oral
administration of 1000 mg/kg of bromoform, but not at lower dosages.
Sedation persisted for approximately 4 hours.  Subcutaneous LDso  values
of 410 mg/kg and 1820 mg/kg bromoform have also been reported  for the
rabbit and mouse, respectively (RTECS 1977, Kutob and Plaa 1962).

     In dogs, deep narcosis occurred after 8-minute exposure to 142,100
mg/m-3 bromoform; mortality occurred after 60 minutes.  Exposure to the
same concentration for 30 minutes produced deep narcosis, but  the dogs
survived (Sax 1979).

     Inhalation by rats of 0.25 mg bromoform per liter of air for 4 hours
per day for 2 months produced disorders in prothrorabin synthesis  and
glycogenesis in the liver and reduced renal filtration capacity;  the
threshold concentration was 0.05 mg/liter (Dykan 1962).

c.   Human Studies

     Little information is available on the adverse health effects of
bromoform in man.  Sax (1979) reports that inhalation of small amounts
of bromoform causes irritation provoking lacrimation, salivation and
facial hyperemia.  Poisonings with bromoform have been reported (Fatty
1963).  Symptoms include listlessness, headache, vertigo, unconscious-
ness, loss of reflexes, and occasionally convulsions.  The primary cause
of death is respiratory failure.

3.   Dibromochloromethane

a.   Metabolism

     Anders et al. (1978) foun'd a slight elevation of blood carbon mono-
xide levels in male rats given 1 mmol/kg of dibromochloromethane by
intraperitoneal injection.  At 2 hours, blood levels of carbon monoxide
were approximately 400 mmol CO/ml compared with a value of 50 mmol CO/ml
for controls.

b.   Animal Studies

i.   Carcinogenesis

     No data are currently available concerning the carcinogenicity of
dibromochloromethane.  The compound is currently being tested (June
1979) for carcinogenicity by the National Cancer Institute according
to its standard bioassay protocol.

ii.  Hutagenesis

     Simmon and Tardiff (1978) reported a positive mutagenic effect in
Salmonella typhimurium TA 100 (with or without metabolic activation)  '
exposed to vapor from 10 ul of dibromochloromethane in a dessicator.

iii. Teratogenesis

     No data are available concerning the teratogenesis of dibromochloro-

iv.  Other Toxicoloqic Effects

     Toxicological data on dibromochloromethane are sparse.  Bowman et al.
(1978) recently reported oral IS^Q values of 800 mg/kg (667-960) and
1200 mg/kg (945-1524) dibromochloromethane tor male and female ICR Swiss
mice, respectively.  Males appeared to be more sensitive than females
to the acute lethal effects of dibromochloromethane.  Sedation and
anesthesia occurred within 30 ainutes following oral administration of
500 mg/kg of this compound and persisted for approximately 4 hours.  Ac
necropsy, livers appeared co have fancy iafiltration, che kidneys vere
pale, and hemorrhaging was noted in the brain, lungs, and adrenals.

c.   Human Studies

     No human data are available.

4.   Dichlorobromomethane

a.   Metabolism

     Male rats injected intraperitoneally with  1 mmol/kg  dichlorobromo-
methane revealed only a small elevation  in blood levels of  carbon
monoxide.  At 2 hours following injection, blood carbon monoxide levels
were approximately 50 mmol and 120 mmol  CO/ml for  control and  dichloro-
bromomethane-treated rats, respectively  (Anders jat al. 1978).

b.   Animal Studies

i.   Carcinogenesis

     Theiss et al. (1977) reported that  dichlorobromomethane produced  an
elevated response in a pulmonary adenoma bioassay  that approached statis-
tical significance.  Strain A male mice  were injected intraperitoneally
three times per week for 24 weeks with 0, 20, 40 or 100 mg/kg  dichloro-
bromome thane.  The incidence of lung tumors per mouse (*  SE) were 0.27 *
0.15, 0.20  0.11 (p =0.724), 0.25  0.11 (p =0.930) and  0.85    0.27
(p  0.067), respectively.
     Dichlorobromomethane is currently being tested (June 1979)  by the
National Cancer Institute according to Its standard bioassay protocol.

ii.  Mutagenesis

     Dichlorobromomethane (50 ul) was mutagenic in Salmonella  typhimurium
strain TA 100, both in the presence and  absence of metabolic activation
(Simmon and Tardiff 1978).

iii. Teratogenesis

     No data are available concerning teratogenic  effects of dichloro-
bromome thane.

iv.  Other Toxicologic Effects

     As is the case for dibromochloromethane, the  toxicologic  data avail-
able for dichlorobromomethane are quite  limited.   Male ICR  Swiss  mice
appear to be more sensitive than females to the acute lethal effects of
dichlorobromomethane.  Oral LD$Q values  of 450 mg/kg (326-621) and 900
mg/kg (811-999) of dichlorobromomethane were recorded for males  and
females of this strain, respectively.  Sedation and anesthesia occurred
within 30 minutes after oral administration of 500 mg/kg  of this  compound
and persisted for approximately & hours.  At necropsy, livers  appeared
to have fatty infiltration,  kidneys were pale, and hetnorrhaging was
observed in the adranals, Lungs, and brain (Bowman  a.1. L973).

c.   Human Studies

     No data are available.

5.   Overview

     Trihalomethanes are present in the chlorinated drinking water
supply of municipalities throughout the United States,  Several epidemi-
ological studies have suggested a correlation between ingestion of
chlorinated water and a higher incidence of cancer.  The four trihalo-
methanes of concern are chloroform, bromoform, dibromochloromethane, and

     Extensive toxieologic data on chloroform indicate that it is rapidly
absorbed through the lungs, from the gastrointestinal tract and, to a
lesser extent, through intact skin.  Species variations exist with regard
to the metabolic handling of chloroform, but some conversion to carbon
dioxide does occur.

     Ingestion of 90-132 mg/kg of chloroform has been shown to induce
both hepatic and renal tumors in experimental animals, although other
experiments at lower dosage levels have yielded contradictory results.
No indications of mutagenic activity have been reported.

     Administration by inhalation to rats during days 6-15 of gestation
resulted in a high incidence of fetal resorption and a few cases of
acudate fetuses.  No evidence of teratogenicity was found in another
study, in which pregnant rats and rabbits were given up to 126 and 50
mg/kg, respectively, by oral intubation.

     Species-, strain-, and sex-related differences exist with respect
to the acute lethal effects of chloroform, and both acute and prolonged
exposure to chloroform results in liver necrosis and kidney damage.

     In man, ingestion of 120 ml of chloroform has been survived, but
serious illness has been reported following ingestion of only 5 ml.
Dermal contact produces burning pain within a few minutes of exposure
and, depending on dose, erythema, hyperemia, and vesication may also

     Limited toxieologic information is available for bromoform,
dibromochloromethane, and dichlorobromomethane.  However, because of
structural similarities to chloroform, all three compounds are cause
for concern with regard to carcinogenic effects.  Preliminary results
in lung adenoma bioassays with bromoform and dichlorobromomethane
support this concern.  In addition, all three of the lesser crihalo-
methanes are mutagenic in the Ames test.  To date, other adverse health
effects have not been demonstrated, and therefore cannot be quanticaced.
                                   / j


1.   Introduction

     This section quantifies  the  exposure  of  persons  in the IKS.  to
trihalomethanes.  Several authors have  attempted  such an analysis
previously (NAS  1978, U.S. EPA  1979a).   However  this discussion will
include some new data, as well  as briefly  review  previous estimations
of exposure.  The exposure routes considered  include  ingestion via
drinking water and food, inhalation, and dermal contact.

2.   Ingestion

a.   Drinking Water

     the formation -of trihalomethanes during  chlorination of water
supplies has been discussed in  Chapters III and IV.   The  concentrations
of these chemicals in drinking  water have  been  measured in several
nationwide surveys.  The National Organics Reconnaissance Survey  (NORS)
(Symons et al. 1975) included sampling  of  trihalomethanes in raw  and
finished water from 80 water supplies.  The National  Organic Monitoring
Survey (NOMS)(U.S.  EPA 1978) also included analysis for these chemicals*in
addition to others, in samples  taken from  113 community water  supplies
during a .12-month period.  The  results from these surveys  have been
analyzed and presented in numerous ways, and have been used  in various
risk assessments (NAS 1978,  U.S. EPA 1979a, Reitz et al. 1978).  As a
result, these data are not discussed extensively here.  The  reader is
referred to the original sources and the discussions cited above.

     Table 19 shows the concentrations  of  trihalomethanes found in the
NORS and NOMS surveys (Symons .et.al. 1975, U.S. EPA 1978).   Figure 6
shows the distribution of total trihalomethanes in drinking  water from
NOMS.  Concentrations of total trihalomethanes are greater than 200 ug/1
(see Table 20) and chloroform levels exceed 100 ug/1  in many of the
locations sampled.

     The formation of trihalomethanes in chlorinated water supplies has
been discussed in Chapter IV.   It is obvious  that water source, treat-
ment method,  and analytical method influence  the observed levels  of
trihalomethanes.  The samples collected for NORS were of  finished water
at the plant, while the terminal  samples reported in NOMS had  been
allowed Co react with residual  chlorine.   Sylvia ejt al.  (1978) showed
that samples collected in the distribution system had much higher levels
than those collected at the plant.  U.S. EPA  (1978) showed that the
terminal samples contained higher levels than those that  had been
quenched (sodium thiosulfate added to react with residual  chlorine).
However, measurements taken in  the distribution system generally  showed
levels between those of the quenched and terminal samples.   Therefore,
use of these  values either underestimates or overestimates actual expo-
sure Levels.

                          SUPPLIES FROM MORS AND MOMS
                                         Concentration (mg/1)

T cihaloBBChanes


Phase I

XF-0. 271

Phase II

XF-0. 47

Phase III
Qaenehed Terminal




Dib romoehlorome chane
IJF-0. 116
XF-0. 183
XF-0. 180
local Trihalomechanes
XF-0. 457
XF-0. 784
HF-0. 295
XF-0. 69 5
hfF  noc found.
-LD  less Chan dececcion limic.
Source:  U.S. EPA (1978)


i    10

10-50     50-100     100-200

 Concentration (yg/l)
        Source: U.S. EPA (1978)


                        GREATER THAN 200 ug/1
Annondale, VA
Brownsville, TX
Camden, AR
Cape Cirardeau, MO
Charleston, SC
Columbus, OH
Hagerstown, MD
Houston, TX
Huntington, WV
Huron, SD
Illwaco, WA
Jackson, MS
Louisville, KY
Melbourne, FL
Montgomery, AL
Newport, RI
Norfolk, VA
Oklahoma City, OK
Omaha, NE
Tampa, FL
Terre Bonne Parish, LA
Wheeling, WV
 Phase II; summer terminal.
Source:  U.S. EPA (1978)

     Table 21 shows the exposures of persons  co chloroform estimated
with various assumptions regarding daily intake of water  and  the  concen-
tration of chloroform.  The range of exposures was estimated  to be  0.02-
1.2 mg chloroform per day.  These estimates are based on  the  assumption
Chat all water used in water-based beverages  comes from the same  source
of tap water.  Though this assumption may be  valid for beverages  such  as
coffee and tea," chloroform levels in soft drinks, beer, wine, etc.  would
correspond to the levels in drinking water in che area of  manufacture
and not necessarily to the levels in the area of consumption.

     Table  22  shows median and maximum exposures  for all  of the four
trihalomethanes.  Though some high concentrations of bromofora,
dibromochloromethane, and  bromodichloromethane  are  found  in drinking
water, exposure  levels  for these other trihalomethanes  are generally
much lower  than  exposure levels for  chloroform.

b-   Food

     Little  information is available on- the  levels  of chloroform  in
food.  McConnell e_t al.  (1975) apparently  have  done  the only  work to
date.  They  analyzed  various  foods in Great  Britain  and found chloroform
levels ranging from 0 ug/kg  to 33 ug/kg.   The higher levels were  found
in cheeses,  olive oil,  tea packets,  and potatoes.  The  source of  this
contaminant  is unknown,  although chloroform  has  been used as  a pesti-
cide.  It is possible that chloroform may  be formed  in  the food from
other organics.   HAS  (1978) used these data  to  estimate human exposure
to chloroform  through food.   For the minimum intake  of  food containing
the minimum  concentration  of  chloroform, exposure  is estimated to be
0.0006 ing/day;  for the  maximum intake and  maximum concentration,  the
exposure would be 0.04  mg/day.  The  average  is  estimated  to be 0.006
ing/day.   These estimates are  based on concentrations of chloroform in
uncooked  food; since  cooking  is likely to  vaporize  some of the chloro-
form, the estimated exposure  levels  are probably  overstated.   No
information  is available on  the levels of  the other  trihalomethanes in'

3.   Air

     Section IV-c described levels of  chloroform in  both  urban and  rural
air.  In order to  simplify these data  for use  in exposure  estimates, the
following assumptions were made:

                                            Concentration of
          Exposure Setting                 Chloroform  (ug/m^)

          rural  atmosphere                       0.07

          urban  atmosphere                       1

          industrial  acaosphere                 50

          indoor acnospnere                       0.5


 _  Exposure Category


   medJan concentration
     (0.059 nig/1)

   maxImi mi concentration
       (0.540 rog/1)


   median concentration
      (0.059 mg/1)

   maximum concentration
      (0.540 mg/1)
                                                     Estimated Exposure to Chloroform  (me/day)
                                     MlnJmum Intake     Maximum Intake     Reference Male      Reference Female
          Includes tan water and water-based beverages.
         2lnuikus - Adult - 365-2180 ml/day.  Taken from ICRP .(1975)
                    Children - 540-790 ml/day
                    reference male - 1650 ml/day
                    reference female - 1200 ml/day
                    reference child - 950 ml/day (It Is unclear why this Is not within  the  range  shown above.)
         Source:   Arthur  D.  Little,  Inc.

                TABLE  22.   ESTIMATED  HUMAN  EXPOSURES  TO
                            TRIHALOMETHANES  VIA DRINKING  WATER
                                                Daily exposure  (mg/day)
                                     Assuming Maximum Adult Assuming Reference M  <=
                                       Intake^ and Maximum    Intake^ and Median
Trihalomethane  Concentration (iag/1) Concentration in Water Concentration in Water
                    Median  Maximum

Chloroform           0-059   0.540             1.2                   0.1

Bromoform            0.004   0.280             0.6                   0.007
Dibromochloromethane 0.004   0.290             0.6                   0.007

Bromodichloromethane 0.014   0.180             0.4                   0.02
 2.18 liter per day
 1*65 liter per day
Source:  Arthur D. Little, Inc.

     These levels were selected  co  represent  typical  conditions,  although
 there  is considerable controversy over what may be  considered  typical
 since  the range in values  is so  great.  The value for the  indoor  occupa-
 tional atmosphere is intended  to represent concentrations  commonly found
 in such locations as beauty parlors, drug stores, photoduplicating rooms,
 etc.   Presumably persons working in these situations  would be  exposed for
 8-hour periods.  Persons using these services would be exposed for shorter
 periods, perhaps up to 4 hours per  exposure,  as in  the case of beauty

     Table 23 gives the estimated exposures associated with the above
 concentrations.  These estimates were developed for a respiratory volume
 for light activity of 19200 1 for 16 hours and 360"0 1  for  8 hours of  resting
 (ICRP  1975).  Higher respiratory volumes, such as those for athletes  or
 other  more active people,  would  result in correspondingly  higher  expo-
 sure levels.

     Monitoring data are insufficient to permit an estimate of  the size of
 the population exposed at  these  levels.   The urban exposure level shown in
Table  23 may be quite common, and exposures at this level  for  24 hours  would
not be unusual.  The estimate of 0.5 mg/day would be expected  for persons who
 work in the highly industrialized cities in the U.S.   Unfortunately,  no
 estimates of exposure can  be made for the other trihalomethanes,  but  the
 exposure levels would be expected to be considerably  lower.

 4.   Dermal Contact

     Absorption of trihalomethanes  through the skin would  be expected to
 occur  in areas where the water supply is chlorinated.   Blank (personal
 communication, as cited in Beech 1980) has measured the permeability  con-
 stant  of 125 x 10" 3 cm/hr  for chloroform.  For the median  concentration
 of 60  ug/1 in water, an exposure of 0.15 mg/hr is estimated for a  total
 body exposure.  If total body immersion occurs for  2 hours/day, the total
 maximum exposure would be  0.3 mg/day.  A more typical  immersion would be
 0.2 hours/day including bathing  and dishwashing (U.S.  EPA  1979c),  resulting
 in an  average exposure of  0.03 mg/day.  Similar calculations can  be made
 for the other trihalomethanes.   Using the same permeability constant  and
 median concentrations in water,  exposures of  0.002-0.006 mg/day were
 estimated for this route.  Maximum  exposures  ranged from 0.08-0.14 mg/day.

     Beech est al. (1980) have measured the concentration of trihalo-
 methanes in swimming pools in Florida and found mean  concentrations of
 156 ug/1 in freshwater pools (mainly chloroform) and  657 ug/1  in  saline
 pools  (mainly bromoform).  The maximum concentrations  were 430 ug/1 and
 1278 ug/1, respectively.   These  authors suggested that trihalomethanes
 are continuously formed in swimming pool water.  Using the permeability
 constant as above for chloroform and bromoform, several estimates  of
 exposure vere made for a six-year old child, as shown  in Table  24.

 Exposure Situation                    Concentration  (ug/tn3) Exposure (ing/day)

 Rural                    .                      0.07               0.002

 Urban (24 hours)                               1                  0.02

 Urban (8 hours) indoor (16 hours)           1, 0.5                0.02

 Industrial (8 hours) indoor (16 hours)      50,0.5                0.5
Source:  Arthur D. Little, Inc. 

                      IN SWIMMING POOLS
                                           Exposure (mg/day)
                                     Typical^              Maximum

Type of Pool

freshwater                             0.2                   1.3
saltwater                              0.7                   4.2
1                                   42
 Assumes a total exposure (0.88 x 10  cm ) of 1 hour per day at
 the mean concentrations of 156 ug/1 and 657 ug/1 for chloroform
 and bromoform, respectively.
 Assumes an exposure of 3 hours per day at the maximum concentrations
 of 430 ug/1 and 1278 ug/L respectively.

Note:  The permeability constant of 125 x 10   cm/hour was used
       for all calculations.
Source:  AOL estimates, from Beech (1980)

5.   Conclusions

     Both inhalation and ingestion of chloroform in drinking water
appear to be important exposure routes.  The relative importance of
these two exposure routes depends upon the location.  Inhalation may
be more important in highly industrialized areas, whereas chloroform
in drinking water may be more important where water is chlorinated and
levels in air are low.  Ingestion of chloroform in food does not appear
to be an important exposure route.  Dermal contact of persons spending
long periods in swimming pools may result in high exposures to chloroform.

     Exposure to the other trihalomethanes occurs primarily through
drinking water, although dermal contact with these compounds may result
in significant exposure in some situations.  The results in Table 24
suggest that dermal exposures can be high for persons spending a signifi-
cant amount of time in swimming pools.  It should be noted that-inhalation
exposures would probably be occurring at the same time.   Beech (1980)
calculated that inhalation exposures would be about one-half as large as
dermal exposures under the same conditions.

Alavanga, M.; Goldstein, I.; Susser, M.  A case control study of gastro-
intestinal and urinary tract cancer mortality and drinking water chlor-
ination.  Chapter 30.  Jolley, R.L.; Gorchev, H.; Hamilton, D.H. Jr.
eds.' Water chlorination:  environmental impact and health effects.
Vol. 2.  Ann Arbor, Michigan:  Ann Arbor Science Publishers, Inc.; 1978:

Anders, M.W.; Stevens, J.L.; Sprague, R.W.; Shaath, Z.; Ahmed; A.E.
Metabolism of haloforms to carbon monoxide. II.  _In vivo studies.  Drug
Metab. 6(5):556-560; 1978.

Beech, J.A.  Estimated worst case trihalomethane body burden of a child
using a swimming pool.  Medical Hypotheses 6:303-307; 1980.

Beech, J.A.; Diaz, R.; Ordaz, C.; Falomeque, B.  Nitrates, chlorates
and"trihalomethanes in swimming pool water.  Accepted for publication.
Am. J. Public Health; 1980.

Bowman, F.; Borzelleca, J.F.; Munson, A.E.  The toxicity of some halo-
methanes in mice.  Toxicol. Appl. Pharmacol. 44(1) -.213-216; 1978.

Brown, D.M.; Langley, P.F.; Smith, D.; Taylor, D.C.  Metabolism of
chloroform - I:  the metabolism of  (14C) chloroform by different species.
Xenobiotica 4(3):151-163;  1974.

Brownlee, K.A. e^ al.  The up-and-down method with small samples.  Amer.
Statist. Assoc. J. 48:262; 1933.  (As cited in U.S. EPA 1977).

Capel, I.D.; Williams, D.C.  The effect of chloroform ingestion on the
growth of some murine tumours.  IRCS Medical Science 6:435; 1978.

Cantor, K.P.; Hoover, R.; Mason, T.J.; McCabe, L.J.  Associations of
cancer mortality with halomethanes in drinking water.  J. Nat. Cancer
Inst. 61(4):979-985; 1978.

Cresteil, T.; Beaune, P.; Leroux, J.P.; Lange, M.; Mansuy, D.  Biotrans-
formation of chloroform by rat and human liver microsomes; in vitro
effect on some enzyme activities and mechanism of irreversible binding to
tnacromolecules.  Chem. Biol. Interact. 24(2):153-165; 1979.

DeRouen, T.A.; Diem, J.E.  Relationships between cancer mortality in
Louisiana drinking-water source and other possible causative agents.
Incidence of Cancer in Humans.  Hiatt, H.H.; Watson, J.D.; Winston, J.A.
eds.  Proceedings of the Cold Spring Harbor Conferences on origins of
human cancer. Vol. 4.  Cold Spring Harbor:  Cold Spring Harbor Laboratory;
1975:  1&0-1A3.

DeSalva, S.; Volpe, A.; Leigh, G.; Regan, T.  Long-term safety studies
of a chloroform-containing dentifrice and mouth-rinse in man.  Food
Cosmet. Toxicol. 13(5):529-532; 1975.

Dykan, V.A.  Changes in liver and kidney functions due to methylene
bromide and bromoform.  Nauchn. Tr. Ukr. Nauch.-Issled. Inst. Gig. Tr.
Profzabol. 29:82-90; 1962.  (As cited in HAS 1977).

Eschenbrenner, A.B.; Miller, E.  Induction of hepatomas in mice by
repeated oral administration of chloroform with observation on sex
differences.  J. Nat. Cancer Inst. 5:251-255; 1945.

Fry, B.J.; Taylor, I.; Hathaway, D.E.  Pulmonary eliminations of chloro-
form and its metabolites in man.  Arch. Int. Phannacodyn. Ther. 196:98;

Grant, W.M.  Toxicology of the eye.  Second Edition.  Springfield,
IL:  Charles C. Thomas; 1974:  267-268.  (As cited in U.S. EPA 1979a).

Heywood, R.; Sortwell, R.J.; Noel, P.R.; Street, A.E.; Prentice, D.E.;
Roe, F.J.; Wadsworth, P.F.; Worden, A.N.; Van Abbe, N.J.  Safety evalua-
tion of toothpaste containing chloroform. III. Long-term study in beagle
dogs.  J.. Environ. Pathol. Toxicol. 2(3):835-851; 1979.
Hill, R.'N.; Clemens, T.L.; Liu, D.K. e al.  Genetic control of chloro-
form toxicity in mice.  Science 190:159-161; 1975.

Hill, R.N.  Differential toxicity of chloroform in the mouse.  Ann. N.Y.
Acad. Sci. 298:170-175; 1978.

Ilet, K.F.; Reid, W.P.; Sipes, I.G.; Krishna, G.  Chloroform toxicity in
mice:  correlation of renal and hepatic necrosis with covalent binding
of metabolites to tissue macromolecules.  Exp. Mol. Path. 19:215-229;
1973.  (As cited in U.S. EPA 1979b).

International Agency for Research on Cancer (IARC).  Working group on the
evaluation of the carcinogenic risk of chemicals to man.  IARC
monograph on the evaluation of the carcinogenic risk of chemicals to man.
1:61-65; 1971.

International Commission on Radiological Protection (ICRP).  Report of
the Task Group on Reference Man.  New York, NY:  Pergamon Press; Adapted
October 1974.

Jones, W.M.; Marguis, G.; Stephen, C.R.  Hepatotoxicity of inhalation
anesthetic drugs. Anesthesiology 19:715-723; 1958.   (As cited in U.S.
EPA 1979a).

Kimura, S.T.; Ebert, O.M.; Dodge, P.W.  Acute toxicity and liai:s of
solvent residue for 16 organic solvents.  Toxicol.  Appl. Pharmacoi.
19:699-704; 1971.

Klassen, C.D.; Flaa, G.L.  Relative effects of chlorinated hydrocarbons
on liver and kidney function in mice.  Toxicol. Appl. Pharmacol. 9:139-
151; 1966.

Kraybill, H.F.  Carcinogenesis induced by trace contaminants in potable
water.  Bull. N.Y. Acad. Med.- 54(4) :413-427; 1978.

Kucob, S.D.; Plaa, G.L.  A procedure for estimating the hepatoxic poten-
tial of certain industrial solvents.   Toxicol. Appl. Pharmacol. 4:354-
361; 1962.

Lucas, G.H.S.  A study of the fate and toxicity of bromine and chlorine
containing anesthetics.  J. Pharm. Exp. Ther. 34:223-227; 1928.

Maiten, J.E.; .Spruit, D.; Boemaars, H.G.M.; deKeizer, M.J.M.  Horny
layer injury by solvents-.  Verufsdermatosen 16:135-147; 1968.  (As cited
in U.S. EPA 1979b).

McConnell, G.; Ferguson, D.M.; Pearson, C.R.  Chlorinated hydrocarbons
and the environment.  Endeavor 34(121):13-18; 1975.

Miklashevskii, V.E.; Tugarinova, V.N.; Rakhmanina, M.L.; Yakovleva, G.P.
Toxicity of chloroform administered perorally.  Hyg. Sanit. 31:320-322;

National Academy of Sciences (NAS).  Drinking water and health.
Washington, DC:  NAS; 1977.

National Academy of Sciences (NAS).  Chloroform, carbon tetrachloride,
and other halomethanes:  An environmental assessment.  Washington, DC:
NAS; 1978.  294 p.

National Cancer Institute (NCI).  Report on carcinogenesis bioassay of
chloroform.  March 1, 1976.

Page, T.; Harris, R.H.; Estein, S.S.   Drinking water and cancer mortality
in Louisiana.  Science 193:55-57; 1976.

Palmer, A.K.; Street, A.E.; Roe, F.J.; Worden, A.N.; Van Abbe, N.J.
Safety evaluation of toothpaste containing chloroform. II. Long-term
studies in rats.  J. Environ. Pathol. Toxicol. 2(3):321-833; 1979.

Patty, F.A.  Industrial hygiene and toxicology, Vol. II. 2nd Ed.  New
York, NY:  Interscience Publishers; 1963.

Petzold, G.L.; Swenberg, J.A.  Detection of DNA damage induced in vivo
following exposure of rats to carcinogens.  Cancer Res. 38(6):1589-1594;

?ohl, L.R.; Booshan, 3.; Whittaker, N.F.; Krishna, G.  Phosgene:  A
mecabolice of chlorofora.  Biochem. Biophys. Res. Commun. 79(3):53&-59I;


Poobalasingham, N.; Payne,. J.P.  The uptake and elimination of chloro-
form in man.  Br. J. Anesth. 50(4):325-329; 1978.

Powers, M.B.; Voelker, R.W.  Evaluation of the oncogenic potential of
chloroform by long-term oral administration in rodents.  (Abstract)
Toxicol. Appl. Pharmacol. 37(1):179; 1976.

Registry of Toxic Effects of Chemical Substances  (RTECS).  Washington,
DC:  Department of Health, Education, and Welfare, National Institute of
Occupational Safety and Health; 1977.

Reitz, R.H.; Gehring, P.J.; Park, C.N.  Carcinogenic risk estimation for
chloroform - an alternative to EPA's procedures.  J. Food & Cosmet.
Toxicol. 16:511-514; 1978.

Renne, R.A.; Ferrell, J.F.; Voelker, R.E.; Powers, M.B.  Pathology of
long-term oral administration of chloroform in rodents.  Toxicol. Appl.
Pharmacol. 37(1):179-180; 1976.

Roe, F.J.C.; Carter, R.L.; Mitchley, B.C.V.  Tests of miscellaneous sub-
stances for carcinogenesis.  Test of chloroform and 8-hydroxyquinoline for
carcinogenicity using newborn mice.  A.R. Brit. Emp. Cane. Camp. 46:13;
1968.  (As cited in IARC 1971).

Roe, F.J.C.; Palmar, A.K.; Worden, A.N.; Van Abbe, N.J.  Safety evalua-
tion of toothpaste containing chloroform.  Part I. Long-term studies in
mice.  Abstract.  Plaa, G.L.; Duncan, W.A. eds.  International congress
on toxicology; 1978.  New York, NY:  Academic Press; 1979.

Rudali, G.  A-propos de 1'activite oncogene de quelques hydrocarbures
halogenes utilises en therapeutique.  UICC Monogr. Ser. 7:138; 1967.
(As cited in IARC 1971).
Sax, N.I.  Dangerous properties of industrial materials.  5th ed.  New
York, NY:  Van Nostrand Reinhold Company; 1979.

Schroeder, H.G.  Acute and delayed chloroform poisoning  a case report.
Brit. J. Anaesth. 37:972-975; 1965.

Schwetz, B.A.; Leong, B.K.J.; Gehring, P.J.  Embryo- and feto-toxicity
of inhaled chloroform in rats.  Toxicol. Appl. Pharmacol.  28:442-451;

Simmon, V.F.; Tardiff, R.G.  The mutagenic activity of halogenated
compounds found in chlorinated drinking water.  Water Chlorination
2:417-431; 1978.

Stevens, J.L.; Anders, M.W.  Studies on the mechanism of Che metabolism
of haloforns co carbon monoxide.   Toxicol. Appl. Pharmacol. 45(1):297;

SCurrock, J.  Lack of mutagenic efface of haloethane or chloroform on
cultured cells using Che azaguanine test system.  Brit. J. Anaesth.
49(3):207-210; 1977.

Sylvia, A.E.; Miller, J.D.; Baxter, W.S.  Preliminary investigations
into the presence of haloforms in Massachusetts drinking water supplies.
J. M.E. Water Works.  March: 30-40; 1978.

Symons, J.M.; Bellar, T.A.; Carswell, J.K.; OeMarco, J.; Kropp, K.L.;
Robeck, G.G.; Seeger, D.R.; Slocum, C.J.; Smith, B.L.; Stevens, A.A.
National organics reconnaissance survey for halogenated organics.  J.
Am.  Water Works Assn. 67:634-646; 1975.

Taylor, D.C.; Brown, D.M.; Keeble, R.; Langley, P.F.  Metabolism of
chloroform - II: A sex difference in the metabolism of (14C) chloroform
in mice.  Xenobiotica 4(3):165-174; 1974.

Theiss, J.D.; Stoner, G.D.; Shimkin, M.B.; Weisburger, E.K.  Test for
carcinogenicity of organic contaminants of United States drinking waters
by pulmonary tumor response in Strain A mice.  Cancer Res. 37:2717-2720;

Thompson, D.J.; Warner, S.D.; Robinson, V.B.  Teratology studies on
orally administered chloroform in the rat and rabbit.  Toxicol. Appl.
Pharmacol. 29:348-357; 1974.

Timms, R.M.; Moser,  K.M.  Toxicity secondary to intravenously admini-
stered chloroform in humans.  Arch. Intern. Med. 135:1601-1603; 1975.

Torkelson, T.R.; Oyen, F.; Rowe, V.K.  The toxicity of chloroform as
determined by single and repeated exposure of laboratory animals.  Am.
Md.  Hyg. Assoc. J. 37(12):697-705; 1976.

Uehleke, H.; Werner, T.; Griem, H.; Kramer, M.  Metabolic activation of
haloalkanes and tests in vitro for mutagenicity.  Xenobiotica 7(7):393-
400; 1977.

U.S. Environmental Protection Agency (U.S. EPA).  Water quality criterion
document.  Interim draft. No. II.  Washington, DC:  Office of Water
Planning and Standards, U.S. EPA; November 3, 1977.

U.S. Environmental Protection Agency (U.S. EPA).  National organics
monitoring survey.  Unpubl.  Washington, DC:  Technical Support Division,
Office of Water Supply, U.S. EPA; 1978.

U.S. Environmental Protection Agency (U.S. EPA).  Statement of basis and
purpose for an amendment to the National Interim Primary Drinking Water
Regulations on Trihalomethanes.  Washington, DC:  Office of Drinking
Water, Criteria and Standards Division, U.S. E?A; 1979a.

U.S. Environmental Protection Agency  (U.S. EFA).  Chloroform  The
consent decree ambient water quality  criteria document.  Washington,  DC:
Office of Water1 Planning and Standards, U.S. EPA; 1979b.

U.S. Environmental Protection Agency  (U.S. EFA).  Identification  and
evaluation of waterborae routes of exposure from other  than food  and
drinking water.  .Report No. EPA-440/4-79-016.  Washington, DC:  Office
of Water Planning and Standards, U.S. EPA; 1979c.

Van Dyke, R.A.; Chemoweth, M.B.; Van  Poznak, A.  Metabolism of volatile
anesthetics - I:  conversion in vivo  of several anesthetics to 1*C02
and chloride.  Biochem. Pharmacol. 13:1239-1247; 1964.

Vessell, E.S.; Lang, C.M.; White, W.J.; Passanante, G.T.; Hill, R.N.
_et al.  Environmental and genetic factors affecting the response  of
laboratory animals to drugs.  Fed. Proc. 35:1125-1132;  1976.

Wallace, C.J.  Hepatitis and nephrosis due to cough syrup containing
chloroform.  Calif. Med. 73:442; 1959.  (As cited in U.S. EPA 1979a).

Winslow, S.G.; Gerstner, H.B.  Health aspects of chloroform:  a review.
Drug. Chem. Toxicol. 1(3):259-276; 1978.

                              CHAPTER VI



1.   Introduction

     This section provides information about the levels of trihalometh-
anes that disrupt the normal behavior and metabolic processes of aquatic
organisms.  Despite a thorough literature search, no toxicity data for
dibromochloromethane or dichlorobromomethane were found.  Moreover,
information on chloroform and bromoform was limited.  In assessing
threshold effects levels, it is important to note that, with few excep-
tions, the bioassays were conducted under static conditions, and the
toxicant concentrations were calculated rather than measured.  Conse-
quently, the LC5Q values (concentrations lethal to 50% of test organisms)
reported could underestimate actual toxicity, as a result of toxicant
evaporation and metabolism by test organisms.

2.   Chloroform

     Earlier studies on the aquatic toxicity of chloroform reported
"negative" (avoidance) reactions as a measure of effects levels.  Clayberg
(1917) observed "positive reactions" (not defined) in brown bullheads and
white suckers in 214 mg/1 of chloroform, and "a few negative reactions
... in somewhat higher concentrations."  Jones (1974) recorded an
avoidance reaction by the ten-spined stickleback in 100 mg/1.  Fifty
percent of a group of goldfish were anesthetized by a 167 mg/1 solution
of chloroform (Gherkin and Catchpool 1964).  This effect was observed to
decrease with increasing temperature.

     More recent research on the acute toxicity of chloroform has yielded
lower overall effects levels than the studies described above.  The
lowest reported 96-hour LC5Q value for a marine organism is 28 mg/1, for
the dab (Pearson and McConnell 197S).  Equally susceptible is Daphnia
magna. with a 48-hour LC5Q value of 28.9 mg/1 chloroform (U.S. EPA 1978a).
The "chronic value" (as defined by U.S. EPA) for this species is 2.5 mg/1.
The 96-hour LC5Q for the pink shrimp, a marine invertebrate, is 81.5 mg/1,
as reported by Bentley.et.al. (1975).

     Clayberg (1917) observed mortality in the orange-spotted sunfish
after 1 hour of exposure to concentrations ranging from 107 mg/1 to
153 ag/1.

     Anderson t al. (1979) conducted 96-hour acute toxicity bioassays
with largemouth bass, rainbow trout, and bluegill sunfish.  The respec-
tive LCjQ values for chlorofora vere 45-56 oig/1, 15-22 ag/1, and 13-22
aig/1.  In a comparative scudy, Bencley ec al. (1975) found rainbow crouc

to be more sensitive to chloroform than bluegill sunfish, with respec-
tive 96-hour LCso'sof 66.8 mg/1 and 115.0 mg/1 in water 35 mg/1 hardness.
When the hardness was increased to 200 mg/1 and the bioassays repeated,
96-hour LCso's values decreased to 43.8 mg/1 and 110.0 mg/1, respectively,
a reduction indicating significantly greater toxicity in hard water for
the rainbow trout.  The data available regarding the toxicity of chloro-
form to aquatic organisms are summarized in Table 25.

3.   Bromoform

     The data for bromoform are limited and are all based on static
bioassays.  The bluegill sunfish was the only freshwater finfish tested
for sensitivity to bromoform.  The 96-hour LC$Q for this species was
29.3 mg/1.  Daphnia magna experienced median lethality after 48 hours of
exposure to 46.5 mg/1 bromoform.  For the alga (Selenastrum capricornutum).
the 96-hour EC5Q (effects observed in 50% of test organisms) level was
172 mg/1 with respect to chlorophyll-^ activity, and 116 mg/1 as deter-
mined by cell number (U.S. EPA 1978b).

     The effects of bromoform were also measured for three marine spe-
cies.  The 96-hour LC$Q for the sheepshead minnow was 17.9 mg/1 bromo-
form, while the. chronic value was 9.2- mg/1.  The mysid shrimp was
apparently somewhat less susceptible to bromoform, with a 96-hour LC5Q
value of 24.4 mg/1.  The alga, Skeletonema costatum, was tested for
bromoform effects in terms of chlorophyll-^ activity and proportion of
cells affected.  The respective 96-hour 50 concentrations were 12.3
mg/1 and 11.5 mg/1, respectively (U.S.  EPA 1978b).

     In a study by Gibson _et_ al. (1979), five marine species were tested
for sensitivity to bromoform in continuous-flow bioassays.  An LCjQ
value was not obtained for the littleneck clam (Protothaca staminea)
because, in high concentrations of bromoform, the clams closed up and
did not pump water.  However, mortality was observed in 5 mg/1 and
10 mg/1 bromoform.  At concentrations above 10 mg/1, the Atlantic oyster
(Crassostrea virginica) and the quahog (Mereenaria mereenaria) also
closed and did not circulate water, and this resulted in no mortalities
during the 96-hour bioassay.  However,  when exposure was continued for
an additional 3 days, several organisms died.  Ninety-six-hour LC5Q
concentrations were then estimated at 40 mg/1 for the oyster, and 140
mg/1 for the quahog.  For the brown shrimp (Penaeus aztecus) and the
menhaden (Brevoortia tyrannus). 96-hour LC5Q values were given as 26 mg/1
and 12 mg/1, respectively.  These data are summarized in Table 26.

4.   Summary

     The LC5Q values for the bluegill included the lowest and highest
values recorded, from 13 mg/1 to 115 mg/1 chloroform.  Daphnia magna and
rainbow trout were found to be comparatively sensitive, with LC5Q values
for chloroform of 28.9 mg/1 and 15 mg/1, respectively."  The limited data
do not indicate whether marine or freshwater species are generally more
sensitive to chloroform.  In one bioassay, 'rainbow trout were signifi-
cantly more susceptible in hard water than in soft water.  No data were
available on the toxicity of chloroform to aquatic plants.

                                               TAULE  2).   EFFECTS Of CHLOROFORM OH AQUATIC ORGANISMS
Ciiiiruiit rail tin
      41. H
 lOfc.B'J-IV,!. 7

 (l.lmanda up.)

Ualnbow  trout
 (Sal mo  Kalrdncrl
Rainbow  trout
 (Salmo  Kalrdnerl)
fink shrimp
 (I'mtaciia  duorarum)
fun-up 1 nud bllckleback
Oiange-spotted sunflah
 (l.tipoials huinllls)

BluuBlll sunfish
 (Lepoiala m.icroclilrus)

Illueglll sun Mali
 (Lepomla macrochlma)

 (Carasalua amatua)

Brown bullhead
 (Icculiiriiii nebulosus)
     *zn.U       White uuckcr
                        j^ouiiib I'ooiimirboul)
                             Hardness (ug/1)
                             Salinity o/oo     pll   Temperature     Teat Type   Test Duration       Effects
                                     TABLE 26.   EFFECTS OF BHOMOFUKM ON AQUATIC OKCANISMS

12. J

l.lttleneck clan
(Protothaca atamtnea) Marine
Shecpsliead minnow
(Cyprlnodon varlegatiis)
(Skctutonema costacuai)
cell number
(Brevoortla tyrannuu) Marine
Sheepshead nlnnuw
(Cyprlnodon varleaatiia) Marine
Mysld shrimp
(Hyuldopsls bahla) Marine
Brown shrimp
(Penacus aztecus) Marine
Blueglll sun f I ah
(I.epouia macrochlrua) Freshwater
Atlantic oyster
(Craaaostrea vlrelnlca) Marine
Paphnla aa^na  Freshwater
(Solcnantruin capr Iconiutuia)
cell number
Teat Duration
96 hr

96 hr
96 hr
96 hr
96 hr
96 hr
96 hr
96 hr
96 hr
48 hr

96 hr
96 hr
Chronic Value


                                                                                                                  et  .il.  (1979)

                                                                                                          U.S.  EPA  (1978)

                                                                                                          U.S.  EHA  (197U)

                                                                                                          U.S.  EPA  (1978)

                                                                                                          Gibson t  al.  (1979)

                                                                                                          U.S.  EPA  (1978)

                                                                                                          U.S.  EPA  (1978)

                                                                                                          Gibson e_t  a_l.  (1979)

                                                                                                          U.S.  EPA  (1978)

                                                                                                          Glbuon et  al^.  (1979)

                                                                                                          U.S.  EPA  (1978)

                                                                                                          U.S EPA (1978)

                                                                                                          U.S.  EPA (1978)

     Although information on bromoform toxicity was equally scarce,
bromoform appears to be somewhat more toxic to aquatic life than chloro-
form.  The LC5Q values ranged between 12 mg/1 for the menhaden and
46.5 mg/1 for Daphnia magna (the one species tested that exhibited less
sensitivity to bromoform than to chloroform).  For the marine alga
species tested, EC^o concentrations were 11.5- mg/1 and 12.3 mg/1,
approximately one order of magnitude less than for the freshwater alga.

     Monitoring data for the trihalomethanes considered here are
extremely sparse; consequently, any generalizations made with regard
to exposure levels must be strictly qualified.  Since the levels
observed were in many cases at or below the detection limits of the
analytic procedure, future improvements in detection ability could
conceivably lower the estimates made below.

     Ambient concentrations of chloroform in water normally fell between
0.1 ug/1 and 10 ug/1, with a small proportion of measurements exceeding
10 ug/1 (see Chapter IV).  In a few instances the levels exceeded 100
ug/L  Nearly all (97%) of the bromoform measurements were between 1.0
ug/1 and 10 ug/1.  Roughly two-thifds of observations concerning both
bromodichloromethane and dibromochloromethane fell into the 0.1-1.0
ug/1 category, with approximately one-third exceeding 1.0 ug/1.  The
Pacific Northwest and California basins were the most intensively sampled
watersheds for the latter two trihalomethanes.  Perhaps coincidentally,
the concentrations in the Pacific Northwest for both were usually between
1.0 ug/1 and 10 ug/1, while California had similar levels of dibromo-
chloromethane .

     In addition, no fish kills attributed to chloroform have been
reported in the data files of the Monitoring and Data Support Division,
Office of Water Regulations and Standards, U.S. EPA.

     Most of the data for bromoform, bromodichloromethane, and dibromo-
chloromethane were remarked, which probably means the chemical was
detected but the concentration was not quantified.  However, the fact
that the levels of all of the trihalomethanes were usually so low as to
approach the detection limits may suggest that ambient exposure levels
in most areas of the U.S. are negligible.  Factors that modify the
toxicity of trihalomethanes for aquatic biota have not been identified  
or studied sufficiently to merit a discussion of geographical variations
in such factors.

Anderson, D.R. _et_aj.  Progress Reporc Covering Period January 1 through
March 31, 1979:  Biocide by-products in aquatic environments.  PNL-2988.
U.S. Nuclear Regulatory Commission; 1979.

Bentley, R.E.; Heitmuller, F.; Sleight, B.H.; Parrish, P.R.  Acute  coxicity
of  chloroform  to bluegill  (Lepomis macrochirus), rainbow  trout  (Salmo
gairdneri). and pink shrimp (Penaeus durorarum).  Contract No. WA-6-1414-B.
Washington, DC:  U.S. Environmental Protection Agency; 1975.

Gherkin, A.; Catchpool, J.F.  Temperature dependence of anesthesia in
goldfish.  Science 144:1460-1462; 1964.

Clayberg, H.D.  The effect of ether and chloroform on certain fishes.
Biol. Bull. 32:234; 1917.

Gibson, C.I.; Tone, F.C.; Wilkinson, P.; Blaylock, J.W.  Toxicity and
effects of bromoform on five marine species.  PNL-3023.  U.S. Nuclear
Regulatory Commission; 1979.  Available from NTIS, Springfield, VA;

Jones, J.R.E.  The reactions of Pygosteus pungitius L. to toxic solutions.
J. Exp. Biol. 24:110; 1974.

Pearson, C.R.; McConnell, G.  Chlorinated C^ and G hydrocarbons in the
marine environment.  Proc.* R. Soc. Lond. B 189:305-322; 1975.

U.S. Environmental Protection Agency (U.S.EPA).  Ambient water quality
criteria.  Criterion document  chloroform.  Washington DC:   Criteria
and Standards Division, Office of Water Planning and Standards, U.S.
EPA; 1978a.  Available from:  NTIS, Springfield, VA; PB 292 427.

U.S. Environmental Protection Agency (U.S. EPA).  Ambient water quality
criteria.  Criterion document  halomethanes.  Washington, DC:  Criteria
and Standards Division, Office of Water Planning and Standards, U.S. EPA;
1978b.  Available from:  NTIS, Springfield, VA; PB 296 797.


     This chapter assesses risk to both humans and non-humans associated
wich exposure to chloroform, bromoform, dibromochloromethane, and
bromodichloromethane.  However, only the risks associated with exposure
to chloroform can actually be quantified, since sufficient information
on exposure and effects was not available for the other three trihalo-

     Several different dose-response models were applied to the available
data concerning the carcinogenic effects of chloroform, and the range of
potential human risk was estimated for various possible exposure levels.
These risk estimates have been compared against other risk estimates
derived from the literature, although the latter usually focus upon
drinking water alone as the pathway of exposure.
     With respect to non-human biota, .little information is available on
the effects of environmental exposure to trihalomethanes; therefore risks
to non-human biota were addressed only in qualitative terms.


1.   Effects of Trihalomethanes

     Chapter V discusses the effects of trihalomethanes in laboratory
animals and humans in some detail.  Tables 27-30 summarize these effects.
Chloroform has been shown to be carcinogenic in rats and mice.  However,
no evidence indicates teratogenicity or mutatagenicity.  Acute effects
attributable to chloroform include liver necrosis and kidney damage.
It is likely that these effects would not be observed in the general
population at the present exposure levels.

     The other three trihalomethanes have not been tested for carcino-
genicity; however, they all are potential mutagens based on the results
of the Ames bioassay.

     As a result of the available data, the  only risks that can be
quantified at this time are the potential cancer risks due to chloroform
exposure, which are considered below.

2.   Carcinogenic!ty of Chloroform

     The potential carcinogenic effects of chloroform upon humans can be
quantitatively estimated through extrapolation of ia vivo laboratory
results.  The available data concerning mammalian effaces were discussed
previously in Chapter V and summarized in Table 27.   For extrapola-
tion purposes, the NCI data-shown in Table 31 have been selected.

Adverse Effect
                                          No Apparent
Species     Lowest Reported Effect Level  Effect Level
 138 mg/kg (diec)
Renal Epithelial
  90 mg/kg (diet)



                                      60 mg/kg (gavage)

                                     490 mg/m3/hr inhal
                           17 mg/kg
                                                                  126 mg/kg

                                                                   50 mg/kg
Hepatic and Renal       Rat
Necrosis                Mouse
               250 mg/kg (orally)
Oral LD
119 mg/kg
Source:  Chapter V.

                                                               No Apparent
Adverse Effect      Species      Lowest Reported Effect Level  Effect Level
Pulmonary Adenoma    Mouse             48 mg/kg ip x 23       4 mg/kg ip x 18
Mutagenicity        Salmonella            50 ul (vapor)
                    TA 100
Teratogenicity                      No data available
Chronic Toxic         	              No data available
Median Oral          Mouse               1400 mg/kg
  Lethal Dose
Source:  Chapter V.

                                                             No Apparent
Adverse Effect    Species      Lowest Reported Effect Level  Effect Level
                  No data available
Mutagenicity     Salmonella
                 TA 100
                    10 ul (vapor)
Chronic Toxic
                  No data available

                  No data available
Median Oral
  Lethal .Dose
800 mg/kg
Source:  Chapter V.

Adverse Effect
                              No Apparent
Lowest Reoorted Effect Level  Effect Level
Pulmonary Adenoma    Mouse
                     100 mg/kg i.p. x 24    40 mg/kg i.p. x 24
TA 100
        50 ul (vapor)
Chronic Toxic
                     No data available

                     No data available
Median Oral
  Lethal Dose
          450 mg/kg
Source:  Chapter V.

Dosage(mg/kg/day)  Response (%)    Tumor Site
NCI Study
male rat

male mouse

female mouse

ICI Study
male mouse



                                                    renal cortex
Source:  Reuber (1979).

demonstracing increased renal cumors in mice, and increased hepatic
tumors in male rats.  The ICI data also indicate carcinogenic activity
at a lower dose, but the NCI data form a larger and more consistent
data base for extrapolation.  It must be noted that interpretation of
these results for human risk assessment is subject to a number of impor-
tant qualifications and assumptions:

       Though positive carcinogenic findings exist, there have
        also been negative findings in tests with several species.
        Thus the carcinogeniclty of chloroform to humans is far
        from certain.

       Assuming that the positive findings indeed provide a
        basis for extrapolation to humans, the estimation of
        equivalent human doses involves considerable uncertainty.

       Due to inadequate understanding of the mechanisms of
        carcinogenesis, there is no scientific basis for selecting
        among several alternate dose-response models, which yield
        widely differing results.

       Apart from the risk via ingestion, there may be rieks due
        to either dermal absorption or inhalation of chloroform.
        However, toxicological data corresponding to these routes
        are not available for extrapolation purposes.

In order to deal with the large uncertainties inherent in extrapolation
to humans, three commonly-used dose-response models have been applied
in order to establish a range of potential human risk.

     The first step in extrapolation- of laboratory data is to determine
the equivalent human dose.  Reitz t al. (1978) showed that considera-
tion of metabolic rates in rodents led to a more consistent interpreta-
tion of laboratory dose-response data than did the U.S. EPA method of
correcting for surface area.  They argue that humans are less sensitive
to chloroform than rodents due to their slower metabolic rate (they
maintain that the effects of chloroform are due to a .toxic metabolite),
and that past efforts at extrapolation have over-estimated the human
risk.  Accordingly, the rodent doses have been used as a basis for
extrapolation in this analyses, without an attempt to convert them to
a human equivalent dose.  If a surface area conversion factor were
introduced, it would decrease the equivalent human dose by a factor of
6 for rat data or 14 for mouse data; thus the conversion would imply
considerably higher human risk.

     The three dose-response models used to extrapolate human risk were
the linear "one-hit" model, the log-probit model, and the multistage
model.  The latter is actually a generalization of the one-hit model,
in which the hazard rate was permitted to be a quadratic rather than a
linear function of the dose.  All of these models are well known in the

literature, and a theoretical discussion may be found in Arthur 0.
Little (1980).  The one-hit models assume that the probability of a
carcinogenic response is described by

          P (response at dose X)  1 - e  ""

where h(x) is the "hazard rate" function.  The log-probit model assumes
that human susceptibility varies with dose, according to a log-normal
distribution.  Due to their differing assumptions, these dose-response
models usually give widely differing results when effects data are
extrapolated from high laboratory doses to the low doses typical of
environmental exposure.

     The results of these extrapolations are shown in Table 32 for human
exposures ranging from 0.01 mg/day to 10 mg/day.  The linear model may
be applied to several different groups of data, yielding a range of risk
estimates.  The best, and most conservative, fit was obtained using the
female mouse data for liver cancer, which indicate a coefficient of
about 8.6 x 10"5 increase in risk per mg/day increase in human dose.
The log-probit extrapolation was performed using a unit slope with
respect to the log-dose, which is generally accepted as a conservative
procedure.  Again, a wide range of estimates can be derived by selecting
different groups of data.  In order to provide a less conservative esti-
mate of the risk, the male rat data were also utilized for kidney cancer,
which predicted risks considerably lower than did the female mouse data.
Finally, the multistage model was fit to the combined data for liver
cancer in male and female mice.

3.   Literature Review of Carcinogenic Risks of Chloroform

     Since chloroform was first identified in drinking water in 1974,
there has been a great deal of interest in the assessment of carcino-
genic risk from trihalomethane exposure (Stokinger 1977, Tardiff 1977,
NAS 1977, MAS 1978, Reitz &t al. 1978, U.S. EPA 1979).  These analyses
contain widely varying results, which are discussed briefly below.

     Tardiff (1977) used four different models for extrapolation of
carcinogenic effects observed in laboratory animals at high doses to
effects that may be anticipated at low doses, but did not convert data
from one species to another.  His analysis was based upon a maximum
human dosage of 0.01 mg/kg chloroform per day, and utilized the NCI data
on the carcinogenicity of this chemical.  His estimates, based on four
different extrapolation models, range from 0 to 0.84 additional cancers
per million population each year.  This is equivalent to a maximum of
60 lifetime tumors per million population at a dose of 0.7 mg/day, which
agrees with the estimates in Table 32.  Tardiff concludes that between
0% and 1.6% of cancer incidence in liver or kidneys may be attributable
to chloroform in drinking water.

     The National Academy of Sciences (1978) ristc estimation was based
on che multistage model of Guess and Crump, and a species conversion of

                       DUE TO CHLOROFORM EXPOSURE
                                         Exposure level
Extrapolation Model                         (me/day)

                                   0.01     0.1       1     10
Linear Model l                     0.86     8.6      86     860

Log-probit Model l         '         N       0.3      32    1300

Log-probit Model2                   N        N       0.8     72

Multi-stage Model3                  .06      .65     6.5     65
Note:  Use of a dose conversion factor would increase all
       estimates by an order of magnitude
N  negligible

  Based on NCI data for female mouse.

  Based on NCI data for male rat (renal tumors).

  Based on NCI data for male and female mouse.

 carcinogenicicy data  based  on  surface  area.   Using Che NCI carcino-
 genicicy data, the  authors  of  Che MS  study  found  that an oral dose
 of  1 ug chloroform  per  day  resulted  in an  estimated lifetime risk of
 cancer in a  70-kg man of  as much as  2  x 10"^.   This is considerably
 greater than the estimate of Tardiff,  when one considers  the extremely
 low dose assumption.  However  the majority of this difference is due to
 the surface  area conversion factor,  which  Tardiff  did  not include.   The
 estimates in Table  32 would be increased by  about  an order of magnitude
 if  that conversion  factor were used.

 4.   Human Exposure Scenarios

     Table 33 estimates chloroform exposures for five  hypothetical
 exposure situations:  a rural  area with and  without a  chlorinated
 drinking water supply,  an urban setting, an  industrial setting,  and
 maximum exposures for an  adult and child.  A number of assumptions  have
 been used to construct  the  total exposure  estimates.   First, 100% absorp-
 tion has been assumed for oral and inhalation routes.   Second,  the
 carcinogenic activity of  chloroform  has been assumed to be similar  for
 these routes, thus  allowing the summation  of exposures for purposes of
 risk evaluation.  There are no data  to support the validity of  these
 assumptions,  but they permit consideration of the  relative contributions
 of  various exposure routes  to  total  risk.  For the rural,  urban,  and
 industrial settings,  the  concentrations chosen are meant  to represent
 typical levels for  these  areas.  The maximum values represent maximum
 concentrations for  all  exposure routes.

     As can  be seen in Table 33, in areas  where drinking water  is
 chlorinated,  drinking water is  the predominant route of exposure.
 However, in  urban areas,  inhalation  represents an  Important additional
 exposure route.  Swimming in chlorinated pools may represent another
 important exposure  route.   What portion of the U.S.  population may  fall
 into each of  the five categories is  unknown,  and there are certainly
 other situations that are not  represented  in Table 33.  Nevertheless,
 this table is meant to  illustrate a range  of typical exposures.

     The maximum risks  of cancer were  estimated for the five exposure
 conditions by use of  the  multistage model  results shown in Table 32.

 5.   Other Trihalomethanes

     Table 34 summarizes  estimates for known routes of exposure  to  other
 trihalomethanes.  Risk estimates of  the type made  for  chloroform  cannot
 be  made for  the other trihalomethanes  since  information on exposure and
effects are lacking.  The typical levels of exposure are much lower  than
for chloroform; however,  exposure in a  few locations is high relative  to
chloroform.   Very little  information is available on the effects of bromo-
form,  dibromochloromethane,  and bromodichloromethane for man.  The only
basis  for comparison  among  the trihalomethanes  is  the oral LDso for mice,
which shows chloroform to be the least  toxic  (Bowman .at. al_.  1978).  Until
 che carcinogenesis bioassays for these  other chemicals are completed,  no
conclusive statements can be tnade about: risk to man.

                                   TABLE  33.  ESTIMATED  EXPOSURE OF HAN TO CHLOROFORM

                                                                                                 Probable Maximum
                              	Exposure in nig/day (Z of  total)	(Lifetime
          lix|osure                                                             Dermal   ~Total   Cancer risk
          Si illation           Drinking Water   	Food	 Inhalation  Contact     (max)   per capita)

          KuraJ area              0  (0) ]      0.0006 -  0.04  (100) .002 (0)   0  (0)        0.04   0.26 x 10~6
            no clilorlnation

          Uural area     _                                                                                  ,
            ,:l, I urination     0.02 -  0.13  (60)  0.0006 -  0.04  (18)  .002 (9)   0.03 (13)    0.2    1.3 x 10~

i-         Urban               .03 -  0.2 (67) 3 0.0006 -  0.04  (13)  0.02 (6)   O.OS'(14)    0.3    2.0 x 10~6
           Industrial *       0.03 - 0.2  (25)   0.0006 -  0.04  (5)    0.5 (60)   0.05 (6)     0.8    5.2 x 10~6

          Maximum            1.2  (46)          0.04  (2)             1 (38)     0.3 (12)     2.6    16.9  x  10~6

                                                                                        5                   fi  6
          Cl.il.l - maximum    0.5  (19)          0.04  (2)             1 (38)     1.1 (41)     2.6   47.3 x 10

           I'ercent of total (using maximum values) percentages  may not add to due to rounding.
           Assumes median level of chloroform In rural areas.
           ^Assumes chloroform concentration of O.lmg/1.                   ,
          ('Assumes concentration of 50 ug/m^ 24 hrs/day.
           Includes swimming.
                   a 25 kg child.
          Source:  See Chapter V.

                                    Exposure (nig/day)
                      Drinking Water   Contact             Total

                       Median  Max. Typical  Max.    Typical   Max.

Bromoform              0.007   0.6   0.001   0.8      0.008    1.4
Dibcomochloromethane   0.007   0.6   0.001   0.07     0.008    0.7

Bromodichloromethane   0.002   0.4   0.003   0.04     0.005    0.4
 This is for child swimming in saline pool 2 hrs/day.
Source:  Chapter V.


      The complete lack of effects data for bromodlchloromethane and
 dibromochloromethane renders a risk analysis for aquatic life impossible.
 Although monitoring and effects information on chloroform and bromoform
 are extremely limited, some tentative conclusions can be drawn.

      On a nationwide scale, no aquatic populations appear to be exposed
 consistently to harmful levels of either chloroform or bromoform.
 Anderson .et al. (1979) state that the acute toxicity levels found in
 their freshwater fish bioassays ."are orders of magnitude above the
 maximum level (24.7 mg/1) found in chlorinated water samples taken
 across the U.S."  However, ambient levels of chloroform between 100 mg/1
 and 1,000 mg/1 have been measured, and the occurrence of spills and dis-
 charges from industries may result in temporary and localized high
 concentrations that could be hazardous to aquatic life.  Unfortunately,
 monitoring data for all of the trihalomethanes are so scarce that it is
 not possible to identify any specific regions that have a potential for
.high levels of exposure to biota.

Anderson, D.R. et al.  Progress Report Covering Period January 1 through
March 31, 1979:  Biocide by-products in aquatic environments.  PNL-2988.
U.S. Nuclear Regulatory Commission; 1979.

Arthur D. Little, Inc.  An approach to exposure and risk assessments for
priority pollutants.  Draft.  Contract 68-01-3857.  Washington, DC:
Monitoring and Data Support Division, U.S. Environmental Protection
Agency; 1980.

Bowman, M.F.; Borzelleca, J.F.; Munson, A.E.  The toxicity of some halo-
methanes in mice.  Toxicol. Appl. Pharmacol. 44(1):213-216; 1978.

National Academy of Sciences (NAS).  Drinking water and health.
Washington, DC:  NAS; 1977.

National Academy of Sciences (NAS).  Chloroform, carbon tetrachloride,
and other halomethanes:  an environmental assessment.  Washington, DC:
NAS; 1978.  294 p.

Reitz, R.H.; Gehring, P.J.; Park, C.N.  Carcinogenic risk estimation for
chloroform - an alternative to EPA's procedures.  J. Food & Cosmet.
Toxicol. 16:511-514; 1978.

Reuber, M.D.  Carcinogenicity of chloroform.  Environ. Health Persp.
31:171-182; 1979.

Stokinger, H.E.  Toxicology and drinking water contaminants.  J. Amer.
Water Works Assoc. 69(7):399-402; 1977.

Tardiff, R.G.  Health effects of organics:  risk and hazard assessment
of ingested chloroform.  J. Amer. Water Works Assoc. 69(12):658-660;

U.S. Environmental Protection Agency (U.S. EPA).  Ambient water quality
criteria.  Criterion document  chloroform.  Washington, DC:  Criteria
and Standards Division, Office of Water Planning and Standards, U.S. EPA;
1979.  Available from:  NTIS, Springfield, VA; PB 292 427.