NATIONAL INTERIM
PRIMARY DRINKING
WATER REGULATIONS
For Use With
Homestudy Course 3014-G
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF WATER SUPPLY
CENTER FOR PROFESSIONAL DEVELOPMENT AND TRAINING
U S DEPARTMENT OF HEALTH AND HUMAN SERVICES
PU8LIC HEALTH SERVICE
Centers for Disease Control
-------
Preface
The National Interim Primary Drinking Water Regulations published
herein were promulgated on December 24, 1975, in accordance with the
provisions of the Safe Drinking Water Act I Public Law 93-523). Additional
Interim Primary Regulations for radioactivity in drinking water were pro-
mulgated on July 9, 1976. These regulations become effective on June 24,
1977, and become in essence the standards by which all public drinking
water supplies are judged.
These regulations will replace the Public Health Service Drinking Water
Standards of 1962.
The background material on which the various Maximum Contaminant
Levels were based, known as the Statement of Basis and Purpose, is included
herein as appendices. The Statement of Basis and Purpose also includes
background material on some contaminants which were omitted from the
Regulations and thus provides an explanation for those omissions.
Certain contaminants which were listed in the Public Health Service
Drinking Water Standards are not included in the National Interim Primary
Drinking Water Regulations because the contaminants are not directly re-
lated to the safety of drinking water but rather are related to the esthetic
quality. Such contaminants, and others, will be listed in Secondary Drinking
Water Regulations, to be published separately.
The National Interim Primary Drinking Water Regulations, including
any amendments or revisions which may be added later, should be useful in
evaluating the quality and safety of all water supplies generally.
Victor J. Kimm
Deputy Assistant Administrator for Water Supply
Environmental Protection Agency
-------
Table of Contents
Subpart A—General
Sec. Pnge
141.1 Applicability 1
141.2 Definitions . 1
141.3 Coverage 2
141.4 Variances and exemptions. 2
141.5 Siting requirements . 3
141.6 Effective date 3
Subpart B—Maximum Contaminant Levels
141.11 Maximum contaminant levels for inorganic chemicals 5
141.12 Maximum contaminant levels for organic chemicals 5
141.13 Maximum contaminant levels (or turbidity 6
141.14 Maximum microbiological contaminant levels 6
141.15 Maximum contaminant levels for radium 226, radium 228, and gross
alpha particle radioactivity in community water systems 7
141.16 Maximum contaminant levels for beta particle and photon radioactivity
from man-made radionuclides in community water systems 7
Subpart C—Monitoring and Analytical Requirements
141.21 Microbiological contaminant sampling and analytical requirements 9
141.22 Turbidity sampling and analytical requirements 12
141.23 Inorganic chemical sampling and analytical requirements 12
141.24 Organic chemical sampling and analytical requirements. 14
141.25 Analytical methods for radioactivity 15
141.26 Monitoring frequency for radioactivity in community water systems. 17
141.27 Alternative analytical techniques 19
141.28 Approved laboratories 20
141.29 Monitoring of consecutive public water systems 20
Subpart D—Reporting, Public Notification, and Record Keeping
141.31 Reporting requirements 21
141.32 Public notification of variances, exemptions, and non-compliance with
regulations 21
141.33 Record maintenance 22
Authority: Sees. 1412, 1414, 1445, and 1450 of the Public Health Service Act, 88
Stat. 1660 ( 42 U.S.C. 300g-l, 300g-3,300j-4, and 300j-9).
Appendix A—-Background Used in Developing the National Interim Primary
Drinking Water Regulation*
Source and Facilities 25
Microbiological Quality 27
Chemical Quality 47
Fluid Intake 48
Arsenic 51
Barium 58
Cadmium 59
Chromium 63
Cyanide 65
Fluoride 66
Lead 69
Mercury 76
Nitrate 81
Organic Chemicals 84
Pesticides 103
-------
Table of Contents (Continued) Pa8e
Selenium 113
Silver 118
Sodium 120
Sulfate 126
Appendix B—Radionuclides
Introduction 129
General Considerations 129
Health Risk from Radionuclides in Drinking Water 131
The Control of Radium in Public Water Systems 134
National CoBt for Radium Removal 135
Impact of Maximum Contaminant Levels for Man-made Radionuclides 137
Monitoring for Radioactivity in Community Water Systems 138
Monitoring Costs for Radium and Alpha Particle Activity 139
Monitoring Costs for Man-made Radioactivity — 141
Appendix I Policy Statement, "Relationship Between Radiation Dose
and Effect" 143
Appendix II Risk to Health from Internal Emitters 146
A. The Dose and Health Risk from Radium Ingestions 146
B. The Relative Health Risk of Radium-228 as Compared
to Radium-226 147
Appendix III Cost and Cost Effectiveness of Radium Removal 150
Appendix IV Dosimetric Calculations for Man-Made Radioactivity 152
A. Calculations Based on NBS Handbook 69 152
B. The Dose from Tritium and Strontium-90 in Drinking
Water 153
C. Average Annual Concentrations Yielding 4 Millirem Per
Year for Two Liter Daily Intake 155
Table TV-2A (Nuclides with ti/, > 24 h) 155
Table IV-2B (Nuclides with twa < 24 h) 157
-------
SUBPART A—GENERAL
Subpart A-General
Section 141.1 Applicability.
This part establishes primary drinking water regulations pursuant to
section 1412 of the Public Health Service Act, as amended by the Safe
Drinking Water Act (Pub. L. 93-523); and related regulations applicable to
public water systems.
Section 141.2 Definitions.
As used in this part, the term:
(a) "Act" means the Public Health Service Act, as amended by the Safe
Drinking Water Act, Pub. L. 93-523.
(b) "Contaminant" means any physical, chemical, biological, or radi-
ological substance or matter in water.
(c) "Maximum, contaminant level" means the maximum permissible level
of a contaminant in water which is delivered to the free flowing outlet of
the ultimate user of a public water system, except in the case of turbidity
where the maximum permissible level is measured at the point of entry to
the distribution system. Contaminants added to the water under circum-
stances controlled by the user, except those resulting from corrosion of pip-
ing and plumbing caused by water quality, are excluded from this definition.
(d) "Person" means an individual, corporation, company, association,
partnership, State, municipality, or Federal agency.
(e) "Public water system" means a system for the provision to the public
of piped water for human consumption, if such system has $t least fifteen
service connections or regularly serves an average of at least twenty-five
individuals daily at least 60 days out of the year. Such term includes (1)
any collection, treatment, storage, and distribution facilities under control
of the operator of such system and used primarily in connection with such
system, and (2) any collection or pretreatment storage facilities not under
such control which are used primarily in connection with such system. A
public water system is either a "community water system" or a "non-com-
munity water system."
(i) "Community water system" means a public water system which serves
at least 15 service connections used by year-round residents or regularly
serves at least 25 year-round residents.
(ii) "Non-community water system" means a public water system that is
not a community water system.
(f) "Sanitary survey" means an onsite review of the water source, facili*
ties, equipment, operation and maintenance of a public water system for the
purpose of evaluating the adequacy of such source, facilities, equipment,
operation and maintenance for producing and distributing safe drinking
water.
(g) "Standard sample" means the aliquot of finished drinking water that
is examined for the presence of coliform bacteria.
1
-------
DRINKING WATER REGULATIONS
(h) "State" means the agency of the State government which has juris
diction over public water systems. During any period when a State does not
have primary enforcement responsibility pursuant to Section 1413 of the
Act, the term "State" means the Regional Administrator, U.S. Environ-
mental Protection Agency.
(i) "Supplier of water" means any person who owns or operates a public
water system.
(j) "Dose equivalent" means the product of the absorbed dose from
ionizing radiation and such factors as account for differences in biological
effectiveness due to the type of radiation and its distribution in the body as
specified by the International Commission on Radiological Units and
Measurements (ICRU).
(k) "Rem" means the unit of dose equivalent from ionizing radiation to
the total body or any internal organ or organ system. A "millirem (mrem)"
is 1/1000 of a rem.
(1) "Picocurie (pCi)" means that quantity of radioactive material pro-
ducing 2.22 nuclear transformations per minute.
(m) "Gross alpha particle activity" means the total radioactivity due to
alpha particle emission as inferred from measurements on a dry sample.
(n) "Man-made beta particle and photon emitters" means all radio-
nuclides emitting beta particles and/or photons listed in Maximum Per-
missible Body Burdens and Maximum Permissible Concentration of Radio-
nuclides in Air or Water for Occupational Exposure, NBS Handbook
69, except the daughter products of thorium-232, uranium-235 and
uranium-238.
(o) "Gross beta particle activity" means the total radioactivity due to
beta particle emission as inferred from measurements on a dry sample.
Section 141.3 Coverage.
This part shall apply to each public water system, unless the pubilic water
system meets all of the following conditions:
(a) Consists only of distribution and storage facilities (and does not
have any collection and treatment facilities) ;
(b) Obtains all of its water from, but is not owned or operated by, a pub-
lic water system to which such regulations apply:
(c) Does not sell water to any person; and
(d) Is not a carrier which conveys passengers in interstate commerce.
Section 141.4 Variances and exemptions.
Variances or exemptions from certain provisions of these regulations
may be granted pursuant to Sections 1415 and 1416 of the Act by the entity
with primary enforcement responsibility. Provisions under Part 142,
National Interim Primary Drinking Water Regulations Implementation—
subpart E (Variances) and subpart F (Exemptions)—apply where EPA
has primary enforcement responsibility.
2
-------
SUBPART A—GENERAL LEVELS
Section 141.5 Siting requirements.
Before a person may enter into a financial commitment for or initiate
construction of a new public water system or increase the capacity of an
existing public water system, he shall notify the State and, to the extent
practicable, avoid locating part or all of the new or expanded facility at a
site which:
(a) Is subject to a significant risk from earthquakes, floods, fires or
other disasters which could cause a breakdown of the public water system or
a portion thereof; or
(b) Except for intake structures, is within the floodplain of a 100-year
flood or is lower than any recorded high tide where appropriate records
exist.
The U.S. Environmental Protection Agency will not seek to override land
use decisions affecting public water systems siting which are made at the
State or local government levels.
Section 141.6 Effective date.
The regulations set forth in this part shall lake effect 18 months after
the date of promulgation.
3
-------
SUBPART B—MAXIMUM CONTAMINANT LEVELS
Subpart It-Maximum Contaminant
Levels
Section 141.11 Maximum contaminant levels for inorganic
chemicals.
(a) The maximum contaminant level for nitrate is applicable to both
community water systems and non-community water systems. The levels for
the other inorganic chemicals apply only to community water systems. Com-
pliance with maximum contaminant levels for inorganic chemicals is cal-
culated pursuant to § 141.23.
(b) The following are the maximum contaminant levels for inorganic
chemicals other than fluoride:
Level,
milligrams
Contaminant per liter
Arsenic 0.05
It iit mm ... 11 ....¦> i i • ——• • 1.
Cadmium ........ ...... 0.010
Chromium 0.05
Mercury 0.002
Nitrate (as N) 10.
Selenium — 0.01
Silver —.— .............. 0.05
(c) When the annual average of the maximum daily air temperatures for
the location in which the community water system is situated is the follow-
ing, the maximum contaminant levels for fluoride are:
Temperature Level,
Degrees Degrees Celsius milligrams
Fahrenheit per liter
53.7 and below 12.0 and below 2.4
53.8 to 58-3 12.1 to 14.6 22
58.4 to 63.8 14.7 to 17.6 2.0
63.9 to 70.6 17.7 to 21.4 1.8
70.7 to 79.2 21.5 to 262 1.6
79.3 to 90.5 26.3 to 32.5 1.4
Section 141.12 Maximum contaminant level* for organic
chemicals.
The following are the maximum contaminant levels for organic chemicals.
They apply only to community water systems. Compliance with maximum
contaminant levels for organic chemicals is calculated pursuant to § 141.24.
Level,
milligrams
per liter
(a) Chlorinated hydrocarbons:
Endrin (1, 2, 3, 4, 10, 10-hexachloro-6,7-epoxy-l, 4, 0.0002
-------
DRINKING WATER REGULATIONS
4a, 5, 6, 7, 8, 8a-octahydro-l, 4-endo, endo-5, 8 - dimethano
naphthalene).
Lindane (1, 2, 3, 4, 5, 6-hexachlorocyclohexane, 0,004
gamma isomer).
Methoxychlor (1, 1, 1-Trichloroethane). 2, 2-bis 0.1
[p-methoxyphenyl].
Toxaphene {C j 0H j 0Clg-Technical chlorinated 0.005
camphene, 67-69 percent chlorine).
(b) Chlorophenoxys:
2,4-D, (2, 4-Dichlorophenoxyacetic acid). 0.1
2, 4, 5-TP Silvex (2, 4, 5-Trlchlorophenoxypropionic acid). 0.01
Section 141.13 Maximum contaminant levels for turbidity.
The maximum contaminant levels for turbidity are applicable to both
community water systems and non-community water systems using surface
water sources in whole or in part. The maximum contaminant levels for
turbidity in drinking water, measured at a representative entry point (s)
to the distribution system, are:
(a) One turbidity unit (TU), as determined by a monthly average pur-
suant to § 141.22, except that five or fewer turbidity units may be allowed
if the supplier of water can demonstrate to the State that the higher tur-
bidity does not do any of the following:
(1) Interfere with disinfection;
(2) Prevent maintenance of an effective disinfectant agent throughout
the distribution system; or
(3) Interfere with microbiological determinations.
(b) Five turbidity units based on an average for two consecutive days
pursuant to § 141.22.
Section 141.14 Maximum microbiological contaminant levels.
The maximum contaminant levels for coliform bacteria, applicable to
community water systems and non-community water systems, are as follows:
(a) When the membrane filter technique pursuant to § 141.21(a) is
used, the number of coliform bacteria shall not exceed any of the following:
(1) One per 100 milliliters as the arithmetic mean of all samples ex-
amined per month pursuant to § 141.21 (b) or (c);
(2) Four per 100 milliliters in more than one sample when less than 20
are examined per month; or
(3) Four per 100 milliliters in more than five percent of the samples
when 20 or more are examined per month.
(b) (1) When the fermentation tube method and 10 milliliter standard
portions pursuant to § 141.21(a) are used, coliform bacteria shall not be
present in any of the following:
(i) more than 10 percent of the portions in any month pursuant to
§ 141.21 (b) or (c);
(il) three or more portions in more than one sample when less than 20
6
-------
SUBPART B—MAXIMUM CONTAMINANT LEVELS
samples are examined per month; or
(iii) three or more portions in more than five percent of the samples
when 20 or more samples are examined per month.
(2) When the fermentation tube method and 100 milliliter standard por-
tions pursuant to § 141.21(a) are used, coliform bacteria shall not be
present in any of the following:
(i) more than 60 percent of the portions in any month pursuant to
§ 141.21 (b) or (c);
(ii) five portions in more than one sample when less than five samples
are examined per month; or
(iii) five portions in more than 20 percent of the samples when five or
more samples are examined per month.
(c) For community or non-community systems that are required to
sample at a rate of less than 4 per month, compliance with paragraphs (a),
(b) (1), or (b) (2) of this section shall be based upon sampling during a
3 month period, except that, at the discretion of the State, compliance may
be based upon sampling during a one-month period.
Section 141.15 Maximum contaminant levels for radium-226,
radium-228, and gross alpha particle radioactivity in com-
munity water systems.
The following are the maximum contaminant levels for radium-226,
radium-228, and gross alpha particle radioactivity:
(a) Combined radium-226 and radium-228—5 pCi/1.
(b) Gross alpha particle activity (including radium-226 but excluding
radon and uranium)—15 pCi/1.
Section 141.16 Maximum contaminant levels for beta particle
and photon radioactivity from man-made radionuclides in
community water systems
(a) The average annual concentration of beta particle and photon radio-
activity from man-made radionuclides in drinking water shall not produce
an annual dose equivalent to the total body or any internal organ greater
than 4 millirem/year.
(b) Except for the radionuclides listed in Table A, the concentration of
man-made radionuclides causing 4 mrem total body or organ dose equiv-
alents shall be calculated on the basis of a 2 liter per day drinking water
intake using the 168 hour data listed in **Maximum Permissible Body Bur-
dens and Maximum Permissible Concentration of Radionuclides in Air or
Water for Occupational Exposure " NBS Handbook 69 as amended August
1963, U.S. Department of Commerce. If two or more radionuclides are
present, the sum of their annual dose equivalent to the total body or to any
organ shall not exceed 4 millirem/year.
7
-------
DRINKING WATER REGULATIONS
Table A.—Average annual concentrations assumed to produce a total body or organ
dose of 4 mrem/yr
Radionuclide
Critical organ
pa
per liter
Tritium —
Strontium-90
_ . . Bone marrow
20,000
8
a
-------
SUBPART C—MONITORING AND ANALYTICAL REQUIREMENTS
Subpart C-Monitoring and Analytical
Requirements
Section 141.21 Microbiological contaminant sampling and
analytical requirements.
(a) Suppliers of water for community water systems and non-community
water systems shall analyze for coliform bacteria for the purpose of deter-
mining compliance with ^ 141.14. Analyses shall be conducted in accord-
ance with the analytical recommendations set forth in "Standard Methods for
the Examination of Water and Wastewater," American Public Health Asso-
ciation, 13th Edition, pp. 662-688, except that a standard sample size shall
be employed. The standard sample used in the membrane filter procedure
shall be 100 milliliters. The standard sample used in the 5 tube most prob-
able number (MPN) procedure (fermentation tube method) shall be 5
times the standard portion. The standard portion is either 10 milliliters or
100 milliliters as described in § 141.14 (b) and (c). The samples shall be
taken at points which are representative of the conditions within the dis-
tribution system.
(b) The supplier of water for a community water system shall take coli-
form density samples at regular time intervals, and in number proportionate
to the population served by the system. In no event shall the frequency be
less than as set forth below:
Minimum number of
Population served: samples per month
25 to 1,000 1
I,001 to 2,500 2
2,501 to 3,300 3
3,301 to 4,100 4
4,101 to 4,900 5
4,901 to 5,800 6
5,801 to 6,700 7
6,701 to 7,600 8
7,601 to 8,500 9
8,501 to 9,400 10
9,401 to 10,300 11
10,301 to 11,100 12
II,101 to 12,000 13
12,001 to 12,900 14
12,901 to 13,700 15
13,701 to 14,600 16
14,601 to 15,500 17
15,501 to 16,300 18
16,301 to 17,200 19
17,201 to 18,100 20
18,101 to 18,900 . 21
18,901 to 19,800 22
19,801 to 20,700 23
Minimum number of
Population served: samples per month
90,001 to 96,000 95
96,001 to 111,000 100
111,001 to 130,000 110
130,001 to 160,000 120
160,001 to 190,000 130
190,001 to 220,000 140
220,001 to 250,000 150
250,001 to 290,000 160
290.000 to 320,000 170
320.001 to 360,000 180
360,001 to 410,000 190
410,001 to 450,000 200
450,001 to 500,000 210
500,001 to 550,000 220
550,001 to 600,000 230
600,001 to 660,000 240
660,001 to 720,000 250
720,001 to 780,000 260
780,001 to 840,000 270
840,001 to 910,000 280
910,001 to 970,000 290
970,001 to 1,050,000 300
1,050,001 to 1,140,000 310
9
-------
DRINKING WATER REGULATIONS
20,701 to 21,500 24 1,140,001 to 1,230,000 320
21,501 to 22,300 25 1,230,001 to 1,320,000 330
22,301 to 23,200 26 1,320,001 to 1,420,000 340
23,201 to 24,000 27 1,420,001 to 1,520,000 350
24,001 to 24,900 28 1,520,001 to 1,630,000 360
24,901 to 25,000 ..... 29 1,630,001 to 1,730,000 370
25,001 to 28,000 30 1,730,001 to 1,850,000 380
28,001 to 33,000 35 1,850,001 to 1,970,000 390
33,001 to 37,000 40 1,970,001 to 2,060,000 400
37,001 to 41,000 45 2,060,001 to 2,270,000 410
41,001 to 46,000 50 2,270,001 to 2,510,000 420
46,001 to 50,000 55 2,510,001 to 2,750,000 430
50,001 to 54,000 60 2,750,001 to 3,020,000 440
54,001 to 59,000 65 3,020,001 to 3,320,000 450
59,001 to 64,000 70 3,320,001 to 3,620,000 460
64,001 to 70,000 75 3,620,001 to 3,960,000 470
70,001 to 76,000 80 3,960,001 to 4,310,000 480
76,001 to 83,000 85 4310,001 to 4,690,000 490
83,001 to 90,000 90 4,690,001 or more 500
Based on a history of no coliform bacterial contamination and on a sanitary
survey by the State showing the water system to be supplied solely by a
protected ground water source and free of sanitary defects, a community
water system serving 25 to 1,000 persons, with written permission from the
State, may reduce this sampling frequency except that in no case shall it be
reduced to less than one per quarter.
(c) The supplier of water for a non-community water system shall sample
for coliform bacteria in each calendar quarter during which the system pro-
vides water to the public. Such sampling shall begin within two years after
the effective date of this part. If the State, on the basis of a sanitary survey,
determines that some other frequency is more appropriate, that frequency
shall be the frequency required under these regulations. Such frequency
shall be confirmed or changed on the basis of subsequent surveys.
(d) (1) When the coliform bacteria in a single sample exceed four per
100 milliliters (§ 141.14(a)), at least two consecutive daily check samples
shall be collected and examined from the same sampling point. Additional
check samples shall be collected daily, or at a frequency established by the
State, until the results obtained from at least two consecutive check samples
show less than one coliform bacterium per 100 milliliters.
(2) When coliform bacteria occur in three or more 10 ml portions of a
single sample (§ 141.14(b) (1)), at least two consecutive daily check
samples shall be collected and examined from the same sampling point.
Additional check samples shall be collected daily, or at a frequency estab-
lished by the State, until the results obtained from at least two consecutive
check samples show no positive tubes.
(3) When coliform bacteria occur in all five of the 100 ml portions of
a single sample (§ 141.14(b) (2)), at least two daily check samples shall
be collected and examined from the same sampling point. Additional check
10
-------
SUBPART C—MONITORING AND ANALYTICAL REQUIREMENTS
samples shall be collected daily, or at a frequency established by the State,
until the results obtained from at least two consecutive check samples show
no positive tubes.
(4) The location at which the check samples were taken pursuant to
paragraphs (d) (I), (2), or (3) of this section shall not be eliminated from
future sampling without approval of the State. The results from all coliform
bacterial analyses performed pursuant to this subpart, except those obtained
from check samples and special purpose samples, shall be used to determine
compliance with the maximum contaminant level for coliform bacteria as
established in § 141.14. Check samples shall not be included in calculating
the total number of samples taken each month to determine compliance with
§ 141.21 (b) or (c).
(e) When the presence of coliform bacteria in water taken from a par-
ticular sampling point has been confirmed by any check samples examined
as directed in paragraphs (d) (1), (2), or (3) of this section, the supplier
of water shall report to the State within 48 hours.
(f) When a maximum contaminant level set forth in paragraphs (a), (b)
or (c) of § 141.14 is exceeded, the supplier of water shall report to the State
and notify the public as prescribed in § 141.31 and § 141.32.
(g) Special purpose samples, such as those taken to determine whether
disinfection practices following pipe placement, replacement, or repair have
been sufficient, shall not be used to determine compliance with § 141.14 or
§ 141.21 (b) or (c).
(h) A supplier of water of a community water system or a non-com-
munity water system may, with the approval of the State and based upon a
sanitary survey, substitute the use of chlorine residual monitoring for not
more than 75 percent of the samples required to be taken by paragraph (b)
of this section, Provided, That the supplier of water takes chlorine residual
samples at points which are representative of the conditions within the dis-
tribution system at the frequency of at least four for each substituted micro*
biological sample. There shall be at Jeaatdaily determinations of chlorine
residual. When the supplier of water exercises the option provided in this
paragraph (h) of this section, he shall maintain no less than 0.2 mg/1 free
chlorine throughout the public water distribution system. When a particular
sampling point has been shown to have a free chlorine residual less than
0.2 mg/1, the water at that location shall be retested as soon as practicable
and in any event within one hour. If the original analysis is confirmed, this
fact shall be reported to the State within 48 hours. Also, if the analysis is
confirmed, a sample for coliform bacterial analysis must be collected from
that sampling point as soon as practicable and preferably within one hour,
and the results of such analysis reported to the State within 48 hours after
the results are known to the supplier of water. Analyses for residual chlorine
shall be made in accordance with "Standard Methods for the Examination
of Water and Wastewater," 13th Ed., pp. 129-132. Compliance with the
11
-------
DRINKING WATER REGULATIONS
maximum contaminant levels for coliform bacteria shall be determined on
the monthly mean or quarterly mean basis specified in § 141.14, including
those samples taken as a result of failure to maintain the required chlorine
residual level. The State may withdraw its approval of the use of chlorine
residual substitution at any time.
Section 141.22 Turbidity sampling and analytical
requirements.
(a) Samples shall be taken by suppliers of water for both community
water systems and non-community water systems at a representative entry
point(s) to the water distribution system at least once per day, for the pur-
pose of making turbidity measurements to determine compliance with
§ 141.13. The measurement shall be made by the Nephelometric Method in
accordance with the recommendations set forth in "Standard Methods for
the Examination of Water and Wastewater," American Public Health Asso-
ciation, 13th Edition, pp. 350-353, or "Methods for Chemical Analysis of
Water and Wastes," pp. 295-298, Environmental Protection Agency, Office
of Technology Transfer, Washington, D.C. 20460, 1974.
(b) If the result of a turbidity analysis indicates that the maximum al-
lowable limit has been exceeded, the sampling and measurement shall be
confirmed by resampling as soon as practicable and preferably within one
hour. If the repeat sample confirms that the maximum allowable limit has
been exceeded, the supplier of water shall report to the State within 48
hours. The repeat sample shall be the sample used for the purpose of cal-
culating the monthly average. If the monthly average of the daily samples
exceeds the maximum allowable limit, or if the average of two samples taken
on consecutive days exceeds 5 TU, the supplier of water shall report to the
State and notify the public as directed in § 141.31 and § 141.32.
(c) Sampling for non-community water systems shall begin within two
years after the effective date of this part.
(d) The requirements of this § 141.22 shall apply only to public water
systems which use water obtained in whole or in part from surface sources.
Section 141.23 Inorganic chemical sampling and analytical
requirements.
(a) Analyses for the purpose of determining compliance with | 141.11
are required as follows:
(1) Analyses for all community Water systems utilizing surface water
sources shall be completed within one year following the effective date of
this part. These analyses shall be repeated at yearly intervals.
(2) Analyses for all community water systems utilizing only ground
water sources shall be completed within two years following the effective
date of this part. These analyses shall be repeated at three-year intervals.
(3) For non-community water systems, whether supplied by surface or
ground water sources, analyses for nitrate shall be completed within two
12
-------
SUBPART C—MONITORING AND ANALYTICAL REQUIREMENTS
years following the effective date of this part. These analyses shall be re-
peated at intervals determined by the State.
(b) If the result of an analysis made pursuant to paragraph (a) indicates
that the level of any contaminant listed in Jj 141.11 exceeds the maximum
contaminant level, the supplier of water shall report to the State within 7
days and initiate three additional analyses at the same sampling point within
one month.
(c) When the average of four analyses made pursuant to paragraph (b)
of this section, rounded to the same number of significant figures as the
maximum contaminant level for the substance in question, exceeds the max-
imum contaminant level, the supplier of water shall notify the State pur-
suant to § 141.31 and give notice to the public pursuant to § 141.32.
Monitoring after public notification shall be at a frequency designated by
the State and shall continue until the maximum contaminant level has not
been exceeded in two successive samples or until a monitoring schedule as
a condition to a variance, exemption or enforcement action shall become
effective.
(d) The provisions of paragraphs (b) and (c) of this section notwith-
standing, compliance with the maximum contaminant level for nitrate shall
be determined on the basis of the mean of two analyses. When a level ex-
ceeding the maximum contaminant level for nitrate is found, a second
analysis shall be initiated within 24 hours, and if the mean of the two an-
alyses exceeds the maximum contaminant level, the supplier of water shall
report his findings to the State pursuant to § 141.31 and shall notify the
public pursuant to § 141.32.
(e) For the initial analyses required by paragraph (a) (1), (2) or (3)
of this section, data for surface waters acquired within one year prior to
the effective date and data for ground waters acquired within 3 years prior
to the effective date of this part may be substituted at the discretion of the
State.
(f) Analyses conducted to determine compliance with § 141.11 shall
be made in accordance with the following methods:
(1) Arsenic—Atomic Absorption Method, "Methods for Chemical An-
alysis of Water and Wastes," pp. 95-96, Environmental Protection Agency,
Office of Technology Transfer, Washington, D.C. 20460, 1974.
(2) Barium—Atomic Absorption Method, "Standard Methods for the
Examination of Water and Wastewater," 13th Edition, pp. 210-215, or
"Methods for Chemical Analysis of Water and Wastes," pp. 97-98, Environ-
mental Protection Agency, Office of Technology Transfer, Washington,
D.C. 20460,1974.
(3) Cadmium—Atomic Absorption Method, "Standard Methods for the
Examination of Water and Wastewater," 13th Edition, p.p. 210-215, or
"Methods for Chemical Analysis of Water and Wastes," pp. 101-103, En-
13
-------
DRINKING WATER REGULATIONS
vironmental Protection Agency, Office of Technology TransfA", Washington,
D.C. 20460, 1974.
(4) Chromium—Atomic Absorption Method, "Standard Methods for the
Examination of Water and Wastewater," 13th Edition, pp. 210-215, or
"Methods for Chemical Analysis of Water and Wastes," pp. 105-106, En-
vironmental Protection Agency, Office of Technology Transfer, Washing-
ton, D.C. 20460, 1974.
(5) Lead—Atomic Absorption Method, "Standard Methods for the Ex-
amination of Water and Wastewater," 13th Edition, pp. 210-215, or
"Methods for Chemical Analysis of Water and Wastes," pp. 112-113, En-
vironmental protection Agency, Office of Technology Transfer, Washing-
ton, D.C. 20460, 1974.
(6) Mercury—Flameless Atomic Absorption Method, "Methods for
Chemical Analysis of Water and Wastes," pp. 113-126, Environmental Pro-
tection Agency, Office of Technology Transfer, Washington, D.C. 20460,
1974.
(7) Nitrate—Brucine Colorimetric Method, "Standard Methods for the
Examination of Water and Wastewater," 13th Edition, pp. 461-464, or
Cadmium Reduction Method, "Methods for Chemical Analysis of Water and
Wastes," pp. 201-206, Environmental Protection Agency, Office of Tech-
nology Transfer, Washington, D.C. 20460, 1974.
(8) Selenium—Atomic Absorption Method, "Methods for Chemical
Analysis of Water and Wastes," p, 145, Environmental Protection Agency,
Office of Technology Transfer, Washington, D.C, 20460, 1974.
(9) Silver—Atomic Absorption Method, "Standard Methods for the Ex-
amination of Water and Wastewater", 13th Edition, pp. 210-215, or
"Methods for Chemical Analysis of Water and Wastes", p. 146, Environ-
mental Protection Agency, Office of Technology Transfer, Washington,
D.C. 20460,1974.
(10) Fluoride—Electrode Method, "Standard Methods for the Examin-
ation of Water and Wastewater", 13th Edition, pp. 172-174, or "Methods for
Chemical Analysis of Water and Wastes," pp. 65-67, Environmental Pro-
tection Agency, Office of Technology Transfer, Washington, D.C. 20460,
1974, or Colorimetric Method with Preliminary Distillation, "Standard
Methods for the Examination of Water and Wastewater," 13th Edition,
pp. 171-172 and 174-176, or "Methods for Chemical Analysis of Water and
Wastes," pp. 59-60, Environmental Protection Agency, Office of Technology
Transfer, Washington, D.C. 20460,1974.
Section 141.24 Organic chemical sampling and analytical
requirements.
(a) An analysis of substances for the purpose of determining compliance
with § 141.12 shall be made as follows:
f 1) For all community water systems utilizing surface water sources, an-
14
-------
SUBPART C—MONITORING AND ANALYTICAL REQUIREMENTS
alyses shall be completed within one year following the effective date of this
part. Samples analyzed shall be collected during the period of the year
designated by the State as the period when contamination by pesticides is
most likely to occur. These analyses shall be repeated at intervals specified
by the State but in no event less frequently than at three year intervals.
(2) For community water systems utilizing only ground water sources,
analyses shall be completed by those systems specified by the State.
(b) If the result of an analysis made pursuant to paragraph (a) of this
section indicates that the level of any contaminant listed in § 141.12 exceeds
the maximum contaminant level, the supplier of water shall report to the
State within 7 days and initiate three additional analyses within one month.
(c) When the average of four analyses made pursuant to paragraph (b)
of this section, rounded to the same number of significant figures as the
maximum contaminant level for the substance in question, exceeds the max-
imum contaminant level, the supplier of water shall report to the State pur-
suant to § 141.31 and give notice to the public pursuant to § 141.32, Mon-
itoring after public notification shall be at a frequency designated by the
State and shall continue until the maximum contaminant level has not been
exceeded in two successive samples or until a monitoring schedule as a con-
dition to a variance, exemption or enforcement action shall become effective.
(d) For the initial analysis required by paragraph la) 11) and (2) of
this section, data for surface water acquired within one year prior to the
effective date of this part and data for ground water acquired within three
years prior to the effective date of this part may be substituted at the dis-
cretion of the State.
(e) Analyses made to determine compliance with § 141.12(a) shall be
made in accordance with "Method for Organochlorine Pesticides in In-
dustrial Effluents," MDQARL, Environmental Protection Agency, Cincin-
nati, Ohio, November 28, 1973.
ff) Analyses made to determine compliance with § 141.12(b) shall be
conducted in accordance with "Methods for Chlorinated Phenoxy Acid Her-
bicides in Industrial Effluents," MDQARL, Environmental Protection
Agency, Cincinnati, Ohio, November 28, 1973.
Section 141.25 Analytical Methods for Radioactivity.
(a) The methods specified in Interim Radiochemical Methodology for
Drinking Water, Environmental Monitoring and Support Laboratory, EPA-
600/4-75*008, US EPA, Cincinnati, Ohio 45268, or those listed below, are
to be used to determine compliance with § § 141.15 and 141.16 (radio-
activity) except in cases where alternative methods have been approved in
accordance with § 141.27.
(1) Gross Alpha and Beta—Method 302 "Gross Alpha and Beta Radio-
activity in Water" Standard Methods for the Examination of Water and
Wastewater, 13th Edition, American Public Health Association, New York,
15
-------
DRINKING WATER REGULATIONS
N.Y., 1971.
(2) Total Radium—Method 304 "Radium in Water by Precipitation"
Ibid.
(3) Radium-226—Method 305 "Radium-226 by Radon in Water" Ibid.
(4) Strontium-89, 90—Method 303 "Total Strontium and Strontium-90
in Water" Ibid.
(5) Tritium—Method 306 "Tritium in Water" Ibid.
(6) Cesium-134—ASTM D-2459 "Gamma Spectrometry in Water," 1975
Annual Book of ASTM Standards, Water and Atmospheric Analysis, Part 31,
American Society for Testing and Materials, Philadelphia, PA. (1975).
(7) Uranium—ASTM D-2907 "Microquantities of Uranium in Water
by Fluorometry," Ibid.
(b) When the identification and measurement of radionuclides other
than those listed in paragraph (a) is required, the following references are
to be used, except in cases where alternative methods have been approved
in accordance with § 141.27.
(1) Procedures for Radiochemical Analysis of Nuclear Reactor Aqueous
Solutions, H. L. Krieger and S. Gold, EPA-R4-73-014. USEPA, Cincinnati,
Ohio, May 1973.
(2) HASL Procedure Manual, Edited by John H. Harley. HASL 300,
ERDA Health and Safety Laboratory, New York, N.Y., 1973.
(c) For the purpose of monitoring radioactivity concentrations in drink-
ing water, the required sensitivity of the radioanalysis is defined in terms of
a detection limit. The detection limit shall be that, concentration which can
be counted with a precision of plus or minus 100 percent at the 95 percent
confidence level (1.96a where 9 is the standard deviation of the net count-
ing rate of the sample).
(1) To determine compliance with § 141.15(a) the detection limit shall
not exceed 1 pCi/1. To determine compliance with § 141.15(b) the detection
limits shall not exceed 3 -pCi/1.
(2) To determine compliance with § 141.16 the detection limits shall not
exceed the concentrations listed in Table B.
Table B.—Detection Limits for Man-made Beta Particle and Photon Emitters
Radionuclide Detection limit
Tritium 1.000 pCi/1.
Strontium-89 10 pCi/I.
Strontium-90 2 pCi/1.
Iodine-131 1 pCi/1.
Cesium-134 10 pCi/1.
Gross beta 4 pCi/1.
Other radionuclides 1/10 of the applicable
limit.
(d) To judge compliance with the maximum contaminant levels listed in
sections 141.15 and 141.16, averages of data shall be used and shall be
16
-------
SUBPART C—MONITORING AND ANALYTICAL REQUIREMENTS
rounded to the same number of significant figures as the maximum contam-
inant level for the substance in question.
Section 141.26 Monitoring Frequency for Radioactivity in Com*
munity water systems.
(a) Monitoring requirements for gross alpha particle activity, radiura-
226 and radium-228.
(1) Initial sampling to determine compliance with § 141.15 shall begin
within two years of the effective date of these regulations and the analysis
shall be completed within three years of the effective date of these regula-
tions. Compliance shall be based on the analysis of an annual composite of
four consecutive quarterly samples or the average of the analyses of four
samples obtained at quarterly intervals.
(1) A gross alpha particle activity measurement may be substituted for
the required radium-226 and radium-228 analysis Provided, That the
measured gross alpha particle activity does not exceed 5 pCi/1 at a con-
fidence level of 95 percent (1.65cr where a is the standard deviation of the
net counting rate of the sample). In localities where radium-228 may be
present in drinking water, it is recommended that the State require radium-
226 and/or radium-228 analyses when the gross alpha particle activity ex-
ceeds 2 pCi/1.
(ii) When the gross alpha particle activity exceeds 5 pCI/1, the same or
an equivalent sample shall be analyzed for radium-226. If the concentration
of radium-226 exceeds 3 pCi/1 the same or an equivalent sample shall be
analyzed for radium-228.
(2) For the initial analysis required by paragraph fa) fl), data ac-
quired within one year prior to the effective date of this part may be sub-
stituted at the discretion of the State.
(3) Suppliers of water shall monitor at least once every four years fol-
lowing the procedure required by paragraph (a) (1). At the discretion of
the State, when an annual record taken in conformance with paragraph (a)
(1) has established that the average annual concentration is less than half
the maximum contaminant levels established by § 141.15, analysis of a
single sample may be substituted for the quarterly sampling procedure re-
quired by paragraph fa) (1).
(i) More frequent monitoring shall be conducted when ordered by the
State in the vicinity of mining or other operations which may contribute
alpha particle radioactivity to either surface or ground water sources of
drinking water.
fii) A supplier of water shall monitor in conformance with paragraph
fa) (1) within one year of the introduction of a new water source for a
community water system. More frequent monitoring shall be conducted
when ordered by the State in the event of possible contamination or when
changes in the distribution system or treatment processing occur which may
increase the concentration of radioactivity in finished water.
17
-------
DRINKING WATER REGULATIONS
(iii) A community water system using two or more sources having dif-
ferent concentrations of radioactivity shall monitor source water, in addi-
tion to water from a free-flowing tap, when ordered by the State.
(iv) Monitoring for compliance with § 141.15 after the initial period
need not include radium-228 except when required by the State, Provided,,
That the average annual concentration of radium-228 has been assayed at
least once using the quarterly sampling procedure required by paragraph
(a) fl).
(v) Suppliers of water shall conduct annual monitoring of any com-
munity water system in which the radium-226 concentration exceeds 3 pCi/1,
when ordered by the State.
(4) If the average annual maximum contaminant level for gross alpha
particle activity or total radium as set forth in § 141.15 is exceeded, the
supplier of a community water system shall give notice to the State pursuant
to § 141.31 and notify the public as required by § 141.32. Monitoring at
quarterly intervals shall be continued until the annual average concentra-
tion no longer exceeds the maximum contaminant level or until a monitoring
schedule as a condition to a variance, exemption or enforcement action shall
become effective.
(b) Monitoring requirements for man-made radioactivity in community
water systems.
(1) Within two years of the effective date of this part, systems using sur-
face water sources and serving more than 100,000 persons and such other
community water systems as are designated by the State shall be monitored
for compliance with § 141.16 by analysis of a composite of four consecutive
quarterly samples or analysis of four quarterly samples. Compliance with
§ 141.16 may be assumed without further analysis if the average annual
concentration of gross beta particle activity is less than 50 pCi/1 and if the
average annual concentrations of tritium and strontium-90 are less than
those listed in Table A, Provided, That if both radionuclides are present the
sum of their annual dose equivalents to bone marrow shall not exceed 4
millirem/year.
(1) If the gross beta particle activity exceeds 50 pCi/1, an analysis of the
sample must be performed to identify the major radioactive constituents
present and the appropriate organ and total body doses shall be calculated
to determine compliance with § 141.16.
(ii) Suppliers of water shall conduct additional monitoring, as ordered
by the State, to determine the concentration of man-made radioactivity in
principal watersheds designated by the State.
(iii) At the discretion of the State, supplies of water utilizing only
ground waters may be required to monitor for man-made radioactivity.
(2) For the initial analysis required by paragraph (b) fl) data acquired
within one year prior to the effective date of this part may be substituted at
the discretion of the State.
18
-------
SUBPART C—MONITORING AND ANALYTICAL REQUIREMENTS
(3) After the initial analysis required by paragraph (b) (1) suppliers
of water shall monitor at least every four years following the procedure
given in paragraph (b) (1).
(4) Within two years of the effective date of these regulations the sup-
plier of any community water system designated by the State as utilizing
waters contaminated by effluents from nuclear facilities shall initiate quar-
terly monitoring for gross beta particle and iodine-131 radioactivity and an-
nual monitoring for strontium-90 and tritium.
(i) Quarterly monitoring for gross beta particle activity shall be based on
the analysis of monthly samples or the analysis of a composite of three
monthly samples. The former is recommended. If the gross beta particle ac-
tivity in a sample exceeds 15 pCi/1, the same or an equivalent sample shall
be analyzed for strontium-39 and cesium-134. If the gross beta particle ac-
tivity exceeds 50 pCi/1, an analysis of the sample must be performed to
identify the major radioactive constituents present and the appropriate
organ and total body doses shall be calculated to determine compliance with
§ 141.16.
lii) For iodine-131, a composite of five consecutive daily samples shall
be analyzed once each quarter. As ordered by the State, more frequent mon-
itoring shall be conducted when iodine-131 is identified in the finished
water.
(iii ) Annual monitoring for strontium-90 and tritium shall be conducted
by means of the analysis of a composite of four consecutive quarterly samples
or analysis of four quarterly samples. The latter procedure is recommended.
(iv) The State may allow the substitution of environmental surveillance
data taken in conjunction with a nuclear facility for direct monitoring of
man-made radioactivity by the supplier of water where the State determines
such data is applicable to a particular community water system.
(5) If the average annual maximum contaminant level for man-made
radioactivity set forth in § 141.16 is exceeded, the operator of a community
water system shall give notice to the State pursuant to § 141.31 and to the
public as required by § 141.32. Monitoring at monthly intervals shall be
continued until the concentration no longer exceeds the maximum contam-
inant level or until a monitoring schedule as a condition to a variance, ex-
emption or enforcement action shall become effective.
Section 141.27 Alternative analytical techniques.
With the written permission of the State, concurred in by the Administra-
tor of the U.S. Environmental Protection Agency, an alternative analytical
technique may be employed. An alternative technique shall be acceptable
only if it is substantially equivalent to the prescribed test in both precision
and accuracy as it relates to the determination of compliance with any max-
imum contaminant level. The use of the alternative analytical technique shall
not decrease the frequency of monitoring required by this part.
19
-------
DRINKING WATER REGULATIONS
Section 141.28 Approved laboratories.
For the purpose of determining compliance with S 141.21 through
§ 141.27, samples may be considered only if they have been analyzed by a
laboratory approved by the State except that measurements for turbidity and
free chlorine residual may be performed by any person acceptable to the
State.
Section 141.29 Monitoring of consecutive public water systems.
When a public water system supplies water to one or more other public
water systems, the State may modify the monitoring requirements imposed
by this part to the extent that the interconnection of the systems justifies
treating them as a single system for monitoring purposes. Any modified
monitoring shall be conducted pursuant to a schedule specified by the State
and concurred in by the Administrator of the U.S. Environmental Pro-
tection Agency.
20
-------
SUBPART I>—REPORTING, PUBLIC NOTIFICATION, AND RECORD KEEPING
Subpart D-Reporting, Public
Notification, and Record Keeping
Section 141.31 Reporting requirements.
(a) Except where a shorter reporting period is specified in this part, the
supplier of water shall report to the State within 40 days following a test,
measurement or analysis required to be made by this part, the results of
that test, measurement or analysis.
(b) The supplier of water shall report to the State within 48 hours the
failure to comply with any primary drinking water regulation (including
failure to comply with monitoring requirements) set forth in this part.
(c) The supplier of water is not required to report analytical results to
the State in cases where a State laboratory performs the analysis and re-
ports the results to the State office which would normally receive such noti-
fication from the supplier.
Section 141.32 Public notification.
(a) If a community water system fails to comply with an applicable max-
imum contaminant level established in Subpart 8, fails to comply with an
applicable testing procedure established in Subpart C of this part, is
granted a variance or an exemption from an applicable maximum contam-
inant level, fails to comply with the requirements of any schedule prescribed
pursuant to a variance or exemption, or fails to perform any monitoring re-
quired pursuant to Section 1445 (a) of the Act, the supplier of water shall
notify persons served by the system of the failure or grant by inclusion of a
notice in the first set of water bills of the system issued after the failure or
grant and in any event by written notice within three months. Such notice
shall be repeated at least once every three months so long as the system's
failure continues or the variance or exemption remains in effect. If the sys-
tem issues water bills less frequently than quarterly, or does not issue water
bills, the notice shall be made by or supplemented by another form of direct
mail.
(b) If a community water system has failed to comply with an applicable
maximum contaminant level, the supplier of water shall notify the public of
such failure, in addition to the notification required by paragraph (a) of
this section, as follows:
(1) By publication on not less than three consecutive days in a news-
paper or newspapers of general circulation in the area served by the system.
Such notice shall be completed within fourteen days after the supplier of
water learns of the failure.
(2) By furnishing a copy of the notice to the radio and television sta-
tions serving the area served by the system. Such notice shall be furnished
within seven days after the supplier of water learns of the failure.
(c) If the area served by a community water system is not served by a
daily newspaper of general circulation, notification by newspaper required
21
-------
DRINKING WATER REGULATIONS
by paragraph (b) of this section shall instead be given by publication on
three consecutive weeks in a weekly newspaper of general circulation serving
the area. If no weekly or daily newspaper of general circulation serves the
area, notice shall be given by posting the notice in post offices within the
area served by the system.
(d) If a non-community water system fails to comply with an applicable
maximum contaminant level established in Subpart B of this part, fails to
comply with an applicable testing procedure established in Subpart C of this
part, is granted a variance or an exemption from an applicable maximum
contaminant level, fails to comply with the requirement of any schedule pre-
scribed pursuant to a variance or exemption or fails to perform any mon-
itoring required pursuant to Section 1445(a) of the Act, the supplier of
water shall give notice of such failure or grant to the persons served by the
system. The form and manner of such notice shall be prescribed by the
State, and shall insure that the public using the system is adequately in-
formed of the failure or grant.
(e) Notices given pursuant to this section shall be written in a manner
reasonably designed to inform fully the users of the system. The notice shall
be conspicuous and shall not use unduly technical language, unduly small
print or other methods which would frustrate the purpose of the notice. The
notice shall disclose all material facts regarding the subject including the
nature of the problem and, when appropriate, a clear statement that a pri-
mary drinking water regulation has been violated and any preventive meas-
ures that should be taken by the public. Where appropriate, or where desig-
nated by the State, bilingual notice shall be given. Notices may include a
balanced explanation of the significance or seriousness to the public health
of the subject of the notice, a fair explanation of steps taken by the system
to correct any problem and the results of any additional sampling.
(f) Notice to the public required by this section may be given bv the
State on behalf of the supplier of water.
(g) In any instance in which notification by mail is required by para-
graph (a) of this section but notification by newspaper or to radio or tele-
vision stations is not required by paragraph lb) of this section, the State
may order the supplier of water to provide notification by newspaper and
to radio and television stations when circumstances make more immediate
or broader notice appropriate to protect the public health.
Section 141.33 Record Maintenance.
Any owner or operator of a public water system subject to the provisions
of this part shall retain on its premises or at a convenient location near its
premises the following records:
(a) Records of bacteriological analyses made pursuant to this part shall
be kept for not less than 5 years. Records of chemical analyses made pur-
suant to this part shall be kept for not less than 10 years. Actual laboratory
22
-------
SUBPART D—REPORTING, PUBLIC NOTIFICATION, AND RECORD KEEPING
reports may be kept, or data may be transferred to tabular summaries, pro-
vided that the following information is included:
(1) The date, place and time of sampling, and the name of the person
who collected the sample;
(2) Identification of the sample as to whether it was a routine distribu-
tion system sample, check sample, raw or process water sample or other
special purpose sample;
(3) Date of analysis;
(4) Laboratory and person responsible for performing analysis;
(5) The analytical technique/method used; and
(6) The results of the analysis.
(b) Records of action taken by the system to correct violations of pri-
mary drinking water regulations shall be kept for a period not less than 3
years after the last action taken with respect to the particular violation
involved.
(c) Copies of any written reports, summaries or communications relat-
ing to sanitary surveys of the system conducted by the system itself, by a
private consultant, or by any local, State or Federal agency, shall be kept for
a period not less than 10 years after completion of the sanitary survey
involved.
(d) Records concerning a variance or exemption granted to the system
shall be kept for a period ending not less than 5 years following the expira-
tion of such variance or exemption.
23
-------
APPENDIX A—DRINKING WATER REGULATIONS
Appendix A
Background Used Li Developing The
National Interim Primary Drinking
Water Regulations
The National Interim Primary Drinking Water Regulations have been
predicated on the best and latest information available at the time of their
promulgation. The concepts and rationale included in this Appendix were
derstanding, judgment, and discretion.
used in arriving at specific limits and should enable those whose responsi-
bility it is to interpret, apply, or enforce the Regulations to do so with un-
A. SOURCE AND FACILITIES
B. MICROBIOLOGICAL QUALITY
C. CHEMICAL QUALITY
A—SOURCE AND FACILITIES
Mounting pollution problems indicate the need for increased attention
to the quality of source waters. Abatement and control of pollution of
sources will significantly aid in producing drinking water that will be in
full compliance with the provisions of these Standards and will be esthet-
ically acceptable to the consumer, but they will never eliminate the need for
well designed water treatment facilities operated by competent personnel.
Production of water that poses no threat to the consumer's health depends
on continuous protection. Because of human frailties associated with pro-
tection, priority should be given to selection of the purest source. Polluted
sources should not be used unless other sources are economically unavail-
able, and then only when personnel, equipment, and operating procedures
can be depended on to purify and otherwise continuously protect the drink*
ink water supply.
Although ground waters obtained from aquifers beneath impervious
strata, and not connected with fragmented or cavernous rock, have been
considered sufficiently protected from bacterial contamination to preclude
need for disinfection, this is frequently not true as ground waters are be-
coming polluted with increasing frequency, and the resulting hazards re-
quire special surveillance. An illustration of such pollution is the presence
of pollutants originating either from sewage or industrial effluents.
Surface waters are subjected to increasing pollution and should never be
used without being effectively disinfected. Because of the increasing hazards
of pollution, the use of surface waters without coagulation and filtration
must be accompanied by adequate past records and intensive surveillance
of the quality of the raw water and the disinfected supply in order to assure
constant protection. This surveillance should include a sanitary survey of
the source and water handling, as well at biological examination of the
supply.
25
-------
DRINKING WATER REGULATIONS
The degree of treatment should be determined by the health hazards in-
volved and the quality of the raw water. When in use, the source should
be under continuous surveillance to assure adequacy of treatment in meet-
ing the hazards of changing pollution conditions. Continuous, effective dis-
infection shall be considered the minimum treatment for any water supply
except for ground waters in which total coliforms can be shown to be con-
tinually absent from the raw water. During times of unavoidable and ex-
cessive pollution of a source already in use, it may become necessary to
provide extraordinary treatment (e.g., exceptionally strong disinfection1, im-
proved coagulation, and/or special operation). If the pollution cannot be
removed satisfactorily by treatment, use of the source should be discon-
tinued until the pollution has been reduced or eliminated.
The adequacy of protection by treatment should be judged, in part, on a
record of the quality of water produced by the treatment plant and the rela-
tion of this quality to the requirements of these Regulations. Evaluation of
adequacy of protection by treatment should also include frequent inspection
of treatment works and their operation. Conscientious operation by well-
trained, skillful, and competent operators is an essential part of protection
by treatment. Operator competency is encouraged by a formal program
leading to operator certification or licensing.
Delivery of a safe water supply depends on adequate protection by na-
tural means or by treatment, and protection of the water in the distribution
system. Minimum protection should include programs that result in the
provision of sufficient and safe materials and equipment to treat and dis-
tribute the water; disinfection of water mains, storage facilities, and other
equipment after each installation, repair, or other modification that may
have subjected them to possible contamination; prevention of health haz-
ards, such as cross-connections or loss of pressure because of overdraft in
excess of the system's capacity; and routine analysis oi water samples and
frequent survey of the water system to evaluate the adequacy of protection.
The fact that the minimum number of samples are taken and analyzed and
found to comply with specific quality requirements of these Standards,
is not sufficient evidence that protection has been adequate. . The protection
procedures and physical facilities must be reviewed along with the results
of water quality analyses to evaluate the adequacy of the supply's protection.
Knowledge of physical defects or of the existence of other health hazards in
the water supply system is evidence of a deficiency in protection of the
water supply. Even though water quality analyses have indicated that the
quality requirements have been met, the deficiencies must be corrected be-
fore the supply can be considered safe.
1 See reference to relationship of chlorine retidual and contact time required to kill
virtues in section on Microbiologics! Quality.
26
-------
APPENDIX A—DRINKING WATER REGULATIONS
B — MICROBIOLOGICAL QUALITY
Colifarm Group
Coliform bacteria traditionally have been the bacteriological tool used to
measure the occurrence and intensity of fecal contamination in stream-pol-
lution investigations for nearly 70 years. During this time, a mass of data
has accumulated to permit a full evalution of the sensitivity and specificity
of this bacterial pollution indicator.
As defined in Standard Methods for the Examination of Water and
Wastewater (1), "the coliform group includes all of the aerobic and faculta-
tive anaerobic, Gram-negative, non-spore-forming rod-shaped bacteria which
ferment lactose with gas formation within 48 hours at 35° C." From this
definition, it becomes immediately apparent that this bacterial grouping is
somewhat artificial in that it embodies a heterogeneous collection of bac-
terial species having only a few broad characteristics in common. Yet, for
practical applications to stream pollution studies, this grouping of selected
bacterial species, which we shall term the "total coliform group," has proved
to be a workable arrangement.
The total coliform group merits consideration as an indicator of pollu-
tion because these bacteria are always present in the normal intestinal tract
of humans and other warm-blooded animals and are eliminated in large
numbers in fecal wastes. Thus, the absence of total coliform bacteria is evi-
dence of a bacteriologically safe water.
Some strains included in the total coliform group have a wide distribution
in the environment but are not common in fecal material. Enterobacter
aerogenes and Enterobacter cloacae are frequently found on various types
of vegetation (2-5) and in materials used in joints and valves (6-7).
The intermediate-aerogenes-cloacae (I.A.C.) subgroups may be found in
fecal discharges, but usually in smaller numbers than Escherichia coli that
is characteristically the predominant coliform in warm-blooded animal in*
testines (8-10). Enterobacter aerogense and intermediate types of organisms
are commonly present in soil (11-14) and in waters polluted some time in
the past. Another subgroup comprises plant pathogens (15) and other or-
ganisms of indefinite taxonomy whose sanitary significance is uncertain. Ail
of these coliform subgroups may be found in sewage and in the polluted
water environment.
Survival Times
Organisms of the I.A.C. group tend to survive longer in water than do
fecal coliform organisms (16-18). The I.A.C. group also tends to be some-
what more resistant to chlorination than E. coli or the commonly occurring
bacterial intestinal pathogens (19-22). Because of these and other reasons,
the relative survival times of the coliform subgroups may be useful in dis-
tinguishing between recent and less recent pollution. In waters recently
contaminated with sewage, it is expected that fecal coliform organisms will
be present in numbers greater than those of the I.A.C. subgroup; but ia
27
-------
DRINKING WATER REGULATIONS
waters that have been contaminated for a considerable length of time or
have been insufficiently chlorinated, organisms of the I.A.C. subgroup may
be more numerous than fecal coliform organisms (23).
Differentiation of Organisms
Because various numbers of the coliform group normally grow in diverse
natural habitats, attempts have been made to differentiate the population in
polluted waters, with specific interest directed toward those coliforms that
are derived from warm-blooded animal contamination. In his pioneering
research, McConkey (23, 24) defined the aerogenes group in terms of
certain fermentation characteristics, ability to produce indole, and reaction
in the Voges-Proskauer test. Other developments refined techniques that
progressed to differentiate the coliform group on the basis of indole pro-
duction, methyl red, and Voges-Proskauer reactions, and citrate utilization
(IMViC tests) into the E. coli, Enterobacter aerogenes, intermediate, and
irregular subgroups (24-28).
In another approach to coliform differentiation, Hajna and Perry (29)
and Vaughn, Levine, and Smith (30) further developed the Eijkman test
(31) to distinguish organisms of fecal origin from those of nonfecal origin
by elevating the incubation temperature for lactose fermentation. Geldreich,
and associates, (31, 32) further refined the procedure and developed ad-
ditional data to indicate the specific correlation of this elevated tempera-
ture procedure to the occurrence of fecal contamination.
Fecal Coliform Measurements
The fecal coliform bacteria, a subgroup of the total coliform population,
does have a direct correlation with fecal contamination from warm-blooded
animals. The principal biochemical characteristic used to identify fecal coli-
form is the ability to ferment lactose with gas production at 44.5° C. Re-
search data have shown that 96.4 percent of the coliforms in human feces
were positive by this test (10). Examination of the excrement from other
warm-blooded animals, including livestock, poultry, cats, dogs, and rodents
(33-34), indicate the fecal coliforms contribute 93.0 to 98.7 percent of the
total coliform population. The predominant fecal coliform type most fre-
quently found in the intestinal flora is E, coli. Occasionally, other coliform
IMViC types may predominate for periods of several months before a shift
occurs in type distribution. For this reason, it is more significant to be
able to measure all coliforms common to the intestinal tract In man, par-
ticularly, there is a significantly greater positive correlation with the
broader fecal coliform concept (96.4 percent) than with identification of
E. coli by the traditional IMViC biochemical reactions (87.2 percent).
Application to Treated Water
The presence of any type of coliform organism in treated water suggests
either inadequate treatment or contamination after post-chlorination (23).
It it true there are some differences between various coliform strains with
28
-------
APPENDIX A—DRINKING WATER REGULATIONS
regard to natural survival and their chlorination resistance, but these are
minor biological variations that are more clearly demonstrated in the labor*
atory than in the water treatment system. The presence of any coliform bac-
teria, fecal or nonfecal, in treated water should not be tolerated.
Insofar as bacterial pathogens are concerned, the coliform group
is considered a reliable indicator of the adequacy of treatment. As an indi-
cator of pollution in drinking water supply systems, and indirectly as an
indicator of protection provided, the coliform group is preferred to fecal
coliform organisms. Whether these considerations can be extended to in-
clude rickettsial and viral organisms has not been definitely determined.
Sample Size
The minimum official sample volume cited in the earlier editions of the
Drinking Water Standards and Standard Methods for the Examination of
Water and Wastewater was either stated or implied to be 50 ml because of
the requirement to inoculate a series of 5 lactose broth fermentation tubes,
each with a 10 ml or 100 ml portion of the sample. Few laboratories ever
routinely employed the larger portions in the multiple tube procedure be-
cause of the attendant problems of preparing, handling and incubating the
larger sized sample bottles that are required. Thus, when the multiple tube
procedure was used, it became a practice to examine only 50 ml. With the
development of the membrane filter procedure for routine potable water
testing, the examination of larger sample volumes became practical, limited
only by the turbidity of water and excessive bacterial populations.
Since many water supplies are sampled infrequently during the month,
it is statistically more meaningful to examine a large sample for greater test
precision with reduced risk of failing to detect some low level occurrence of
coliforms. Increasing the sample portion examined will tighten the base line
sensitivity and is particularly important for measuring the coliform re-
duction capacity of disinfection that approaches the magnitude essential for
control of waterborne virus. Mack et al (35) reported poliovirus type II
could be isolated from a restaurant well water supply using a flocculant in
the 2.5 gallon samples prior to centrifugation to concentrate the low density
virus particles. Bacteriological examinations of 50 ml portions of the un-
concentrated water samples were negative for coliforms. However, coliforms
were found in the concentrated sediment pellets. Future studies on coliform
to virus occurrence in potable water may require further tightening of the
coliform standard, possibly to a one-liter base (36).
The recommendations to increase the sample size to 100 ml for bacterio-
logical examinations of water is supported in the 13th Edition of Standard
Methods where the larger volume is stated as preferred. A study of State
Health Laboratory procedures indicates that 39 or 78 percent of these
laboratory systems are currently using 4 oz sample bottles to collect 100 ml
of sample, and 25 of these State Health Laboratory networks are examining
all public water samples by the membrane filter procedure. These figures
29
-------
DRINKING WATER REGULATIONS
suggest that the stronger position now being proposed on a minimum sample
size of 100 ml for statistically improved coliform monitoring is not
unrealistic in terms of current practice.
Application to Source Waters and Untreated Potable Supplies
In the monitoring of source water quality, fecal coliform measurements
are preferred, being specific for fecal contamination and not subject to
wide-range density fluctuation of doubtful sanitary significance.
Although the total coliform group is the prime measurement of potable
water quality, the use of a fecal coliform measurement in untreated potable
supplies will yield valuable supplemental information. Any untreated
potable supply that contains one or more fecal coliforms per 100 ml should
receive immediate disinfection.
REFERENCES
1. Standard Methods for the Examination of Water and Wastewater, 13th ed. APHA,
A WW A, WPCF, New York (1970).
2. Thomas, S.B. and McQuillin, J. Coli-aerogenes Bacteria Isolated from Grass. J. Appl.
Bacterid, 15: 41 (1952).
3. Fraaer, M.H., Reid, W.B., and Malcolm, J.F. The Occurrence of Coli-aerogenes
Organisms on Plants. J. Appl. Bacterioi. 19:301 (1956).
4. Geldreich, E.E., Kenner, B.A., and Kabler, P.W. The Occurrence of Coliforms,
Fecal Coliforms, and Streptococci on Vegetation and Insects. Appl. Microbiol.
12: 63 (1964).
5. Papavassiliou, J., Tzannetis, S., Yeka, H., and Micfaapoulos, G. Coli-Aerogenes
Bacteria on Plants. J. Appl. Bacterioi, 30 : 219 (1967).
6. Caldwell, E.L., and Parr, L.W. Pump Infection Under Normal Conditions in Con-
troled Experimental Fields. J A WW A. 25: 1107 (1933).
7. Rapp, W.M. and Weir, P. Cotton caulking yarn. JAWWA. 26: 743 (1934).
8. Parr, L.W. The Occurrence and Succession of Coliform Organics in Human Feces,
Am. J. Hyg.27 : 67 (1938).
9. Sears, HJ., Browles, I., and Vchiyama, J.K. Persistence of Individual Strains of
Escherichia coli in the Intestinal Tract of Man. J. Bacterioi. 59 : 293 (1950).
10. Geldreich, E.E., Bordner, R.H., Huff, C.B., Clark, H.F., and Kabler, P.W. Type-
Distribution of Coliform Bacteria in the Feces of Warm-Blooded Animals. JWPCF.
34: 29S (1962).
11. Frank, N. and Skinner, C.E. Coli-Aerogenes Bacteria in Soil. J, Bacterioi. 42:
143 (1941).
12. Taylor, C.B. Coli-Aerogenes Bacteria in Soils, J. Hyg. (Camb.) 49: 162 (1951).
13. Randall, J3. The Sanitary Significance of Coliform Baccilli in Soil. J. Hyg.,
(Camb.) 54: 365 (1956).
14. Geldreich, E.E., Huff, C.B., Bordner, R.H., Kabler, P.W., Clark, H.F. The Fecal
Coli-Aerogenes Flora of Soils from Various Geographical Areas. J. Appl. Bacterioi.
25: 87 (1962).
15. Elrod, R.P. The frwinut-Coliforn Relationship J. Bacterioi, 44: 433 (1942).
16. Parr, L.W. Viability of Coli-Aerogenes Organisms in Cultures and in Various En*
~ironments. J. Infect. Disease 60: 291 (1937).
17. Piatt, A.E. The Viability of Bact. coli and Bact. aero genes in Water: A Method for
the The Rapid Enumeration of These Organisms. J. Hyg. 35: 437 (1935).
18. Taylor, C.B. The Ecology and Significance of the Different Types of Coliform
Bacteria Found in Water. J. Hyg. 42: 23 (1942).
30
-------
APPENDIX A—DRINKING WATER REGULATIONS
19. Tonney, F.O., Greer, F.E., and Danforth, T.F. The Minimal "Chlorine Death Point"
of Bacteria. Am. J. Pub. Health 18: 1259 (1928).
20. Heathman, L.S., Pierce, S.O., and Kabler, P.W, Resistance of Various Strains of
E typhi and Coli-Aerogenes to Chlorine and Chloramine. Pub. Health Rpts., 51:
1367 <1936).»
21. Butterfield, C.T., et. al Influence of pH and Temperature on the Survival of Coli-
forms and Enteric Pathogens when Exposed to Free Chlorine. Pub. Health Rpts.
58: 1837 (1943).
22. Kabler, P.W. Relative Resistence of Coiiform Organisms and Enteric Pathogens
in the Disinfection of Water with Chlorine. JAWWA, 43: 553 (1951).
23. Kabler, P.W. and Clark, H.F. Coiiform Group and Fecal Coiiform Organisms as
Indicators of Pollution in Drinking Water. JAWWA. 52: 1577 (I960).
24. MacConkey, A. Lactose-Fermenting Bacteria in Feces. J. Hyg. 5: 333 (1905).
25. MacConkey, A. Further Observations on the Differentiation of Lactose-Fermenting
Bacteria, with Special Reference to Those of Intestional Origin. J. Hyg. 9: 86
(1909).
26. Rogers, L.A., Clark, W.M., and Davis, BJ. The Colon Group of Bacteria. J. Infect.
Disease 14: 411 (1914).
27. Clark, W.M., and Lubs, W.A. The Differentiation of Bacteria of the Colon-Aero-
genes Family by the Use of Indicators. J. Infect. Disease 17: 160 (1915).
28. Koser, S.A. Differential Tests for Colon-Aerogenes Group in Relation to Sanitary
Quality of Water. J. Infect. Disease 35: 14 (1924).
29. Hajna, A.A. and Perry, C.A. A Comparison of the Eijkman Test with other Tests
for Determining E. coli. J, Bacterioi, 30; 479 (1935).
30. Vaughn, R.H., Levine, M., and Smith, H.A. A Buffered Boric Acid Lactose Medium
for Enrichment and Presumptive Identification of Escherichia coli. Food Res. 16: 10
(1951).
31. Eijkman, C. Die Garungsprobe bei 46 ais Hilfsmittel bei der Trinkwasserunter-
suchung. Centr. Bakteriol. Parasitenk., Abt. I, Orig., 37: 742 (1904).
32. Geldreich, E.E., Clark, H.F., Kabler, P.W., Huff, C.B., and Bordner, R.H. The
Colifonn Group. II, Reactions in EC Medium at 45 C. Appl. Microbiol. 6: 347
(1958).
33. Geldreich, E.E. Sanitary Significance of Fecal Colifortns in the Environment. U.S.
Dept. of the Interior, FWPCA Publ. WP 20-3 (1966).
34. Geldreich, E.E., Best, L.C., Kenner, B.A., and VonDonset, D J., The Bacteriological
Aspects of Stormwater Pollution. JWPCF. 40: 1861 (1968).
35. Mack, N.M., Tu, Y.S. and Coohon, D.B., Isolation of Poliomyelitis Virus from a
Contaminated Well. H.S.M.H.A. Health Reports (In press).
36. Geldreich, E.E. and Clarke, N.A., The Coiiform Test: A Criterion for the Viral
Safety of Water. Proc. 13th Water Quality Conference, College of Engineering,
University of Illinois, pp. 103-113 (1971).
Substitution of Residual Chlorine Measurement for Total Coiiform
Measurement
The best method of assuring the microbiological safety of drinking water
is to maintain good clarity, provide adequate disinfection, including main-
tenance of a disinfectant residual, and to make frequent measurements of
the total coiiform density in the distributed water. In the 1962 U.S. Public
31
-------
drinking water regulations
Health Service Drinking Water Standards, the major emphasis was on the
measurement of total coliform densities and a sampling frequency graph
relating number of samples per month to population served was included.
The sampling frequency ranged from two per month for populations of
2,000 and less to over 500 per month, for a population of 8 million.
The effectiveness of this approach for assuring microbiological safety
was evaluated during the 1969 Community Water Supply Survey. The re-
suits of this evaluation by McCabe, et al,, (1) are paraphrased below.
Microbiological Quality
To determine the status of the bacteriological surveillance program in
each of the 969 water supply systems investigated, records in the State and
county health departments were examined for the number of bacteriological
samples taken and their results during the previous 12 months of record.
Based on this information, only 10 percent had bacteriological surveillance
programs that met the "criteria," while 90 percent either did not collect
sufficient samples, or collected samples that showed poor bacterial quality,
or both. The table below summarizes the results.
Bacteriological Surveillance
500 or 501 Greater than All
Population Less 100,000 100,000 Populations
Number of Systems -
446 501
22
969
Percent of Systems
Met Criteria
36
10
Did no! meet
'
Criteria ..... .... —
95 85
64
90
Sampling Frequency
Insufficient samples were taken in more than one of the previous 12
months of record from 827 systems (85 percent of the survey total). Even
considering a sampling rate reduced by 50 percent of that called for in the
criteria, 670 systems (69 percent of the survey total) still would not have
collected sufficient samples.
Recommendation
The water utility should be responsible for water quality control, but the
bacteriological surveillance collection requirements are not being met in
most small water systems even though only two samples per month are re-
quired. A more practical technique must be developed if the public's health
is to be protected. If all systems were chlorinated, a residual chlorine deter-
mination might be a more practical way of characterizing safety.
The validity of the recommendation that the measurement of chlorine
residual might be a substitute for some total coliform measurements has
been investigated by Buelow and Walton (2) Because the recommended
rate of sample collection could not be or was not being used, alternative
32
-------
APPENDIX A—DRINKING WATER REGULATIONS
methods of indicating safety were considered. One suggestion was to sub-
stitute the measurement of chlorine residual for some of the bacteriological
samples. This method has the advantage of being easy to perform, and thus
providing an immediate indication of safety. Further, data from London,
U.K.; Cincinnati, Ohio; and the 1969 Community Water Supply Survey
(CWSS) has shown that present sampling locations do not protect all con-
sumers and that chlorine residual can be used to replace some coliform
determinations.
Sampling Location
During 1965-66, the London Metropolitan Water Board using its
Standards, made bacteriological examinations of 11,371 samples of water
entering the distribution system, 947 samples taken from distribution
reservoirs, 2,720 samples taken following pipeline breaks anl 689 samples
from miscellaneous locations (complaints, hospitals, etc.). Most of the
unsatisfactory results were associated with reservoir problems. Main breaks
and miscellaneous samples were responsible for most of the remaining un-
satisfactory samples.
Chlorine Residual
In Cincinnati during the 1969-70 period of free chlorine residual, ap-
proximately 24 samples were collected from each of 143 sampling stations.
None of the samples from 116 of these stations showed presence of coliform,
and 23 of the remaining sampling stations showed coliform bacteria in only
one out of the approximately 24 samples examined. At the other four sta-
tions where 2 or more coliform-positive tests were obtained from the 24
samples, three had no chlorine residual at the time the coliform-positive
samples were collected. The question is raised, therefore, as to the need
for examining samples routinely collected from a large number of stations
scattered throughout the system without regard to the water's residual
chlorine content. Maintaing a free chlorine residual of 0.2 mg/1 in the
Cincinnati, Ohio, distribution system reduced the percentage of coliform
positives to about 1 percent. The table below from the CWSS data, shows
that the presence of a trace or more of chlorine residual drastically reduced
or eliminated total coliforms from distribution system samples.
Percent of Various Types of Water Supply Systems Found to Have Average
Total Couforms Greater than 1/100ml
Non-
Chlorinated
With Any
Type of System
Chlorinated
No Residual
Detectable Residual
Spring .
39
17
0
Combined Spring and Well —
41
28
0
Well .. „ .
8
5
0
Surface ... —
64
7
2
Combined Surface and Well ...
100
16
3
33
-------
DRINKING WATER REGULATIONS
These findings indicate that a major portion of a distribution system,
exclusive of deadends, reservoirs, etc., could be monitored for bacteriological
safety by the use of chlorine residual. (Emphasis added.) Therefore, when
chlorine substitution is used, determination of total coiiform densitities
should be continued in problem areas, and some samples, as a check, should
be collected in the main part of the distribution system.
These two studies led to the inclusion in the Regulations of Par.
141.21(h) on the substitution of chlorine residual tests for a portion of the
required total coiiform determinations. Par. 141.21(h) states that any sub-
stitution must be approved by the State on the basis of a sanitary survey.
The following four items should be specified by the State:
1. The number and location of samples for which chlorine residuals are
to be substituted.
2. The form and concentration of chlorine residual to be maintained;
3. The frequency of chlorine residual determinations; and
4. The analytical method to be used.
While each approval must be made individually, taking into account in-
dividual circumstances, the following may offer some guidance. The first
requirement is the establishment of the relationship between chlorine resi-
dual and the absence of total coliforms in any given water. This may not be
too difficult in larger supplies where both of these measurements are rou-
tinely made, but it might be quite difficult for the smaller purveyors (where
the most help is needed) who have not been making either measurement.
The number and location of samples for which chlorine residuals are to be
substituted
Total coiiform measurements should continue to be made of the finished
water as it enters the distribution system and at known trouble spots such
as reservoirs and dead ends. Substition can be considered in the free-flowing
portion of the distribution system.
The chlorine residual to be maintained
In general, a low turbidity water with a free chlorine residual of about
0.2 mg/1 at a pH of less than 8.5 will be free from total coliforms although
these conditions may vary from water to water. However, a higher free
chlorine residual or the use of some other disinfectant is required prior to
the water entering the distribution system, where disinfection is practiced,
if initial disinfection is to be adequate.
The frequency of chlorine residual determinations
Because the chlorine residual test is so easy to perform, it is reasonable to
expect the substition of several chlorine residual determinations for each
total coiiform test deleted. In this way wider coverage of the distribution
system can be achieved, thereby increasing the protection to the consumer.
Since, for maximum protection, chlorination must be continuous, it is also
reasonable to expect that a minimum of one daily determination of chlorine
34
-------
APPENDIX A—DRINKING WATER REGULATIONS
residual be performed whenever the chlorine residual option has been
chosen. By limiting the extent of substitution to 75% of the required bacteri-
ological samples, a sufficient number of bacteriological samples will still
be taken to enable the assessment of the adequacy of disinfection and to
assure the continuity of water quality records.
The analytical method to be used
An analytical method free of interferences to eliminate false residuals
must be recommended. For this reason the DPD method is specified.
Finally, when the chlorine residual option is in use and a free chlorine
residual concentration less than that agreed to is measured at a sampling
point, then a sample for total coliform analysis must be taken immediately
from that point.
REFERENCES
1. McCabe, L.J. Symona, J.M., Lee, R.D., and Robeck, G.C. Study of Community Water
Supply Systems. JA WWA. 62 : 670 (1970).
2. Buelow, R.W., and Walton, G. Bacteriological Quality vs. Residual Chlorine. JAWWA.
63: 28 (1971).
General Bacterial Population
The microbial flora in potable water supplies is highly variable in num-
bers and kinds of organisms. Those bacterial groups most frequently en-
countered in potable waters of poor quality include: Pseudomonas, Flavo-
bacterium, Achromobacter, Proteus, Klebsiella, Bacillus, Serratia, Coryne-
bacterium, Spirillum, Clostridium, Arthrobacter, Gallionella, and Leptothrix
(1-5). Substantial populations of some of these organisms occurring in
potable water supplies may bring a new area of health risk to hospitals,
clinics, nurseries, and rest homes (6-11). Although Pseudomonas organisms
are generally considered to be non-pathogenic, they can become a serious
"secondary pathogenic invader" in post-operation infections, burn cases,
and intestinal-urinary tract infections of very young infants and the elderly
population of a community. These organisms can persist and grow in water
containing a minimal nutrient source of nitrogen and carbon. If Pseudo-
monas becomes established in localized sections of the distribution lines, it
may persist for long periods and shed irregularly into the consumer's
potable water supply (12). A continual maintenance of 0.3 to 0.6 mg/l free
chlorine residual will suppress the development of an extensive microbial
flora in all sections of the distribution network.
Flavobacterium strains can be prevalent in drinking water and on water
taps and drinking-fountain bubbler-heads. A recent study of stored emer-
gency water supplies indicated that 23 percent of the samples contained
Flavobacterium organisms with densities ranging from 10 to 26,000 per
1 ml. Flavob
-------
DRINKING WATER REGULATIONS
cause it can become a primary pathogen in persons who have undergone
surgery (13).
Klebsiella pneumoniae is another secondary invader that produces human
infection of the respiratory system, genitourinary system, nose and throat,
and occasionally this organism has been reported as the cause of meningitis
and septicemia (14). Klebsiella pneumoniae, like Enterobacter aero genes,
(15) can multiply in very minimal nutrients that may be found in slime
accumulations in distribution pipes, water taps, air chambers, and aerators.
Coliform Suppression
The inhibitory influence of various organisms in the bacterial flora of
water may be important factor that could negate detection of the coliform
group (16-17,). Strains of Pseudomonas, Sarcina, Micrococcus, Flavo-
bacterium, Proteus, Bacillus, Actinomycetes, and yeast have been shown to
suppress the detection of the coliform indicator group (18-21). These or-
ganisms can coexist in water, but when introduced into lactose broth they
multiply at a rapid rate, intensifying the factor of coliform inhibition (22).
Suspensions of various antagnoistic organisms in a density range of 10,000
to 20,000 per 1 ml, added to lactose tubes simultaneously with a suspension
of 10 E. coli per 1 ml, resulted in reduction in coliform detection (19). This
loss of test sensitivity ranged from 28 to 97 percent, depending on the com-
bination of the mixed strains.
Data from the National Community Water Supply Survey (23) on bac-
teriological quality of distribution water from the 969 public water supplies
were analyzed (Table 1) for bacterial plate count relationship to detection
of total coliforms and fecal coliforms. It is interesting to note that there was
a significant increase in total and fecal coliform detection when the bacterial
counts increased up to 500 per 1 ml. However, further increase in the de-
tection of either coliform parameter did not occur when the bacterial count
per 1 ml was beyond 500 organisms. There was, in fact, progressively de-
creased detection of both coliform parameters as the bacterial count con-
tinued to rise. This could indicate an aftergrowth of bacteria in distribution
system water or a breakpoint where coliform detection was desensitized by
the occurrence of a large general bacterial population that included organ-
isms known to suppress coliform recovery.
Control of the General Bacterial Population
Density limits for the general bacterial population must be related, in
part, to a need to control undesirable water quality deterioration and prac-
tical attainment for water throughout the distribution system. This necessity
for monitoring the general bacterial population is most essential in those
supplies that do not maintain any chlorine residual in the distribution lines
and in special applications involving desalinization. This bacteriological
measurement would serve as a quality control on water treatment processes
and sanitation of distribution line sections and storage tanks that could
36
-------
APPENDIX A—DRINKING WATER REGULATIONS
be shedding various quantities of organisms into the system, thereby de-
grading the water quality.
Table 1.—Bacterial Plate Count vs. Coliform Detection in Distribution Water Networks
for 969 Public Water Supplies _
General Bacterial Population* Total Coliform Fecal Coliform
Density Range Number of
per 1 ml
Samples
Occurrences
Percent
Occurrences
Percent
1 - 10
1013
47
4.6
22
2 2
11 - 30
371
28
7.5
12
3.2
31 • 100
396
72
18.2
28
7.1
101 - 300
272
48
17.6
20
7.4
301 ¦ 500
120
30
25.0
11
92
501 - 1,000
110
21
19.1
9
8.2
1,000
164
31
18.9
5
3.0
TOTAL
2446
277
.—
107
—
* Standard Plate Count (48 brs. incubation, 35° C)
Practical attainment of a low general bacterial population can best be
judged by a study of data from the National Community Water Supply
Survey. Data presented in Table 2 demonstrate the effectiveness of chlorine
residual in controlling the general bacterial population in a variety of com*
munity water supply distribution systems, Although the number of samples
on each distribution system in this special study was small, it does reflect
bacterial quality conditions in numerous large and small water systems
examined in each of the eight metroplitan areas and the entire State of
Vermont
These data indicate that the general bacterial population in distribution
lines can be controlled to a value below 500 organisms per 1 ml by main-
taining a residual chlorine level in the system. Increasing the chlorine resi-
dual above 0.3 mg/1 to levels of 0.6 and 1.0 mg/1 did not further reduce
the bacterial population by any appreciable amount. Restricting such bac-
terial densities to a limit of 500 organisms per ml is, therefore, not only
attainable in the distribution system, but is also desirable to prevent loss in
coliform test sensitivity definitely observed at approximate densities of 1000
organisms per ml, thereby producing a safety factor of at least two.
Table 2.—The Effect of Varying Levels of Residual Chlorine on the Total Plant Count
in Potable Water Distribution Systems*
Standard Plate Residual Chlorine (mg/1)
Count**
0.0
0.01
0.1
0.2
03
0.4
0.5
0.6
<1
8.1***
14.6
19.7 r
12.8
16.4
17.9
4.5
17.9
1 • 10
20.4
29-2
38 J?
48.9
45-5
51.3
59.1
42.9
11 - 100
37.3
33.7
28.9
26.6
23.6
23.1
31.8
28.6
101 • 500
18.6
11.2
7.9
9.6
12.7
5.1
4£
10.7
501 - 1000
5.6
6.7
L3
2.1
U
0
0
0
>1000
10.0
4.5
3.9
0
0
2.6
0
0
Number of
Samples
520
89
76
94
55
39
22
28
37
-------
DRINKING WATER REGULATIONS
•Data from a survey of community water supply systems in 9 metropolitan areas (23)
••Standard Plate Count (48 hrs. incubation, 35°C)
•**AU values are percent of samples that had the indicated standard plate count.
Any application of a limit for the general bacterial population in potable
water will require a definition of medium, incubation temperature, and in-
cubation time so as to standardize the population to be measured. The 13th
edition of Standard Methods for the Examination of Water and Wastewater
does specify these requirements for a Standard Plate Count (SPC) to be
used in collection of water quality control data. Because many organisms
present in potable waters are attenuated, initial growth in plate count agar
frequently is slow; thus, incubation time should be extended to 48 hours at
35 C. This time extension will permit a more meaningful standard count
of the viable bacterial population. Samples must be collected in bottles
previously sterilized within 30 days and adequately protected from dust
accumulation. Examination for a Standard Plate Count should be initiated
within 8 hours of collection. This time may be extended to periods up to
30 hours only if these samples are transported in iced containers.
With maintenance of a chlorine residual and turbidity of less than one
Turbidity Unit, the need for a bacteriological measurement of the distribu-
tion system may become less critical. For this reason, it is recommended
that such water supplies be monitored routinely for baseline data on the
general bacterial population and correlated with chlorine residual and tur-
bidity measurements in the distribution lines. It is also recommended that
water plant personnel be alert to unusual circumstances that may make it
desirable to monitor the general bacterial population more often in a check
of water plant treatment efficiencies.
For these reasons, the general bacterial population should be limited to
500 organisms per I ml in distribution water. In theory, the limitation of
the general bacterial population to some practical low level would also in-
directly and proportionally limit any antagonistic organisms that could sup-
press coliform detection and reduce the exposure and dosage level for health
effect organisms that might be present.
While no maximum contaminant level for general bacterial populations
is included in the Interim Primary Drinking Water Regulations, it is
recommended that the limit mentioned above be used as an operational
guide in assessing the quality of drinking water delivered.
REFERENCES
1. Willis, A.T. Anaerobic Bacilli in a Treated Water Supply. J. Appl. Bacteriol.
20: 61 (1957).
2. Lueschow, LA. and Mackenthon, K.M. Detection and Enumeration of Iron Bac-
teria in Municipal Water Supplies. JAWWA. 54 : 751 (1962).
3. Clark, F.M., Scott, R.M. and Bone, E. Heterotrophic Iron Precipitating Bacteria,
JAWWA. 59: 1036 (1967).
4. Victoreen, H.T. Soil Bacteria and Color Problem in Distribution System*. JAWWA.
61: 429 (1969).
38
-------
APPENDIX A—DRINKING WATER REGULATIONS
5. Victoreen, H.T. Panel Discussion on Bacteriological Testing of Potable Waters.
Am. Water Works Assoc. Annual Conference. June 21-26, 197(£ Washington, D.C.
6. Culp, R.L. Disease Due to "Non-pathogenic" Bacteria. JAWWA 61: 157 (1969).
7. Roueche, B. The Annals of Medicine. Three Sick Babies. The New Yorker, Oct. 5,
1968.
8. Hunter, C.A. and P.R. Ensign. An Epidemic of Diarrhea in a New-born Nursery
Caused by P. aeruginosa. AJ. Pub. Health 37: 116 11947).
9. Drake, C.H. and Hoff, J.C. Miscellaneous Infection Section VI- Pseudomona»
aeruginosa Infections, pp. 635-639. In: Diagnostic Procedures and Reagents, A.H.
Harris and M.B. Coleman, editors, Am. Pub. Health Assoc. New York, 4th ed.
(1963).
10. Smith, W.W. Survival after Radiation Exposure • Influence of a Disturbed En-
vironment. Nucleonics 10: 80 (1952).
11. Maiztegui, J.I. et al, Cram-Negative Rod Bacteremia with a Discussion of In-
fections Caused by Herella Species. Am. J. Surgery 107 : 701 (1964).
12. Cross, D.F., Benchimoi, A., and Dimond, E.G. The Faucet Aerator-A Source of
Psettdamonas Infection. New England J. Med. 274: 1430 (1966).
13. Herman, L.C. and Himmelsbach, C.K. Detection and Control of Hospital Sources of
Flavobacteria. Hospitals. J. Axh. Hospital Assn. 59: (1965).
14. Leiguarda, R.H. and Polazzolo, A.Z.Q.D., Bacteria of Genua Klebsiella in Water.
Rev. Obr. Sanit. Nac., (Argentina) 38: 169 (1956).
15. Nunez, WJ. and Colmer, A.R. Differentiation of Aerobacter-Klebsiella Isolated
from Sugarcane. Appl. Microbiol. 16: 1875 (1965).
16. Waksman, S.A. Antagonistic Relations of Microorganisms. Bacteriol. Reviews
5: 231 (1941).
17. Schiavone, E.L. and Passerini, L.M.D. The Genus Pseudomonas aeruginosa in the
Judgment of the Potability of Drinking Water. Sem. Med., (B. Aires) 111-. 1151
(1957).
18. Kligler, LJ. Non-lactose Fermenting Bacteria from Polluted Wells and Sub-soil,
J. Bacteriol. 4: 35 (1919).
19. Hutchinson, D., Weaver, R.H. and Scherago, M. The Incidence and Significance of
Microorganisms Antagonistic to Escherichia coli in Water. J. Bacteriol. 45: 29
(1943).
20. Fisher, G. The Antagonistic Effect of Aerobic Sporulating Bacteria on the Coll-
Aerogenes Group. Z. Immam Forsch 107: 16 (1950).
21. Weaver, R.H. and Boiter, T. Antibiotic-Producing Species of Bacillus from Well
Water. Trans. Kentucky Acad. Sci. 13: 183 (1951).
22. Reitter, R. and Seligmann, R. Pseudomonas aeruginosa in Drinking Water.
J. Appl. Bacteriol. 20: 145 (1957).
23. McCabe, LJ., Symons, J.M., Lee, R.D., and Robeck, G.G. Study of Community
Water Supply Systems. JAWWA. 62: 670 (1970).
Enteric Viruses in Water
Viruses of fecal and/or urinary origin from any species of animal may
pollute water. Especially numerous, and of particular importance to health,
are those viruses of human enteric origin. They include poliovirttses, cox-
sackieviruses, echo viruses, adenoviruses, reoviruses, and the infectiua hep-
atitis virus(es). Each group or subgroup consists of a number of different
39
-------
DRINKING WATER REGULATIONS
serological types so that to date more then 100 different human enteric
viruses are recognized.
Most infections with enteric viruses are mild, and many infections are
subclinical, i.e., the infected individual is not sick. It is, however, generally
agreed that all human viruses are pathogens, and as clinical experience ac-
cumulates, it is evident that the enteric viruses have at least two distinct
effects on man: (a) the acute effect, e.g., poliomyelitis, meningitis, in-
fectious hepatitis, etc. (b) the delayed effect, e.g., spontaneous abortion,
congenital heart anomalies, insulin-dependent diabetes, malignancies, etc.
All available evidence to date indicates that the acute clinical effects of
enteric virus infection are many times more common than the delayed clin-
ical effects which appear to be extremely rare and, in many cases,
speculative.
Mosley (1) reviewed the literature in 1968 and cited 50 waterborne out-
breaks of infectious hepatitis and 8 waterborne outbreaks of poliomyelitis.
Nine of these infectious hepatitis outbreaks occurred in the United States,
and 3 of these were reportedly from chlorinated municipal supplies. One is
not certain, however, whether these 3 water supplies were really adequately
treated. Only one of the 8 polimyelitis epidemics occurred in the United
States, and this was the result of cross-connection contamination. Since
Mosley's publication there have been three other reports of waterborne in-
fectious hepatitis outbreaks in this country, all reportedly due to either
sewage pollution of well water or cross-connection contamination. An esti-
mated 20,000 - 40,000 cases of infectious hepatitis were reported in Delhi,
India, in 1955-56 (2) atrributable to a municipal water supply source
heavily overloaded with raw sewage. This outbreak, however, was not ac-
companied by noticeable increases of typhoid fever or other enterobacterial
diseases, suggesting that, in practice, the virus (es) of infectious hepatitis
may be more resistant to chlorine or chloramines than are vegetative bac-
teria. Weibel and co-workers (3) listed 142 outbreaks of gastroenteritis dur-
ing the period of 1946 to 1970 in which epidemiologic evidence suggested
a waterborne nature. More than 18,000 persons were affected in these out-
breaks. Mosley (1) suspected that a significant portion of these cases must
have been caused by viruses.
It is well recognized that many raw water sources in this country are
polluted with enteric viruses. Thus, water supplies from such sources depend
entirely upon the treatment processes used to eliminate these pollutants.
Even though the processes may be perfectly effective, an occasional break-
down in the plant or any marginal practice of treatment could still allow
the pollutants to reach the finished water supplies. It should be noted
that Coin and his associates (4) have reported the recovery of viruses
from raw and finished waters in Paris, France. Coin estimated that the Paris
water probably contained one tissue culture unit of virus per 250 liters. Very
recently, Mack et al (5) reported that poliovirus was recovered in water
40
-------
APPENDIX A—DRINKING WATER REGULATIONS
from a deep well in Michigan. Although the well had a history of positive
coliforms, coliforms and virus were not recovered from an unconcentrated
water sample; only after a 2.5 gallon sample of water was subjected to high
speed centrifugation were both virus and coliforms recovered. This study
would seem to indicate that the present method of using the coliform test is
not adequate to provide assurance of the non-presence of viruses. In sum-
mary, in the United States, most waterborne virus disease outbreaks have
resulted from contamination of poorly treated drinking water by sewage
either directly or through cross-connections. Overt outbreaks of virus disease
from properly treated municipal water supplies are not known to have
occurred. Proper treatment of surface water usually means clarification
followed by effective disinfection.
Chang (6), however, has theorized that some water supplies that practice
only marginal treatment may contain low levels of human viruses, and that
this small amount of virus might initiate infection or disease in susceptible
individuals. He believes that such individuals might thus serve as "index
cases" and further spread the virus by person-to-person contact. Whether
this hypothesis is true, can be proved only by an intensive survey for viruses
in numerous drinking water supplies in this country, and such a survey has
never been conducted. If viruses were detected in a survey of drinking water
supplies, it would be necessary to conduct in-depth epidemiological studies
to determine if actual infection or disease was being caused by these agents.
Additionally, it would be necessary to determine what modifications would
be required in the water treatment processes to eliminate these viruses.
The relative number of viruses and coliform organisms in domestic sew-
age is important in assessing the significance of the coliform test and the
"virus safety" of water. Calculations by Clarke et al (7) have indicated the
following virus-coliform ratios in feces, sewage, and polluted waters.
Calculated Virus-Coliform Ratios
Virus Coliform Ratio
Feces 200/gm 13xl0?gm 1:65,000
Sewage 500/100 ml 46x109100ml 1:92,000
Polluted Surface Water ... 1/100 ml 5xl04/100 ml 1:50,000
It is apparent that coliform organisms far outnumber human enteric
viruses in feces, sewage, and polluted surface water. It should be emphasized
that these calculated ratios are only approximations and that they would be
subject to wide variations and radical changes, particularly during a virus
disease epidemic. Additionally, both bacteria and virus populations in sew-
age and polluted waters are subject to reductions, at different rates, from die-
off, adsorption, sedimentation, dilution, and various other undetermined
causes; thus, the coliform-virus ratio changes, depending upon conditions
resulting from the combined effect of all factors preesnt. Thus, one must take
41
-------
DRINKING WATER REGULATIONS
into consideration the most unfavorable conditions although they may be en-
countered very infrequently. Such conditions may impose considerable de-
mands on the indicator system and treatment processes.
The efficacy of various water treatment processes in removing or inactivat-
ing viruses has recently been reviewed by Chang (6) and also in a Commit-
tee Report, "Engineering Evaluation of Virus Hazards in Water" (8). These
reports indicate that natural "die-off" cannot be relied upon for the elimin-
ation of viruses in water. Laboratory pilot plant studies indicate that com-
bination of coagulation and sand filtration is capable of reducing virus
populations up to 99.7 percent if such treatments are properly carried out
(9). It should be noted, however, that a floe breakthrough, sufficient to
cause a turbidity of as little as 0.5 Turbidity Units, was usually accom-
panied by a virus breakthrough in a pilot plant unit seeded with high doses
of virus (9). Disinfection, however, is the only reliable process by which
water can be made free of virus. In the post, there have been numerous
studies conducted on the chlorination of viruses. Recent work by Liu, et al
(10), has confirmed early observations and has reemphasized two possible
weaknesses in these early reports: (a) the number of virus types studied
was very small, thus generalization on such results is not without pitfalls,
(b) the early chlorination studies were usually conducted with reasonably
pure virus suspensions derived from tissue cultures or animal tissue and
may not represent the physical state of the virus as it exists under natural
conditions (clumped, embedded in protective material, etc.) which would
make the virus much more resistant to disinfectants. Thus, it is imperative
that good clarification processes be used on turbid waters to reduce their
turbidity levels that will ensure effective disinfection. Additionally, Liu's
data show the wide variation in resistance to chlorine exhibited by viruses,
e.g., four minutes were required to inactivate 99.99 percent of a reovirus
population as contrasted to 60 minutes to achieve the same percent inacti-
vation of coxsackievirus.
Virology techniques have not yet been perfected to a point where they
can be used to routinely monitor water for viruses. Considerable progress
on method development, however, has been made in the past decade. The
methods potentially useful include: two-phase polymer separation (11),
membrane filtration (12), adsorption on and elution from chemicals (13,
14, 15), and the gauze pad technique (16) to name a few. From the con-
certed efforts of virus-water laboratories throughout the world, it is hoped
that a simple and effective method will become available for viral examina-
tion of water. In the interim, control laboratories having access to facilities
for vims isolation and identification should be encouraged to use available
procedures for evaluating the occurrence of human enteric viruses in treated
waters.
As noted above, no simple and effective method for the viral examination
of wateT is available at this time. When such a method is developed, and
42
-------
APPENDIX a—DRINKING WATER REGULATIONS
when there are sufficient data to provide the necessary basis, a maximum
contaminant level for virus will be proposed
REFERENCES
1. Mosley, J.W. Transmission of Viral Diseases by Drinking Water. Transmission of
Viruses by the Water Route, edited by G. Berg, John Wiley and Sons, New York,
pp. 5-25, 1968.
2. Viswanathan, R. Epidemiology. Indian J. Med. Res. 45: 1, (1957).
3. Weibel, S.R., Dixon, F.R., Weidner, R.B., and McCabe, LJ. Waterborne-disease
Outbreaks 1946-1970. JAWWA. 56: 947 (1964).
4. Coin, L., Menetriet, M.L., L&boade, J., and Hanivon, M.C. Modem Microbiological
and Virological Aspects of Water Pollution. Second International Conference on
Water Pollution Research, Tokyo, Japan, pp. 1-10. (1964).
5. Mack, N.W., Lu, Y.S., and Coohoon, D.B. Isolation of Poliomyelitis Virus from a
Contaminated Well. Health Services Repts, 87: 271 (1972).
6. Chang, S.L. Waterborne Virus Infections and Their Prevention. Bull. Wld. Hlth.
Org. 38: 401 (1968).
7. Clarke, N.A., Berg, G., Kabler, P.W., and Chang, S.L. Human Enteric Viruses in
Water*. Source, Survival, and Removability, First International Conference on
Water Pollution Research, London, England, pp. 523-542 (1962).
8. Committee Report "Engineering Evaluation of Virus Hazard in Water," JASCE,
SED, pp. 111-161 (1970).
9. Robeck, G.G., Clarke, N.A., and Dostal, K.A. Effectiveness of Water Treatment
Processes in Virus Removal. JAWWA, 54: 1275 (1962).
10. Liu, O.C. Effect of Chlorination on Human Enteric Viruses in Partially Treated
Water from Potomac Estuary. Progress Report. Environmental Protection Agency,
Division of Water Hygiene, 1970.
11. Shuval, H.I., Fattal, B., Cymbalists, S., and Goldblum, N. The Phase-Separation
Method for The Concentration and Detection of Viruses in Water. Water Res.
3: 225 (1969),
12. Rao, N.U. and Labzoffsky, N.A. A Simple Method for Detection of Low Concen-
tration of Viruses in Large Volumes of Water by the Membrane Filter Technique.
Can. J. Microbiol., IS: 399 (1969).
13. Rao, V.C., Sullivan, R., Read, R.B., and Clarke, N.A, A Simple Method for Con-
centrating and Detecting Viruses in Water. JAWWA. 60: 1288 (1968).
14. Wallis, G. and Melnick, J.L. Concentration of Viruses on Aluminum Phosphate and
Aluminum Hydroxide Precipitates. Transmission of Viruses by the Water Route,
Edited by G. Berg, J. Wiley and Sons, pp. 129-138, 1968.
15. Wallis, G,, Gristein, S-, and Melnick, J.L. Concentration of Viruses from Sewage
and Excreta on Insoluble Polyelectrolytes. Appl. Microbiol. 18: 1007 (1969).
16. Hoff, J.C., Lee, R.D., and Becker, R.G. Evaluation of a Method for Concentra-
tion of Microorganisms in Water. APHA Proc. (1967).
T urbidity
Drinking water should be low in turbidity prior to disinfection and at
the consumer's tap for the following reasons:
(1) Several studies have demonstrated that the presence of particulate
matter in water interferes with effective disinfection. Neefe, Baty, Reinhold,
and Stokes (1) added from 40 to 50 ppm of feces containing the causative
agent of infectious hepatitis to distilled water. They then treated this water
43
-------
DRINKING WATER REGULATIONS
by varying techniques and fed the resultant liquid to human volunteers. One
portion of the water that was disinfected to a total chlorine residual after
30 minutes of 1.1 mg/1 caused hepatitis in 2 of the 5 volunteers. A similar
experiment in which the water was first coagulated and then filtered, prior
to disinfection to the same concentration of total residual, produced no
hepatitis in 5 volunteers. This was repeated with 7 additional volunteers,
and again no infectious hepatitis occurred.
Chang, Woodward and Kabler (2) showed that nematode worms can in-
gest enteric bacterial pathogens as well as virus, and that the nematode-
borne organisms are completely protected against chlorinations even when
more than 90 percent of the carrier worms are immobilized.
Walton (3) analyzed data from three waterworks treating surface waters
by chlorination only. Coliform bacteria were detected in the chlorinated
water at only one waterworks, the one that treated a Great Lakes water that
usually did not have turbidities greater than 10 turbidity units (TU), but
occasionally contained turbidities as great as 100 TU.
Sanderson and Kelly (4) studied an impounded water supply receiving
no treatment other than chlorination. The concentration of free chlorine
residual in samples from household taps after a minimum of 30 minutes
contact time varied from 0.1 to 0.5 mg/1 and the total chlorine residual was
between 0.7 and 1 mg/1. These samples consistently yielded confirmed coli-
form organisms. Turbidities in these samples varied from 4 to 84 TU, and
microscopic examination showed iron rust and plankton to be present. They
concluded "... coliform bacteria were imbedded in particles of turbidity
and were probably never in contact with the active agent. Viruses, being
smaller than bacteria, are much more likely to escape the action of chlorine
in a natural water. Thus, it would be essential to treat water by coagulation
and filtration to nearly zero turbidity if chlorination is to be effective as a
viricidal process."
Hudson (5) reanalyzed the data of Walton, above, relating them to the
hepatitis incidence for some of the cities that Walton studied plus a few
others. A summary of his analysis is shown in Table I. Woodward does,
however, in a companion discussion warn against over interpreting such
limited data and urges more field and laboratory research to clearly
demonstrate the facts.
Table I.—Filtered-Water Quality and Hepatitis Incidence, 1953
Final
Average
Chlorine
Turbidity
Residual
Hepatitis
City
TU
mg/1
cases/100,00 people
G
0.15
0.1
3.0
C
0.10
03
4.7
H
0.25
0.3
4.9
B
0.2
—
8.6
M
0.3
0.4
31.0
A
1.0
0.7
130.0
44
-------
APPENDIX A—DRINKING WATER REGULATIONS
Tracey, Camarena, and Wing (6) noted that during 1963, in San Fran-
cisco, California, 33 percent of all the colilorm samples showed S positive
tubes, in spite of the presence of chlorine residual. During the period of
greatest coliform persistence, the turbidity of this unfiltered supply was be-
tween 5 and 10 TU.
Finally, Robeck, Clarke, and Dostal (7) showed by laboratory demonstra-
tion that virus penetration through a granular filter was accompanied by a
breakthrough of floe, as measured by an increase in effluent turbidity above
0.5 turbidity unit in a pilot unit seeded with an extremely high dose of virus.
These 7 studies show the importance of having a low turbidity water prior
to disinfection and entrance into the distribution system.
(2) The 1969 Community Water Supply Survey (8) revealed that un-
pleasant tastes and odors were among the most common customer com-
plaints. While organics and inorganics in finished water do cause tastes
and odors, these problems are often aggravated by the reaction of chlorine
with foreign substances. Maintenance of a low turbidity will permit distribu-
tion with less likelihood of increasing taste and odor problems.
(3) Regrowth of microorganisms in a distribution system is often stimu-
lated if organic matter (food) is present. An example of this possibility
occurred in a Pittsburgh hospital (9). One source of this food is biological
forms such as algae which may contribute to gross turbidity. Therefore, the
maintenance of low turibidity water will reduce the level of this microbial
food and maintain a cleanliness that will help prevent regrowth of bacteria
and the growth of other microorganisms.
(4) The purpose of maintaining a chlorine residual in a distribution
system is to have a biocidal material present throughout the system so that
the consumer will be protected if the integrity of the system is violated. Be-
cause the suspended material that causes turbidity may exert a chlorine de-
mand, the maintenance of a low turbidity water throughout the distribution
system wil facilitate the provision of proper chlorine residual.
For these reasons, the limit for turbidity is one (1) Turbidity Unit (TU)
as the water enters the distribution system. A properly operated water treat-
ment plant employing coagulants and granular filtration should have no dif-
ficulty in consistently producing a finished water conforming to this limit.
REFERENCES
1. Neefe, J.R., Baty, J.B., Reinhold, J.G., and Stokes, J. Inactivation of The Virus of
Infectious Hepatitis in Drinking Water. Am. J. Pub. Health 37: 365 (1947),
2. Chang, SX-, Woodward, R.L., end Kabler, P.K. Survey of Freeliving Nematodes
and Aroebas in Municipal Supplies. JAWWA. 52: 613 (May 1960).
3. Walton, 0. Effectiveness of Water Treatment Processes As Measured by Coliform
Reduction. U.S. Department of Health, Education and Welfare, Public Health
Service, Publ. No. 898, 68 p. (1961).
4. Sanderson, W.W, and Kelly, S. Discuuion of "Human Enteric Viruses in Water:
Source, Survival and Removability" by Clarke. N.A., Berg, G., Kabler, P.K., and
Chang, S.L. Interaat. Conf. on Water Poll. Res., pp. 536-541, London, September
1962, Petgunon Press. (1964>.
45
-------
PRINKING WATER REGULATIONS
5. Hudson, H.E., Jr. High-quality Water Production and Viral Disease. JAWWA.
S4: 1265-1272 (Oct 1962).
6. Tracy, H.W., Cam arena, V.M., and Wing, F. Coliform Persistence in Higlily Chlor-
inated Waters. JAWWA. 58: 1151 (1966).
7. Robeck, G.G., Clarke, N.A., and Dostal, K.A. Effectiveness of Water Treatment
Processes in Virus Removal. JAWWA. 54: 1275-1290 (1962).
8. McCabe, LJ., Symons, J.M., Lee, R.O., and Robeck, G.G. Survey of Community
Water Supply Systems. JAWWA. 62 : 670 (1970).
9. Roueche, B. Annals of Medicine. Three Sick Babies. The New Yorker, Oct. 5, 1968.
46
-------
APPENDIX A—DRINKING WATER REGULATIONS
C—CHEMICAL QUALITY
The following pages present detailed data and the reasoning used in
reaching the various limits.
In general, limits are based on the fact that the substances enumerated
represent hazards to the health of man. In arriving at specific limits, the
total environmental exposure of man to a stated specific toxicant has been
considered. An attempt has been made to set lifetime limits at the lowest
practical level in order to minimize the amount of a toxicant contributed
by water, particularly when other sources such as milk, food, or air are
known to represent the major exposure to man.
The Regulations are regarded as a standard of quality that is generally at-
tainable by good water quality control practices. Poor practice is an in«
herent health hazard. The policy has been to set limits that are not so low as
to be impracticable nor so high as to encourage pollution of water.
No attempt has been made to prescribe specific limits for every toxic or
undesirable contaminant that might enter a public water supply. While the
need for continued attention to chemical contaminants of water is recog-
nized, the Regulations are limited to need and available scientific data or im-
plications on which judgments can be made. Standards for innumerable sub-
stances which are rarely found in water would require an impossible burden
of analytical examination.
The following table indicates the percent of samples analyzed in the Com-
munity Water Supply Study which exceeded 75% of the 1962 PHS Drinking
Water Standards limits. This table shows the relationship of the existing
quality of water analyzed during the study to the drinking water standards
in effect at that time.
Percent or Samples in the Community Water
Supply Study with Value Exceedinc 75% of Each Limit
in the 1962 Drinking Water Standards
Constituent
DWS Limit
DWS Limit X 0.75
Percent of
Samples Exceeding
Arsenic
0.05 mg/1
0.0375 mg/1
1.24%
Barium
1 mg/1
0.75 mg/1
0.08%
Cadmium
0.010 mg/l
0.0075 mg/1
1.45%
Chloride
250 mg/1
187.5 mg/1
1.56%
Chromium
0.05 mg/1
0.0375 mg/1
1.43%
Color
15 C.U.
11.25 C.U.
3.54%
Copper
1 mg/1
0.75 mg/1
2.47%
Cyanide
0J2 mg/1
0.15 mg/1
0.00%
Foaming Agent*
0.5 mg/1
0.375 mg/1
0.08%
Iron
0.3 mg/1
0.225 mg/1
15.81%
Lead
0.05 mg/I
0.0375 mg/1
3.32%
Manganese
0.05 mg/I
0.0375 mg/1
11.91%
Nitrate
45 mg/1
33.75 mg/1
3.46%
Selenium
0.01 mg/1
0.0075 mg/1
8.35%
Silver
0.05 mg/1
0.0375 mg/l
0.00%
Sulfate
250 mg/1
187.5 mg/l
3.37%
Zinc
5 mg/1
3.75 mg/l
0.35%
47
-------
DRINKING WATER REGULATIONS
DAILY FLUID INTAKE
For the purpose of these Regulations, a daily intake of water or water
based fluids of two liters was assumed. This figure was taken as being
representative of the fluid consumption of a normal adult male, and was
obtained by consulting standard textbooks on physiology and numerous
journal articles concerning water consumption.
It was realized that tremendous variation in individual consumption
would exist, but since women and children drink less than the average man,
it was decided that a large percent of the population would consume less
than two liters a day,
There have been numerous reports of individuals or groups of persons
who consume abnormally large quantities of water or waterbased fluids.
For example, the consumption of six liters of beer in a day (1, 2) is not
unknown. However, it should be noted that anyone who consumes this
quantity of beer would be getting more than 240 ml {V2 pint) of pure
alcohol which is close to the maximum tolerable dose for a day.
The Boy's Life Magazine (1971) (3) survey indicated that 8% of 10-17
year-old boys drink more than 8 soft drinks per day. This survey can be
viewed from another angle and a statement made that 92% of such boys
drink leas than 8 soft drinks per day. It would probably be valid to state that
the average consumption is far les than 8.
Guyton (1951) (4) properly indicates that diseased persons having
diabetes insipidus consume great quantities of water a day but even raising
the "daily fluid intake'1 to 6 liters a day would not protect these individuals
who excrete up to 15 or more liters of urine per day. It might also be
pointed out that diabetes insipidus is a relatively rare disease and that these
patients could not be considered average consumers.
Welch, et al (5) show that at temperatures up to 75°F 2 liters or less of
fluid are drunk per day by adult males.
Molnar, et al (6) found that average fluid intake in the desert was 5.90
liters per day with a standard deviation of 42.03 whereas average fluid
intake in the tropics was 3.26 liters with a standard deviation of *1.09.
These men were performing their normal duties including truck driving,
guard duty, hiking, etc. Five percent of the men in the tropics drank as
little as 1 liter a day.
Wyndham and Strydom (7) indicated that marathon runners lost be-
tween 1,500 and 4,200 mi of sweat in 20 miles of running at about 60°F.
To replace their fluid that day would require from 2.5 to 5 liters of water.
In "Clinical Nutrition" (8) the normal water loss per day shown for a
normal adult ranges from 1,500 ml • 2,100 ml. The breakdown for a 2,600
ml water intake is shown as 1,500 from fluids, 800 ml from food and 300
ml from metabolism.
In "Physiology of Man in the Desert" (9) the average intake of fluid for
91 men in the desert was 5.03 liters with a standard deviation of ±1.67.
48
-------
APPENDIX A—DRINKING WATER REGULATIONS
This indicates that some men only drank three liters a day in a desert en-
vironment where temperatures went as high as 105° F.
In Best and Taylor's book, "The Physiological Basis of Medical Practice,"
(1945) (10) an average adult is shown to require 2,500 ml of water from
all sources under ordinary circumstances. The sources of this water are
shown as:
Solid and semisolid food 1200 cc
Oxidation of food 300 cc
Drinks (water, milk, coffee, beer, etc.) 1000 cc
This reference points out that cooked lean meat contains from 65 to 70
percent water.
It should be noted that certain references refer to water loss per day in-
stead of drinking water intake. Water loss per day is approximately 1%
liters higher than the drinking water intake figure would be.
"Human Designs" (11) by Beck (1971) indicates that between 2200 ml
and 2800 ml are required for an average adult with an average 2500 ml
daily fluid intake. This author, however, reverses the food and drink quan-
tities shown above. Both of these references indicate that 1 cc of water is
required per calorie of food intake.
Two articles relating to the fluid intake of children might be cited here.
One, by Galagan, et al (12), used children from under one year of age to
age ten and showed that total fluid intake per pound of body weights was
highest among infants and decreased with age. The water intake listed
average 0.40 ounces (12 ml) per day per pound of body weight. They also
found that water intake increased directly with increases in temperature.
The second article by Bonham, et al (13) concerns six-year old children
and lists 0.70 ounces (21 ml) per day per pound. This is total fluid and in-
cludes milk. If a child of this age weighed 50 lbs., he would drink about one
liter per day.
The "Bioastronautics Data Book" (14) lists an average of 2400 ml total
water intake but indicates the breakdown as 1,500 ml from drinking water,
600 ml from food and 30 ml from oxidation of food.
More recently, the Task Group on Reference Man (1974) (15) estimated
the water-based fluid intake of an adult man to be 1650 ml/day, with cor-
responding values for an adult woman of 1200 ml/day and for a child of
950 ml/day.
Considering all the information we have available, two liters per day
drinking water consumption for the average man should be a reasonable
estimate. It is twice the amount listed by some authors and 30% higher than
other authors list as an average figure and is therefore defensible as a
reference standard.
49
-------
DRINKING WATER REGULATIONS
REFERENCES
1. McDermott P.H., R.L. Delaney, J.D. Egan, and J.F. Sull "Myocardosis and Cardiac
Failure in Man" JAMA 198:253 1966.
2. Morin, Y.L., A.R. Foley, G. Martineau, and J. Roussel Beer-Drinkers Cardiomyo-
pathy: Forty-Eight Cases." Canadian Medical Assoc, J, 97:881-3, 1967.
3. Boy's Life, National Readership Panel Survey, August 1970 Richard Manvtlle
Research, Inc. 1971.
4. Guyton, A.C. Textbook of Medical Physiology, Second Edition Philadelphia, W.S.
Saunders 1961.
5. Welch, B.E., E.R. Buekirk and P.F. lampietro "Relation of Climate and Tempera-
ture to Food and Water Intake" Metabolism 7:141-8, 1958.
6. Molnar, G.W., EJ. Towbin, R.E. Gosselin, A.H. Brown and E.F. Adolph. "A
Comparative Study of Water, Salt and Heat Exchanges of Men in Tropical and
Desert Environments" A J. Hyg 44:411-33, 1946.
7. Wyndhan, C.H. and N.B. Strydom. "The Danger of and Inadequate Water Intake
During Marathon Running." South Afr. MJ. 43:893-6,1969.
8. Joliffe, N. (Editor), Clinical Nutrition, Second Edition, New York, Harper and
Brothers, 1962.
9. Adolph, E.F., Physiology of Man in the Desert, New York, Interscience Publ. 1946.
10. Best, C.H. and H.B. Taylor, The Physiological Basis of Medical Practice, Eighth
Edition, Baltimore, Williams and Wilkins Co. 1966.
11. Beck, W.S., Human Design, New York, Harcourt Brace Jovanovich, 1971.
12. Galagan, DJ., J.R. Vermillion, G.A. Nevitt, Z.M. Stadt, and R.E. Dart. "Climate
and Fluid Intake," Pub. Health Reg. 72'.484-90, 1957.
13. Bonham, G.H. A.S. Gray, and N. Luttrell. "Fluid Intake Patterns of 6-Year-Old
Children in a Northern Fluoridated Community." Canad. Med. Ass. J. 91:749-51,
1964.
14. Webb, P. (Editor). Bioastronautics Data Book, (National Aeronautics and Space
Agency, Washington, D.C.) NASA-SP 3006. 1964.
15. Snyder, W.S., Chairman, "Report of the Task Group on Reference Man." New York,
Pergamon Press, 1974.
50
-------
APPENDIX A—DRINKING WATER REGULATIONS
ARSENIC
The high toxicity of arsenic and its widespread occurrence in the environ-
ment necessitate the setting of a limit on the concentration of arsenic in
drinking water.
The presence of arsenic in nature is due mainly to natural desposits of
the metalloid and to its extensive use as a pesticidal agent. Arsenic concen-
trations in soils range from less than one part per million (mg/1) to several
hundred mg/1 in those areas where arsenical sprays have been used for
years. Despite relatively high concentrations of arsenic in soils, plants rarely
take up enough of the element to constitute a risk to human health (1, 2).
Despite the diminishing use of arsenicals as pesticides, presently several ar-
senites are used as herbicides and some arsenates as insecticides. In 1964,
farmers in the U.S. used a combined total of approximately 15 million
pounds of arsenicals (3).
The chemical forms of arsenic consist of trivalent and pentavalent inor-
ganic compounds and trivalent and pentavalent organic agents. It is not
known which forms of arsenic occur in the drinking water. Although com-
binations of all forms are possible, it can be reasonably assumed that the
pentavalent inorganic form is the most prevalent. Conditions that favor
chemical and biological oxidation promote the shift to the pentavalent
specie; and conversely, those that favor reduction will shift the equilibrium
to the trivalent state.
The population is exposed to arsenic in a number of ways. Arsenic is
still used, albeit infrequently, to treat leukemia, certain types of anemia,
and certain skin diseases (4). In the diet, vegetables and grain contain an
average of 0.44 ppm and meats an average of 0.5 ppm of arsenic (5).
Organic arsenicals are deliberately introduced into the diet of poultry and
pigs as growth stimulators and pesticides. The Food and Drug Administra-
tion has set tolerance limits for residues of arsenicals on fruits and vege-
tables (3.5 mg as As203 per kg) and in meat (0.5 to 2.0 mg as As per kg)
(6). Shellfish are the dietary components that usually contain the highest
concentrations of arsenic, up to 170 mg/kg (2, 7, 8),
For the entire U.S., the arsenic concentrations in air range from a trace
to 0.75 ug/m3 (9). Airborne arsenic is usually the result of operating cotton
gins, manufacturing arsenicals, and burning coal.
Arsenic content of drinking water ranges from a trace in most U.S. sup-
plies to approximately 0.1 mg/1 (10). No adverse health effects have been
reported from the ingestion of water containing 0.1 mg/l of arsenic.
The toxicity of arsenic is well known, and the ingestion of as little as 100
mg can result in severe poisoning. In general, inorganic arsenicals are more
toxic to man and experimental animals than the organic analogs; and ar-
senic in the pentavalent state is less toxic than that in the trivalent form.
Inorganic arsenic is absorbed readily from the gaatro-intestinal tract, the
lungs, and to a lesser extent from the akin, and becomes distributed through-
51
-------
DRINKING WATER REGULATIONS
out the body tissues and fluids (4). Inorganic arsenicals appear to be slowly
oxidized in vivo from the trivalent to the pentavalent state; however, there
is no evidence that the reduction of pentavalent arsenic occurs within the
body (5, 11-13). Inorganic arsenicals are potent inhibitors of the intra-
cellular sulfhydryl (-SH) enzymes involved in cellular oxidations (14).
Arsenic is excreted via urine, feces, sweat, and the epithelium of the skin
(15-20). A single dose is usually excreted largely in the urine during the
first 24 to 48 hours after administration; but elimination of the remainder
of the dose continues for 7 to 10 days thereafter. During chronic exposure
arsenic accumulates mainly in bone, muscle, and skin, and to a smaller
degree in liver and kidneys. After cessation of continuous exposure, arsenic
excretion may last up to 70 days (14).
A number of chronic oral toxicity studies with inorganic arsenite and
arensate (21-25) demonstrated the minimum-effect and no-effect levels in
dogs, rats, and mice. Three generations of breeding mice were exposed to
5 ppm of arsenite in the diet with no observable effects on reproduction. At
high doses (i.e., 200 mg/1 or greater) arsenic is a physiological antagonist
of thyroid hormones in the rat (26). Arsenic is also an antagonist of selen-
ium and has been reported to counteract the toxicity of seleniferous foods
when added to argicultural animals' feed water (27, 28). Rats fed shrimp
meat containing a high concentration of arsenic retain very little of the
dement as compared to rats fed the same concentrations of either arsenic
trioxide or calcium arsenate (29), suggesting that the arsenic in shellfish
tissues may be less toxic to mammals than that ingested in other forms.
In man, subacute and chronic arsenic poisoning may be insidious and
pernicious. In mild chronic poisoning, the only symptoms present are fa-
tigue and loss of energy. The following symptoms may be observed in more
severe intoxication: gastrointestinal catarrh, kidney degeneration, tendency
to edema, polyneuritis, liver cirrhosis, bone marrow injury, and exfoliate
dermatitis (30, 31). In 1962, thirty-two school-age children developed a
dermatosis associated with cutaneous exposure to arsenic trioxide (32, 33).
It has been claimed that individuals become tolerant to arsenic. However,
this apparent effect is probably due to the ingestion of the relatively in-
soluble, coarse powder, since no true tolerance has ever been demonstrated
(14).
Since the early nineteenth century, arsenic was believed to be a carcino-
gen; however, evidence from animal experiments and human experience has
accumulated to strongly suggest that arsenicals do not produce cancer. One
exception is a report from Taiwan showing a dose-response curve relating
skin cancer incidence to the arsenic content of drinking water (44). Some
reports incriminated arsenic as a carcinogen (34, 35), but it was later
learned that agents other than the metalloid were responsible for such can-
cers (36). Sommers and McManua (37) reported several cases of cancer
in individual! who had at some time in their lives been exposed to thera-
52
-------
APPENDIX A—DRINKING WATER REGULATIONS
peutic doses of arsenic trioxide (usually in Fowler's Solution). Patients dis-
played characteristic arsenic keratosis, but there was no direct evidence that
arsenic was the etiologic agent in the production of the carcinoma.
Properly controlled studies (38, 39) have demonstrated that industrial
workers do not have an increased prevalence of cancer despite continued
exposure to high concentrations of arsenic trioxide. In the study by Pinto
and Bennett (39), the exposure was estimated by comparing the arsenic ex-
creted in urine of control and exposed populations. In the experimental
group, some workers who had been exposed to arsenic trioxide for up to 40
years, excreted 0.82 mg of arsenic per liter, or more than six times the con-
centration of the control population. In addition, attempts to demonstrate
through animal studies that arsenic is tumorigenic have met with failure
(23, 35, 40-42). The possible co-carcinogenic role of arsenic trioxide in the
production of methycholanthrene-induced skin tumors has been investigated
and found to have no significant effect (43).
However, some recent evidence supports the view that arsenic is cardio-
genic. Industrial workers in a plant manufacturing arsenic powder were
exposed to arsenic dust and showed a higher incidence of skin and lung can-
cer than other occupational groups (44, 45, 46). Ulceration of the nasal sep-
tum appears to be a common finding among workers exposed to inorganic
arsenic. The incidence of skin cancer has also been reported to be unusually
high in areas of England where arsenic was present in drinking water at a
level of 12 mg/1 (47). More recently Lee and Fraumeni found that the mor-
tality rate of white male smelter workers exposed to both arsenic trioxide
and sulfur dioxide exceeded the expected mortality rate by a statistically
significant margin and found that lung cancer deaths among these workers
increased with increasing lengths of exposure to arsenic trioxide. They
concluded that their findings were "consistent with the hypothesis that ex-
posure to high levels of arsenic trioxide, perhaps in interaction with sulphur
dioxide or unidentified chemicals in the work environment, is responsible
for the three-fold excess of respiratory cancer deaths among smelter
workers" (48),
Similarly, Ott, et al., found, in a study for the Dow Chemical Company,
that exposed employees in a dry arsenical manufacturing plant experienced
a three-fold increase in lung cancer over the rate for non-exposed employees
(49).
Baetjer, et al., in a study for the Allied Chemical Company, found that
19 of the 27 deaths occuring in this population between 1960 and 1972 were
due to cancer as compared to an expected number, based on figures adjusted
for age, race, and sex, of 7.3 cancer-related deaths (50).
Additional medical problems relating to arsenic content of drinking water
have been reported from several other countries. Several epidemiological
studies in Taiwan (51-55) have reported the correlation between increased
incidence of hyperkeratosis and skin cancer with the consumption of water
53
-------
DRINKING WATER REGULATIONS
with arsenic content higher than 0.3 mg/I. A similar problem has been re-
ported in Argentina (56-58). Dermatological manifestations of arsenicism
were noted in children of Antofagasto, Chile, who used a water supply with
0.8 rag/1. A new water supply was provided, and preliminary data show that
arsenic levels of hair have decreased, and further study will be made of the
health of persons born since the change in supply (59). Arsenicism affect-
ing two members of a family where the arsenic content of the family's well
varied between 0.5 and 2.75 mg/l over a period of several months* was re-
ported in Nevada (60). A study in California found that a greater propor-
tion of the population had elevated concentrations of arsenic in the hair
when the drinking water had more than 0.12 mg/l than when it was below
this concentration, but illness was not noted (61). In none of the cited inci-
dents of apparent correlation of arsenic in drinking water with increased
incidence of hyperkeratosis and skin cancer has there been any confirmed
evidence that arsenic was the etiologic agent in the production of
carcinomas.
Arsenic is a geochemical pollutant, and when it occurs in an area it can
be expected to be tn the air, food, and water, but in other cases it is due to
industrial pollution. In some epidemiological studies it is difficult to deter-
mine which exposure is the greater problem. A recent study (62) of metal-
lic air pollutants showed that arsenic levels of hair were related to exposure
from this source, but other exposures were not quantitated. The Taiwan
studies were able to compare quite similar populations that differed only in
the water intake. Deep wells contained arsenic, but persons using shallow
wells were not exposed.
The change in water supply in Chile provided a unique experience to
demonstrate the effect of arsenic in drinking water in spite of other arsenic
exposures.
It is estimated that the total intake of arsenic from food is an average of
900 ug/day (5). At a concentration of 0.05 mg per liter and an average in-
take of 2 liters of water per day, the intake from water would not exceed 100
ug per day, or approximately 10 percent of the total ingested arsenic.
In light of our present knowledge concerning the potential health hazard
from the ingestion of arsenic, the concentration of arsenic in the drinking
water shall not exceed 0.05 mg/L
REFERENCES
1. Underwood, EJ., Trace Elements in Human and Animal Nutrition. New York;
Academic Press, Inc., 1956, pp. 372-364.
2. Monier-WUHftms, G.W.: Trace Elements in Food. New York: John Wiley & Sons,
Inc., 1949, pp. 162-206.
3. Quantities of Pesticides Used by Fanners in 1964. Agriculture Economic Report
No. 131, Economic Research Service, U.S. Department of Agriculture, 1968.
4. Salimaa, T. (ed.) in A Manual of Pharmacology and its Applications to Thera-
peutics and Toxicology. Philadelphia: W,B. Saunders Co., 1957.
54
-------
APPENDIX A—DRINKING WATER REGULATIONS
5. Schroeder, H.A., and Balaesa, J. J. Abnormal Trace MetalB in Man. J. Chron. Dis.
19, 85-106, 1966.
6. Code of Federal Regulations, Title 21, Sections 120. 192/3/5/6 and 133g. 33.
7. Coulson, EJ., Remington, R.E., and Lynch, K.M. Metabolism in the Rate of the
Naturally Occurring Arsenic of Shrimp a* Compared with Arsenic Trioxide, J. Nu-
trition, 10, 255*270, 1935.
8. Ellis, M.M., Westfall, B.A., and Ellis, M.D, Arsenic in Freshwater Fish. Indust.
and Engineer Chem., 33, 1331-1332, 1941 (Experimental Station Report 87. p. 740,
1941).
9. Air Pollution Measurements of the National Air Sampling Network • Analyses of
Suspended Particulates 1963, U.S. Dept. of Health, Education, and Welfare, Public
Health Service, Cincinnati, Ohio, 1965.
10. McCabe, LJ., Symons, J.M., Lee, R.D., and Robeck, G.G. Survey of Community
Water Supply Systems, JAWWA, 62, (11), 670-687, 1970,
11. Overby, L.R., and Fredrickson, RJL, /. Agr. Food Ckem., 11, 78 ,1963.
12. Peoples, S.A. Ann, N.Y. Acad. Set., Ill, 644, 1964.
13. Winkler, W.O. /. Assoc. Offic. Agr. Chemists, 45, 80, 1962.
14. DuBois, K.P. and Ceiling, E.M.K. Textbook of Toxicology. New York, N. Y., Oxford
University Press, 1959, pp. 132-135.
15. Hunter, F.T., Kip, A.F., and Irvine, J.W. Radiotracer Studies on Arsenic Injected as
Potassium Arsenite: I. Excretion and Localization of Tissues. J. Pharmacol. Exper.
Therap., 76 207-220, 1942.
16. Lowry, O.H., Hunter, F.T., Kip, A.F., and Irvine, J.W. Radiotracer Studies on
Arsenic Injected as Potassium Arsenite: II. Chemical Distribution in Tissues.
J. Pharmacol. Exper. Therap. 76, 221-225, 1942.
17. Dupont, O. Ariel, I., and Warren, S.L. The Distribution of Radioactive Arsenic in
Normal and Tumor-Rearing Rabbits, Am. /. Syph. 26, 96*118, 1942.
18. Duncoff, H.S., Neal W.B., Straube, R.L., Jacobaon, L. O., and Brues, A.M. Biologi*
cal Studies with Arsenic: II. Excretion and Tissue Localization. Proc. Soc. Exper.
Biol. Med. 69, 548, 1948.
19. Musil, J. and Dejmal, V. Experimental and Qinical Administration of Radio-
arsenic. Casopis lek. cesk. 96, 1543-6, 1957; Chem. Abstr. 14008, 1958.
20. Crema, A. Distribution et elimination de {'arsenic 76 chez la souris norm ale et
cancereuse. Arch, Internal. Pharmaeodyn. 103, 57-70, 1955.
21. Sollman, 1921. Cited in Sollmann T. (ed.) in a Manual of Pharmacology and Ita
Application to Therapeutics and Toxicology. Philadelphia: W.B. Saunders Co.,
1948, p. 874.
22. Schroeder, H.A. and Balaesa, J J. Arsenic, Germanium, and Tin in Mice. J. Nu-
trition, 92, 245, 1967.
23. Kanisawa, M. and Schroeder, H.A. Life Term Studies on the Effects of Arsenic,
Germanium, Tin, and Vanadium on Spontaneous Tumors in Mice. Cancer Jle-
search 27, 1192, 1967.
24. Schroeder, H.A., Kanisawa, M., Frost, D.V., and Mitchener, M.N. Nutr. 96, 37
1968.
25. Byron, W.R., Bierbower, G.W., Brouwer, J.B., and Hansen, W.H. Tox. Appl.
Pharmacol, 10 (1): 132-147,1967.
26. Hesse, E. Klin. Wehnschr. 12,1060, 1933.
27. DuBois, K.P., Moxon, A.L., and Olson, O.E, Further Studies on the Effectiveness
of Arsenic in Preventing Selenium Poisoning. J. Nutrition 19, 477-482, 1940.
28. Moxon, A.L. The Influence of Arensic on Selenium Poisoning in Hogs, in Pro*
ceedings of the South Dakota Academy of Sciences, 1941, vol. 21, pp. 34-36.
29. Calvary, H.O. Chronic Effects of Ingested Lead and Arsenic. JAMA. Ill, 1722-
1729, 1938.
55
-------
DRINKING WATER REGULATIONS
30. Goodman, L.S. and Gilman, A. (eds.) The Pharmacological Basis of Therapeutics,
3rd Edition. New York, N.Y., The MacMilian Co., 1965, pp. 944-951.
31. DiPalma, J.R. Drill's Pharmacology in Medicine, 3rd Edition. New York, N.Y.,
McGraw-Hill Book Company, 1965, pp. 860-862.
32. Birmingham, D.J., Key, M.M., and Holaday, D.A. An Outbreak of Dermatoses in a
Mining Community - Report of Environmental and Medical Surveys. U.S. Dept.
of Health, Education, and Welfare, TR-12, April, 1954.
33. Birmingham, D.J., Key, M.M., Holaday, D.A., and Perone, V.B. An Outbreak of
Arsenical Dermatosis in a Mining Community. Arch. Dermatol. 91, 457, 1965.
34. Paris, J.A. Pharmacologia: Comprehending the Art of Prescribing upon Fixed
and Scientific Principles Together With The History of Medicinal Substances, 3rd
Edit., p. 132, London: Philips, 1820.
35. Buchanan, W.D. Toxicity of Arsenic Compounds. New Jersey: Van Nostrand, 1962.
36. Frost, D.V. Aracnicals in Biology • Retioapect and Prospect. Federation Proceedings
26, 184, 1967.
37. Sommers, S.C. and McManus, R.G. Multiple Arsenical Cancers of Skin and In-
ternal Organs. Cancer 6, 347-359, 1953.
38. Snegireff, L.S., and Lombard, O.M. Arsenic and Cancer. Arch. Industr. Hyg.
Occupational Med. 4, 199, 19S1.
39. Pinto, S.S., and Bennett, B.M. Effect of Arsenic Trioxide Exposure on Mortality.
Arch. Environ. Health 7, 583, 1963.
40. Baroni, C., Van Each, G.J., and Saffiotti, U. Carcinogenesis Tests of Two Inor-
ganic Arsenical*. Arch. Environ. Health 7, 688, 1963.
41. Boutwell, R.K. J. Agr. Food Chem. 11, 381, 1963.
42. Heuper, W.G., and Payne, W.W. Arch Environ. Health 5,445, 1962.
43. Milner, J.E. The Effect of Ingested Arsenic on Methylcholanthrene-Induced Skin
Tumors in Mice. Arch. Environ. Health, 18, 7-11, 1969.
44. Hill, A.B., Faning, E.L., Perry, F., Bowler, R.G., Bucknell, H.M., Druett, H.A., and
Schilling, R.S.F. Studies in the Incidence of Cancer in a Factory Handling Inor-
ganic Compounds of Arsenic. Brit. J. Indust. Med. 5: 1 (1948).
45. Doll, R. Occupational Lung Cancer: A Review. Brit. J. Indus. Med, 16: 181 (1959).
46. Merewether, E. R. A. Industrial Medicine and Hygiene. Vol. 3, Butterworth & Co.,
London, pp. 196-205 (1956).
47 Neubauer, 0. Arsenical Cancer: A Review. Brit, J. Cancer 1: 192 (1947).
48. Lee, A.M. and Fraumeni, J.F-, Jr. Arsenic and Respiratory Cancer in Man—an
Occupational Study. J. Natl. Cancer Inst. 42: 1045 (1969).
49. Ott, M., Holder, B., Gorden, H. Respiratory Cancer and Occupational Exposure to
Arsenicals. to be published in Archives of Environmental Health, Cited In: Federal
Register, 40FR Pt. 3, p. 3395, January 21, 1975.
50. Baetjer, A., Levin, M. Lillenfeid, A. Analysis of Mortality Experience of Allied
Chemical Plant Cited In: Federal Register, 40FR Part 3, p. 3395, January 21, 1975.
51. Tseng, W.P., Chu, H.M., How, S.W., Fong, J.M., Lin, C.S., and Yeh, S. Prevalence
of Skin Cancer in an Endemic Area of Chronic Arsenicism in Taiwan, /, iVat.
Cancer Inst^ 40, 454, 1968.
52. Chen, K.P., Wu, H., and Wu, T. Epidemiologic Studies on Blackfoot Disease in
Taiwan: 3. Physiochemicai Characteristics of Drinking Water in Endemic Black-
foot Disease Areas. Memoirs of the College of Medicine, National Taiwan Uni-
versity, Vol. Ill, No. 1, 2, pp. 116-129,1962.
53. Wu, H., and Chen, K. Epidemiologic Studies on Blackfoot Disease: 1. Prevalence
and Incidence of the Disease by Age, Sex, Year, Occupation, and Geographic Dis-
tribution. Memoirs of the College of Medicine, National Taiwan University, Vol.
Ill, No. 1, pp 33-50, 1961.
56
-------
APPENDIX A—DRINKING WATER REGULATIONS
54. Yeh, S., How., S.W., and Lin, C.S. Arsenical Cancer of the Skin. Cancer 21, 312-
339, 1968.
55. Chen. K., and Wu, H. Epidemiologic Studies on Blackfoot Disease: 2. A Study of
Source of Drinking Water in Relation to the Disease. /. Formosan Med. Assoc. 61,
(7), 611-617, 1962.
56. Arquello, R.A., Cenget, D.D., and Telo, E.E. Cancer y arsenicismo regional en-
demico el Cordoba. Rev. argent. desmoltosif, 22, 461-487 (1938).
57. Bergoglio, R.M. Mortalidad por cancer enzonaa de aquas anenicales dela Provincia
de Cordoba, Republica Argentina. Prensa. Med. Agent, 51, 99-998 (1964).
58. Trelleg, Larghi and Daiz. El problems sanitarro de las aquas destinadas a la bebida
humane con contenidos elevadea de arienico, vanadio, y flos, Saneamienta. Jan-
March 1970.
59. Borgono, J.M., and Greiber, R. Epidemiological Study of Arsenicism in the City of
Antofagasto. Proceedings of the University of Missouri's 5th Annual Conference on
Trace Substances in Environmental Health (in press).
60. Craun, G. and McCabe, LJ. Waterborne Disease Outbreaks, 1961-1970. Presented
at the Annual Meeting of the American Water Work* Association, June 1971.
61. Goldsmith, J.R., Deane, M., Thorn, J., and Gentry, G., Evaluation of Health Im-
plications of Elevated Arsenic in Well Water. Water Research, 6, 1133-1136, 1972.
62. Hammer, D.I., Finklea, J.F., Hendricks, R.H., Shydral, C.M., and Horton, RJ.M.
Hair Trace Metals Levels and Environmental Exposure. Am. Jour, of Epidemiology,
93, 84-92 (1971).
57
-------
DRINKING WATER REGULATIONS
BARIUM
Barium is recognized as a general muscle stimulant, including especially
the heart muscle (1). The fatal dose for man is considered to be from
0.8-0.9 g as the chloride (550-600 mg Ba). Most fatalities have occurred
from mistaken use of barium salts incorporated in rat poison. Barium
is capable of causing nerve block (2) and in small or moderate doses pro-
duced transient increase in blood pressure by vasoconstriction (3). Aspi-
rated barium sulfate has been reported to result in granuloma of the lung
(4) and other sites in man (5). Thus, evidence exists for high acute toxicity
of ingested soluble barium salts, and for chronic irreversible changes in
tissues resulting from the actual despostion of insoluble forms of barium in
sufficient amounts at a localized site. On the other hand, the recent litera-
ture reports no accumulation of barium in bone, muscle, or kidney from
experimentally administered barium salts in animals (6). Most of the ad-
ministered dose appeared in the liver with far lesser amounts in the lungs
and spleen. This substantiates the prior finding of no measurable amounts
of barium in bones or soft tissues of man (7). Later, more accurate analysis
of human bone (British) showed 7 ug Ba/g ashed sample (8), but no in-
crease in bone barium occurred from birth to death. Small amounts of
barium have been shown to go to the skeleton of animals when tracer
amounts of barium-140 were used (9), but no determinations of barium
have been made in animals to which barium had been repeatedly adminis-
tered for long periods.
No study appears to have been made of the amounts of barium that may
be tolerated in drinking water or of effects from prolonged feeding of
barium salts from which an acceptable water guideline may be set. A ra-
tional basis for a water guideline may be derived from the threshold limit of
0.5 mg Ba/m3 air set by the American Conference of Governmental Indus-
trial Hygienists (10) by procedures that have been discussed (11). By as-
suming that 75% of the barium inhaled is absorbed into the blood stream
and that 90% is a reasonable factor for absorption via the gastrointestinal
tract, a value of 2 mg/1 can be derived as an approximate limiting concen-
tration for a healthy adult population. The introduction of a safety factor to
account for heterogeneous populations results in the derivation of lmg/1 as
a limit that should constitute a "no effect" level in water. Because of the
seriousness of the toxic effects of barium on the heart, blood vessels, and
nerves, drinking water shall not contain barium in a concentration exceed-
ing lmg/1.
REFERENCES
1. Sollman, T.H. (Ed.) A Manual of Pharmacology. W.B. Saunders Co., Philadelphia,
pp. 665-667 (1957).
2. Lorente de No, R., and Feng, T.P. Analysis of Effect of Barium upon Nerve with
Particular Reference to Rhythmic Activity. J. Cell Comp. Physiol. 28: 397 (1946).
3. Gotsev. T. Blutdruck und Herztatigkeit. Ill Mitteilung: Kreislaufwirkung von
Barium. Naunyn Schiedeberg Arch. Exper. Path. 203: 264 (1944).
58
-------
APPENDIX a—DRINKING WATER REGULATIONS
4. Fite, F. Granuloma of Lung Due to Radiographic Contrast Medium. AMA Arch.
Path. 59: 673 (1955).
5. Kay S. Tissue Reaction to Barium Sulfate Contrast Medium. AMA Arch. Path. 57:
279 (1954). Ibid: Kay 5., and Chay. Sun Hak: Results of Intraperitoneal Injection
of Barium Sulfate Contrast Medium 59: 388 (1955).
6. Arnolt, R.I. Fijacion y determinacion quimica del bario en organos. Rev. Col.
Farm. Nac. (Rosario) 7 : 75 (1940).
7. Gerlach, W., and Muller, R. Occurrence of Strontium and Barium in Human Or-
gans and Excreta. Arch. Path. Anat. (Virchows) 294 : 210 (1934).
8. Sowden, W., and Stitch, S.R. Trace Elements in Human Tissue. Estimation of the
Concentrations of Stable Strontium and Barium in Human Bone. Biochem. J. 67:
104 (1957).
9. Bauer, G.C.H., Carlsson, A., and Lindquist, B. A Comparative Study of Metabolism
of 140 Ba and 45 Ca in Rats. Biochem. J. 63: 535 (1956).
10. American Conference of Governmental Industrial Hygienists. Theshold Limit Values
of 1958. A.M.A. Arch. Indust. Health 18: 178 (1958).
U. Stokinger, H.E., and Woodward, R.L. Toxicologic Methods for Establishing Drink-
ing Water Standards. JAWWA 50 : 515 (1958).
CADMIUM
As far as is known, cadmium is biologically a nonessential, non-beneficial
element of high toxic potential. Evidence for the serious toxic potential of
cadmium is provided by: (a) poisoning from cadmium-contaminated food
(X) and beverages (2); (b) epidemiologic evidence that cadmium may be
associated with renal arterial hypertension under certain conditions (3);
(c| epidemiologic association of cadmium with "Itai-itai" disease in Japan
(4); and (d) long-term oral toxicity studies in animals.
The possibility of cadmium being a water contaminant has been reported
in 1954 ( 5); seepage of cadmium into ground water from electroplating
plants has resulted in cadmium concentrations ranging from 0.01 to 3.2
mg/1. Other sources of cadmium contamination in water arise from zinc*
galvanized iron in which cadmium is a contaminant. The average concen-
tration of cadmium in drinking water from community supplies is 1.3 ug per
liter in the United States. Slight amounts are common, with 63 percent of
samples taken at household taps showing 1 ug per liter or more. Only 0.3
percent of tap samples would be expected to exceed the limits of 10 ug per
liter (6).
Several instances have been reported of poisoning from eating substances
contaminated with cadmium. A group of school children were made ill by
mating popsicles containing 13 to 15 mg/1 cadmium (1). This is commonly
considered the emetic threshold concentration for cadmium. It has been
stated (7) that the concentration and not the absolute amount determines
the acute cadmium toxicity; equivalent concentrations of cadmium in water
*re likewise considered more toxic than equivalent concentrations in food
Probably because of the antagonistic effect of components in the food.
Chronic oral toxicity studies in rats, in which cadmium chloride was
added to various diets at levels of 15, 45, 75, and 135 ppm cadmium,
showed marked anemia, retarded growth, and in many instances death at the
59
-------
DRINKING WATER REGULATIONS
135 ppm level. At lower cadmium levels, anemia developed later; only one
cadmium-fed animai had marked anemia at the 15 ppm level. Bleaching of
the incisor teeth occurred in rats at all levels, except in some animals at 15
ppm. A low protein diet increased cadmium toxicity. A maximal "no ef-
fect" level was thus not established in the above studies (8). A dietary
relation to cadmium toxicity has been reported by others (9).
Fifty mg/1 of cadmium administered as cadmium chloride in food and
drinking water to rats resulted in a reduction of blood hemoglobin and less-
ened dental pigmentation. Cadmium did not decrease experimental caries
(10).
In a study specifically designed to determine the effects of drinking water
contaminated with cadmium, five groups of rats were exposed to drinking
water containing levels from 0.1 to 10 mg/1. Although no effects of cadmium
toxicity were noted, the content of cadmium in the kidney and liver in-
creased in direct proportion to the dose at all levels including 0.1 mg/1. At
the end of one year, tissue concentrations approximately doubled those at
six months. Toxic effects were evident in a three-month study at 50 mg/l
(11). Later work has confirmed the virtual absence of turnover of absorbed
cadmium (12). More recently, the accumulation of cadmium in renal and
hepatic tissue with age has been documented in man (13).
Recent epidemiological evidence strongly suggests that cadmium ingestion
is associated with a disease syndrome referred to as "Itai-itai" in Japan (4).
The disease syndrome is characterized by decalcification of bones, pro-
teinuria, glycosuria and increased serum alkaline phosphatase, and other
more subjective symptoms. Similar clinical manifestations have been noted
in cadmium workers (14). Yamagatta and Shigematsu (15) have estimated
the current daily intake of cadmium in an endemic "Itai-itai** area as
600 ug. The authors from a geological and topographical survey as well as
knowledge of local customs, concluded that the daily cadmium intake on
the endemic area was probably higher in the past. They concluded that
600 ug per day would not cause 'itai-itai" disease. The average ingestion of
cadmium is 59 ug/day in non-polluted areas of Japan.
The association of cardiovascular disease, particularly hypertension, with
ingestion of cadmium remains unsettled. Conflicting evidence has been
found both in man (3, 16) and in anmials (17, 18). It is notable that hyper-
tension has not been associated with 'itai-itai" disease (19).
The main sources of cadmium exposure in the United States to the gen-
eral population appear to be the diet and cigarette smoking. R.E. Duggan
and P.E. Corneliussen (20) of the FDA in a market basket survey of five
geographic regions in the U.S. found the "daily intake" of cadmium to be
50 ug in 1969 and 30 ug in 1970. Each market basket represented a 2-week
diet constructed for a 16-19 year-old male. Murthy and associates found the
cadmium intake of children to be 92 ug per day from a study of institutional
diets (21). Other estimates are also generally higher than FDA's — ranging
60
-------
APPENDIX A—DRINKING WATER REGULATIONS
from 67 to 200 ug/day. A review of these data suggest 75 ug is a reason-
able estimate of average daily dietary intake (22, 23, 24, 25).
Cigarette smoking has also been shown to be important. Twenty cigarettes
per day will probably cause the inhalation of 2-4 ug of cadmium (26). How-
ever, the absorption rate associated with cigarette smoke inhalation is much
larger than that associated with food ingestion. Lewis (27) has shown in
autopsy studies that men who smoke one or more packages of cigarettes per
day have a mean cadmium concentration in the renal cortex (wet weight)
double the level in a control group of non-smokers. Hammer (24) in similar
studies also found renal wet weight concentrations for those smoking lx/2 or
more packages of cigarettes per day to be more than twice as high as for
non-smokers. In terms of effective body burden, then, cigarette smoking
may double the level derived from food intake alone.
Exactly what exposure to cadmium will cause proteinuria, the earliest
manifestation of chronic cadmium poisoning, is unknown. From animal ex-
periments and very limited human observation in cases of industrial ex-
posure, it is believed that a cadmium level of 200 ppm wet weight in the
renal cortex will be associated with proteinuria. (However, it should be
noted that in one case a level of 446 ppm was found by Axelsson and Pis*
cator without proteinuria) (29). It has been estimated that with 5% gas-
trointenstinal absorption, rapiod excretion of 10% of the absorbed dose, and
0.05% daily excretion of the total body burden, it would take 50 years with
a daily ingestion of 352 ug of Cd to attain the critical level of 200 ppm wet
weight in the renal cortex. The percentage absorption in man is unknown. If
the gastrointestinal absorption of cadmium in man really is about 3%, it
would probably take about 500-600 ug ingested per day to cause proteinuria.
Concentration of cadmium shall be limited to 0.010 mg/1 in drinking
water. At this level it would contribute 20 ug per day to the diet of a person
ingesting 2 liters of water per day. Added to an assumed diet of 75 ug/day,
this would provide about a four-fold safety factor. This does not, however,
take cigarette smoking into account.
REFERENCES
1. Frant, S., and Kieeman, I. Cadmium "Food Poisoning" J.A.M.A., 117, 86 (1941).
2. Cangelosi, J.T. Acute Cadmium Metal Poisoning. U.W. Nav. Med. Bull., pp. 39
and 408 (1941).
3. Schroeder, H.A. Cadmium as a Factor in Hypertension, J. Chron. Dis. 18, 647-656
(1965).
4. Murata, I., Hirono, T., Saeki, Y., and Nakagawa, S. Cadmium Enteropathy, Renal
Osteomalacia ("Itsi-itai" Disease in Japan). Bull. Soc. Int. Chir. 1, 34-42 (1970).
5. Lieber, M., and Welsch, W.F. Contamination of Ground Water by Cadmium.
J.A.W.W.A., 46, p. 51 (1954).
6. McCabe, L.J., Problem of Trace Metals in Water Supply. Proceedings of 16th An-
nual Sanitary Engineering Conference, University of Illinois (1974).
7. Potts, A.M., Simon, F.P., Tobias, J.M., Postel, S., Swift, M.N., Patt, J.M., and
Gerlad, R.W. Distribution and Fate of Cadmium in the Body. Arch. Ind. Hyg. 2,
p. 175 (1950).
61
-------
DRINKING WATER REGULATIONS
8. Fitzhugh, O.G., and Meiller, FJ. Chronic Toxicity of Cadmium. J. Pharm. 72 p.
15 (1941).
9. Wilson, R.H., and De Eds, F. Importance of Diet in Studies of Chronic Toxicity.
Arch. Ind. Hyg. 1, p. 73 (1950).
10. Ginn, J.T., and Volker, J.F. Effect of Cd and F on Rat Dentition. Proc. Soc. Exptl.
Biol. Med. 57, p. 189 (1944).
11. Decker, L.E., Byerrum, R.U., Decker, C.F., Hoppert, C.A., and Langham, R.F.
Chronic Toxicity Studies, I. Cadmium Administered in Drinking Water to Rats.
A.M.A. Arch. Ind. Health, 18, p. 228 (1958).
12. Cotzias, G.C., Borg, D.C., and Seleck, B. Virtual Absence of Turnover in Cadmium
Metabolism: Cd Studies in the Mouse. J. Physiol. 201, 927-930 (1961).
13. Schroeder, H.A., Balassa, J J., and Hogencamp, J.C. Abnormal Trace Metals in
Man: Cadmium. J. Chron. Dis. 14, 236-258 (1961).
14. Piscator, M. Proteinuria in Chronic Cadmium Poisoning. I. An Electrophoretic and
Chemical Study of Urinary and Serum Proteins from Workers with Chronic Cad-
mium Poisoning. Arch. Environ. Health 4, 607-621 (1962).
15. Yamagata, N., and Shigematsu, I. Cadmium Pollution in Perspective. Bui. Inst.
Public Health 19, 1-27 (1970).
16. Morgan, J.M. Tissue Cadmium Concentration in Man. Arch. Intern. Med. 123,
405-408 (1969).
17. Kanisawa, M. and Schroeder, J.A. Renal Arteriolar Changes in Hypertensive Rats
Given Cadmium in Drinking Water. Exp. & Mole. Path. 10, 81-98 (1969).
18. Lener, J. and Bibr, B. Cadmium Content in Some Foodstuffs In Respect of Its
Biological Effects. Vitalstoffe Zivilisationsdrankheiten 15, 139-141 (1970).
19. Nogawa, K., and Kawano, D.A. Survey of The Blood Pressure of Women Suspected
of Itai-itai Disease. Juzen Med. Soc. J. 77, 357.363 (1969).
20. Duggan, R.E. and Cornelius sen, P.E., Dietary Intake of Pesticide Chemicals in the
United States (III), June 1968-April 1970, Pest. Mon. Journal., 5, No. 4, 331-341
(March 1972).
21. Murthy, G.K., Rhea, U. and Peeler, J.T., Levels of Antimony, Cadmium, Chronium,
Cobalt, Manganese and Zinc in Institutional Total Diets, Env. Sc. and Tech. S
(5): 436-442 (May 1971).
22. Kirkpatrick, D.C., and Coffin, D.E., The Trace Metal Content of Representative
Canadian Diets in 1970 and 1971. Can. Inst. Food Sci. Technol. J. 7 : 56 (1974).
23. Meranger, J., and Smith, D.C. The Heavy Metal Content of a Typical Canadian
Diet. Can. J. Of Pub. Health, 63: 53 (1972).
24. Schroeder, H.A., Nason, A.P., Tipton, I.H., and Balassa, J J., Essential Trace
Metal* in Man: Zinc Relation to Environmental Cadmium, J. Chron. Diseases 20:
179 (1967).
25. Tipton, LH* and Stewart, P.L., Analytical Methods for the Determination of Trace
Elements-Standard Man Series. Proc. Univ. Missouri 3rd Ann. Conf. on Trace
Substances in Environmental Health, 1969, Univ. of Missouri, Columbia, Mo.
(1970).
26. Friberg, L, Piscator, M., and Nordberg, D., Cadmium in the Environment, Chemical
Rubber Company Press, Cleveland, Ohio p. 25 (1971).
27. Lewis, G. P., Jusko, W. J., Coughlin, L.L. and Hartz, S., Cadmium Accumulation
in Man: Influence of Smoking, Occupation, Alcoholic Habit and Disease. /. Chron.
Dis.,2S, 717 (1972).
28. Hammer, D.L., Calocci, A.V., Hasselblad, V., Williams, M.E. and Pinkerton, C.
Cadmium and Lead in Autopsy Tissues, /our. Occ. Med,., IS, No. 12 (Dec. 1973).
29. Friberg, L., Piscator, M. and Nordberg, G., Cadmium in The Environment, Chem-
ical Rubber Company Press (1971), p. 85.
62
-------
APPENDIX A—DRINKING WATER REGULATIONS
CHROMIUM
Chromium, particularly in the hexavalent state, is toxic to man, produces
lung tumors when inhaled, and readily induces skin sensitizations. Chrom-
ium occurs in some foods, in air including cigarette smoke, and in some
water supplies (see Table I). It is usually in an oxidized state in chlorinated
or aerated waters, but measurements for total chromium are easily made by
atomic absorption, so the somewhat conversative total value is used for this
guideline.
Table I
U.S. urban air concentrations range, 1965 (1) 0*0.028 ug/m3
Chromium content in cigarette tobacco (2) 1.4 ug/cigarette
Chromium in foods cooked in stainless-steel ware (3) 0-0.35 mg/100 g
Chromium concentration range in water supplies 1969 (4) 0-0.08 mg/1
Comparatively little data are available on the incidence and frequency of
distribution of chromium in foods. Although most information has limited
applicability, one study (5) determined the occurrence of chromium and
other elements in institutional diets. In that investigation, the concentrations
of chromium in foods ranged from 0,175 to 0.470 mg/kg.
Chromium has not been proved to be an essential or a beneficial element
in the body. However, some studies suggest that chromium may indeed by
essential in minute quantities (5, 6, 7). At present, the levels of chromium
that can be tolerated by man for a lifetime without adverse effects on health
ate still undetermined. A family of four individuals is known to have drunk
water for periods of 3 years at a level as high as 0.45 milligrams chromium
per liter without known effects on their health, as determined by a single
medical examination (8).
A study by MacKenzie et al (8) was designed to determine the toxicity to
rats of chromate (Cr") and chromic (Cr") ion at various levels in the
drinking water. This study showed no evidence of toxic responses after one
year at levels from 0.45 to 25 mg/1 by the tests employed, viz., body weight,
food consumption, blood changes and mortality. Significant accumulation of
chromium in the tissues occurred abruptly at concentrations above 5 mg/I;
however, no study has been made of the effects of chromium on a cancer-
susceptible strain of animal. Recent studies demonstrated that 0.1 mg of
potassium dichromate per kg enhances the secretory and motor activity of
the intestines of the dog (10).
From these and other studies of toxicity (11-15), it would appear that a
concentration of 0.05 mg/1 of chromium incorporates a reasonable factor of
'afety to avoid any hazard to human health.
In addition, the possibility of dermal effects from bathing in water con-
taining 0.05 mg/1 would likewise appear remote, although chromium is
recognized as a potent sensitizer of the skin (3). Therefore, drinking water
*hall not contain more than 0.05 mg/1 of chromium.
63
-------
DRINKING WATER REGULATIONS
REFERENCES
1. U.S. Pubic Health Service, National Air Pollution Control Administration. Pre-
liminary Air Pollution Survey of Chromium and its Compounds. A Literature Re-
view. U.S. Dept. of Commerce, National Bureau of Standards, Clearinghouse of
Federal Scientific and Technical Information, Springfield, VA, 22151.
2. Cogbill, E.C., and Hobbs, M.E. Transfer of Metallic Constituents of Cigarettes to
the Main-stream Smoke. Tobacco Sci. 144: 68 (1957),
3. Denton, C.R., Keenan, E.G., and Birmingham, D.G. The Chromium Content of
Cement and Its Significance in Cement Dermatitis, J. Invest. Derm. 23: 184 (1954).
4. McCabe, LJ., Symons, J.M., Lee, R.D., and Robeck, G.G. Survey of Community
Water Supply Systems- JAWWA. 62: 670 (1970).
5. Muxthy, G.K., Rhea, U., and Peeler, J.T. Levels of Antimony, Cadmium, Chrom-
ium, Cobalt, Manganese, and Zinc in Institutional Diets. Envir, Sci. Techno!. S:
436 (1971).
6. Schroeder, H.A., Balaeea, J J,, and Tipton, I.H. Abnormal Trace Metals in Man -
Chromium. J. Chron. Disease 15: 941 (1962).
7. Hopkins, L.L. Chromium Nutrition in Man. Proceedings of Univ. of Missouri's
4th Annual Conference on Trace Substances in Environmental Health, pp. 285-289
(1970).
8. Davids, H.W., and Lieber, M. Underground Water Contamination by Chromium
Wastes. Water Sewage Works 98: 528 (1951).
9. MacKeniie, R.D., Byerrum, R.U., Decker, C.F., Hoppert, C.A., and Langham, R.F.
Chronic Toxicity Studies II Hexavalent and Trivaient Chromium Administered in
Drinking Water to Rats. A.M.A. Arch. Industr. Health 18: 232 (1958).
10. Nauotova, M,X. Effect of Potassium Bichromate on Secretory and Motor Activity
of Intestine. Gigiena Truda I Professionally e Zabolevaniya 9: 52 (1965).
IL Gross, W.G., and Heller, V.G, Chromwea in Animal Nutrition. J. Indust. Hyg.
Toxicol. 28: 52 (1946).
12. Brard, M.D. Study of Toxicology of Some Chromium Compounds. J. Pbarm. Chim.
21: 5 (1935).
13. Conn, L.W., Webster, H.L., and Johnson, A.H. Chromium Toxicology. Absorption
of Chromium by The Rat When Milk Containing Chromium Lactate was Fed. Fed.
Am. J. Hyg. 15: 760 11932).
14. Schroeder, H.A., Vinton, W.H., and Balassa, J.J, Effect of Chromium, Cadmium, and
Other Trace Metals on The Growth and Survival of Mice. J. Nutrition 80: 39
(1965).
15. Schroeder, H.A., Vinton, W.H. and Balaasa, J.J. Effects of Chromium, Cadmium,
and Lead on The Growth and Survival of Rats. J. Nutrition 80: 48 (1965).
64
-------
APPENDIX A—DRINKING WATER REGULATIONS
CYANIDE
Cyanide in reasonable doses (10 mg or less) is readily converted to thio-
cyanate in the human body and is thus much less toxic for man than fish.
Usually, lethal toxic effects occur only when the detoxifying mechanism is
overwhelmed. The oral toxicity of cyanide for man is shown in the following
table.
Oral Toxicity of Cyanide fob Man
Literature
Dosage
Response
Citations
2.9-4. 7 mgA
Noninjurious
(1)
10 mg, single dose
Noninjurious
(2)
19 mg/1 in water
Calculated from threshold
(3)
limit for air to be safe
50-60 mg, single dose
Fatal.
(4)
Proper chlorination to a free chlorine residual under neutral or alkaline
conditions will reduce cyanide to very low levels. The acute oral toxicity of
cyanogen chloride, the chlorination product of hydrogen cyanide, is approx-
imately one-twentieth that of hydrogen cyanide (5). It should be noted that
at a pH of B.5 cyanide is readily converted to cyanate which is much less
toxic,
Because of the above considerations, and because cyanide occurs, however
rarely, in drinking water primarily as the result of spills or other accidents,
there appears to be no justification for establishing a maximum contaminant
level for cyanide.
REFERENCES
1. Smith, O.M. The Detection of Poisons in Public Water Supplies. Water Works Eng.
97: 1293 (1944).
2. Bodansky, M., and Levy, M.D. I: Some Factors Influencing the Dctoxication of
Cyanides in Health and Disease. Arch. Int. Med. 3i: 373 (1923).
3. Stokinger, H.E., and Woodward, R.L. Toxicologic Methods for Establishing Drink-
ing Water Standards. JAWWA 50: 515 (1958).
4. Annon. The Merck Index. Ed. 6. Merck & Co. Inc., Railway, NJ. p. 508 (1952).
5. Spector, W.S. Handbook of Toxicology. Tech. Rpt. No. 55-16, Wright-Patterson Air
Force Base, Ohio, Wright Air. Devel. Center, Air Res, and Devel. Command, (1955),
65
-------
DRINKING WATER REGULATIONS
FLUORIDE
The Food and Nutrition Board of the National Research Council has
stated that fluoride is a normal constituent of all diets and is an essential nu-
trient (1). In addition, fluoride in drinking water will prevent dental caries.
When the concentration is optimum, no ill effects will result, and the caries
rate will be 60-65 percent below the rates in communities with little or no
fluoride (2, 3).
Excessive fluoride in drinking water supplies produces objectionable den-
tal fluorosis which increases with increasing fluoride concentration above
the recommended upper control limits. In the United States, this is the only
harmful effect observed to result from fluoride found in drinking water
(4, 5, 6, 7, 8, 9, 10, 11). Other expected effects from excessively high intake
levels are: (a) bone changes when water containing 8-20 mg fluoride per
liter (8-20 mg/1) is consumed over a long period of time (7); (b) crippling
fluorosis when 20 or more mg of fluoride from all sources is consumed per
day for 20 or more years (12); (c) death when 2,250-4,500 mg.of fluoride
(5,000-10,000 mg sodium fluoride) is consumed in a single dose (7).
The optimum fluoride level (see Table 1) for a given community depends
on climatic conditions because the amount of water (and consequently the
amount of fluoride) ingested by children is primarily influenced by air tem-
perature. This relationship was first studied and reported by Galagan and
Associates in the 1950*s (13, 14, 15, 16), but has been further investigated
and supported by Richards, et al (17) in 1967. The control limits for fluor-
ide supplementation, as shown in Table 1, are simply the optimum concen-
trations for a given temperature zone, as determined by the Public Health
Service, DHEW, from the data cited, plus or minus 0.1 mg/liter.
Many communities with water supplies containing {ess fluoride than the
concentration shown as the lower limit for the appropriate air temperature
range have provided fluoride supplementation (18, 19, 20, 21). Other com-
munities with excessively high natural fluoride levels have effectively re-
duced fluorosis by partial defluoridation and by change to a water source
with more acceptable fluoride concentration (22, 23, 24).
Richards, et al (17) reported the degree of fluorosis among children
where the community water supply fluoride content was somewhat above
the optimum value. From such evidence, it is apparent that an approval limit
(see Table 1) slightly higher than the optimum range can be tolerated with-
out any mottling of teeth, so where fluorides are native to the water supply,
this concentration is acceptable. Higher levels should be reduced by treat-
ment or blending with other sources lower in fluoride content. In such a
case, the optimum value should be sought and maintained.
66
-------
APPENDIX A—DRINKING WATER REGULATIONS
Table
1
Annual Average of
Recommended Control
Maximum Daily Ait
Limits Fluoride
Approval
Temperatures
Concentrations in mg/1
Limit
F
Lower
Optimum
Upper
mg/1
50.0 - 53.7
1.1
12,
1.3
2.4
53.8 - 58.3
1.0
1.1
\2
22
58.4-63.8
0.9
1.0
1.1
2.0
63.9 - 70.6
0.8
0.9
1.0
1.8
70.7 - 79.2
0.7
0.8
0.9
1.6
79,3 - 90.5
0.6
0.7
0.8
1.4
It should be noted that, when supplemental fluoridation is practiced, it is
particularly advantageous to maintain a fluoride concentration at or near
the optimum. The reduction in dental caries experienced at optimal fluoride
concentrations will be diminished by as much as 50% when the fluoride
concentration is 0,2 mg/I below the optimum. (25, 26).
REFERENCES
1. National Research Council, Food Nutrition Board, Recommended Daily Allowance,
Seventh Revised Edition, Publication 1964, National Academy of Sciences, Wash-
ington, D.C. p. 55 ( 1968).
2. Dean, H.T., Arnold, F.A., Jr. and Elvove, E. Domestic Water and Dental Caries.
V. Additional Studies of the Relation of Fluoride Domestic Waters to Dental
Caries Experience in 4,425 White Children, Age 12 to 14 Years, of 13 Cities in 4
States. Pub. Health Rep. 57: 1155 (1942).
3. Dean H.T., Jay, P., Arnold, F.A., Jr. and Elvove, E. Domestic Water and Dental
Caries. II. A Study of 2,832 White Children Aged 12 to 14 Years of 8 Suburban
Chicago Communities, Including Lactobacillus Acidophilus Studies of 1,761 Chil-
dren. Pub. Health Rep. 56: 761 (1941).
4. Dean, H.T, Geographic Distribution of Endemic Dental Fluorosis (Mottled
Enamel), In: Moulton, F.R. (Ed.) Flourme and Dental Health, A.A.A.S. Pub.
No. 19, Washington, D.C., pp. 6-11 (194
-------
DRINKING WATER REGULATIONS
12. Roholm, K. Fluorine Intoxication. A Clinical-Hygienic Study. H.K. Lewis & Co.,
Ltd., London (1937).
13. Galagan, DJ, and Lamson, G.G, Climate and Endemic Dental Fluorosis. Pub.
Health Rep. 68: 497 (1953).
14. Galagan, D.J. Climate and Controlled Fluoridation. J. Am. Dent. Assn. 47: 159
(1953).
15. Galagan, D.J., Vermillion, J.R. Determining Optimum Fluoride Concentrations.
Pub. Health Rep. 72: 491 (1957).
17. Richards, L.F., et al. Determining Optimum Fluoride Levels for Community Water
Supplies in Relation to Temperature. J. Am. Dental. Assn. 75: (1967).
18. Pelton, W.J., and Wisan, J.M. Dentistry in Public Health W.B. Saunders Co.,
Philadelphia pp. 136-162 (1949).
19. Arnold F.A., Jr., Dean, H.T., Jay, P., and Knutson, J.W. Fifteenth Year of The
Grand Rapids Fluoridation Study. J. Am. Dental Assn. 65: 780 (1962).
20. U.S. Public Health Service. Fluoridation Census 1969. National Institutes of Health,
Division of Dental Health. U.S. Government Printing Office, Washington, D.C.
(1970).
21. Maier, F.J. Twenty-five Years of Fluoridation. JAWWA (1970).
22. Dean, H.T. and McKay, F.S. Production of Mottled Enamel Halted by A Change in
Common Water Supply. Am. J. Pub. Health 29: 590 (1939).
23. Dean, H.T., McKay, F.S. and Elvove, E. Mottled Enamel Survey of Bauxite, Ark.
10 Years After A Change In The Common Water Supply. Pub. Health Rep. 53:
1736.
24. Maier, F.J. Partial Refluoridation of Water. Public Works (1960).
25. Chrietzberg, J.E. and Lewis, F.D., Jr. Effect of Inadequate Fluorides in Public
Water Supply on Dental Caries. Ga. Dental J. (1957).
26. Chrietzberg, J.E. and Lewis, J.F. Effect of Modifying Sub-Optimal Fluoride Con-
centration in Public Water Supply. J. Ga. Dental Assn. (1962).
68
-------
APPENDIX A—DRINKING WATER REGULATIONS
LEAD
Lead is well known for its toxicity in both acute and chronic exposures,
Kehoe (1) has pointed out that in technologically developed countries, the
widespread use of lead multiplies the risk of exposure of the population to
excessive lead levels. For this reason, the necessity of constant surveillance of
the lead exposure of the general population via food, air, and water is im-
perative.
The clinical picture of lead intoxication has been well documented (2).
Unfortunately, the general picture of the symptoms is not unique (i.e., gas-
trointestinal disturbances, loss of appetite, fatigue, anemia, motor nerve
paralysis, and encephalopathy) to lead intoxication and often this has re-
sulted in misdiagnosis (3, 4). Several laboratory tests that are sensitive to
increased lead blood levels have been developed for diagnostic purposes, but
their relationship to the effects of lead intoxication are incompletely under-
stood. The most sensitive of these is the inhibition of red cell-aminolevulinic
acid dehydrase (AT,AD) which correlates well with blood lead levels from
5-95 ug/100 g blood (5, 6). Because this is not the rate-limiting step in por-
phyrin biosynthesis, accumulation of aminolevulinic acid (ALA) does not
occur until high blood lead levels are reached. Other such tests, which cor-
relate with blood lead to a lesser degree and at higher levels, are the meas-
urment of urinary coproporphyria, the number of coarsely stippled red-
blood cells and the basophilic quotient (6). These changes, in themselves,
Have little known significance in terms of the danger to the health of the
normal individual, for although red cell life-time can be shown to decrease
(7), high lead concentrations are required for the development of the
anemia typical of lead intoxication (8). Urinary ALA, however, has been
shown to be closely related to elevated lead levels in soft tissues (9, 10) and
is considered to be indicative of a probable health risk (11).
Young children present a special case in lead intoxication, both in terms
of the tolerated intake and the severity of the symptoms (8). Lead ence-
phalopathy is most common in children up to three years of age (12). The
most prevalent source of lead in these cases of childhood poisoning has been
lead-containing paint still found in many older homes (1, 12). Prognosis
°f children with lead encephalopathy is poor, with or without treatment. Up
to 94% of the survivors have been found to have psychological abnormal-
ities (13). It is still unknown whether smaller intakes of lead without ence-
phalopathy or subclinical lead poisoning causes mental retardation or psy-
chological abnormalities. Several studies in man and animals suggest this
(14, 15, 16, 17), but a well-controlled prospective study in man has yet to
be done. AT,AD in baby rats' brains is suppressed by excess lead (18); how-
ever, the significance of this finding to humans is unknown. Some groups
of individuals who experienced lead intoxication at an early age and sur-
vived have demonstrated a high incidence of chronic nephritis in later hfe
(19). Recent work has demonstrated a high incidence of aminoaciduria and
69
-------
DRINKING WATER REGULATIONS
other biochemical changes of kidney disease in children in Boston with ex-
cessive lead exposure (17). A recent study found anemia in children with
blood levels from 37-60 ug/100 ml to be common (20). There is evidence
that lead in high doses in animals affects the immunological system (21, 22,
23, 24); this, however, has not yet been demonstrated in man.
The average daily intake of lead via the diet was 0.3 mg in 1940 (25) and
rarely exceeded 0.6 mg. Data obtained subsequent to 1940 indicate that the
intake of lead appears to have decreased slightly since that time (1, 26). In*
haled lead contributes about 40% to total body burden of lead (1, 27) in
the average population. Cigarette smoking in some studies in the past has
also been associated with slightly elevated blood lead levels (3).
Accumulation of lead with age in non-occupationally exposed individuals
has been demonstrated (26, 28, 29). The bulk of this lead distributes to
bone, while soft tissues levels vary only slightly from normal even with high
body burdens (30). Blood levels vary only slightly from normal even with
high body burdens (30). Blood levels of lead in persons without unusual
exposure to lead range up to 40 ug/100 g and average about 26 ug/100 g
(1). The U.S. Public Health Service (31) considers 40 ug/100 g lead or
over in whole blood in older children and adults on two separate occasions
as evidence suggestive of undue absorption, either past or present. Levels of
50-79 ug/100 g require immediate evaluation as a potential poisoning case.
Eighty ug/100 g or greater is considered to be unequivocal lead poisoning.
The 40 ug/100 g lead level in blood probably has a biological effect as the
National Academy of Science Lead Panel (11) concluded:
.. the exponential increase in ALA excretion associated with blood lead
content above approximately 40 ug/100 g of blood signifies inhibition of
ALAD that is significant physiologically in vivo."
In addition animal experiments show beginning renal injury at about the
same exposure level causing urinary ALA increase (32).
Blood lead is increased in urban vs. suburban (28, 33, 34), near to vs.
distant from large motorways (35, 36) and in occupational exposure to
areas of high traffic density (37, 38, 39). Lead in soil has epidemioiogically
been implicated in increased blood lead in children (40).
The World Health Organization Committee (41), assuming 10% of lead
from food and water is absorbed, established in adults a "Provisional toler-
able weekly intake" of 3 mg of lead per person (the maximum lead exposure
the average person can tolerate without increased body burden). (Kehoe
considers 600 ug per day the limit). Assuming 10% absorption from the
gastrointestinal tract, approximately 40 ug of lead per day would be ab-
sorbed, by the WHO standard. With the average diet containing 100-300 ug
lead per day, and the average urban air containing 1 to 3 ug/m3 of air, the
average urban man would absorb 16 to 48 ug of lead per day. (The con-
tribution from 1 ug/m3 lead in air at 20 m respiratory volume with 30%
absorption is 6 ug). Just from food and air alone, some urban dwellers
70
-------
APPENDIX a—DRINKING WATER REGULATIONS
would have excessive exposure by the WHO standard. Urban children are
further exposed by dust with levels of over 1000 ug/g (40, 42, 43) and be-
cause airborne lead particles vary in density inversely from the distance
from the ground (44, 45). Rural children have significantly less exposure
than do urban children to these sources. Additionally, children have in-
creased risk, because they have food and air intakes proportionally greater
than their size and they might absorb a larger percentage from their gut,
possibly 50% of ingested lead (46). Lead might also have a greater effect
on their developing neurological, hematological, and immunological sys-
tems (18, 20-24, 47, 48). Likewise, fetuses of mothers unduly exposed may
be at risk (49, 51), and Mclntire concluded that there is a definite fetal risk
maximal in the first trimester from intrauterine exposure to increased lead
in maternal blood (52).
The lead concentrations in finished water ranged from 0 to 0.64 mg/liter
in the Community Water Supply Study conducted in 1969 (53). Of the 969
water supplies surveyed, 1.4% exceeded 0.05 mg/liter of lead in drinking
water. Five of the water supplies in this sample had sufficient lead to equal
or exceed the estimated maximum safe level of lead intake (600 ug/day)
without considering the additional contribution to the total intake by other
routes of exposure. Under certain conditions, (acidic soft water, in partic-
ular), water can possess sufficient plumbosolvency to result in appreciable
concentrations of lead in water standing in lead pipes overnight (54).
As a result of the narrow range between the lead exposure of the average
American in every day life and exposure which is considered excessive (es-
pecially in children) it is imperative that lead in water be maintained within
rather strict limits. Since a survey (55) of lead in surface water of the
U.S.A. and Puerto Rico found only 3 of 726 surface waters to exceed 0.05
®g/l; the standard of 0.05 mg/1 should be obtainable. For a child one to
three years old drinking one liter of water a day (probably the most a child
would drink), the contribution would be 0.05 mg/1 x 1.0 liter equals 0.050
The diet is estimated by scaling down the average adult s diet to be 150-
200 ug (56). Assuming the fraction of lead absorbed is the same for lead
in food and water, water would contribute 25 to 33% of the lead normally
ingested. For an adult drinking 2 liters per day, the contribution would be
0.1 mg/0.3 mg, or 33% of food. At lower concentrations, for example, 0.015
mg/l, the average concentration in drinking water, the contribution of water
in an adult or child would be less than 10% of that of food.
It should be reemphasized that the major risk of lead in water is to small
children (50). The potentially signifcant sources of lead exposure to chil-
dren which have been documented include paint, dust (40, 42, 43), canned
miUc (58, 59), tooth paste (60, 61), toys, newsprint ink (62, 63), and air.
Although paint is most strongly implicated eqidemiologically, there is grow-
ing evidence that others, such as dust, are important (40). There is a serious
problem with excess lead in children; it is well documented. It can lead to
71
-------
DRINKING WATER REGULATIONS
lead poisoning. Lead poisoning does cause death and morbidity in children.
A survey of 21 screening programs (64) testing 344, 657 children between
1969 and 1971 found 26.1% or over 80,000 children with blood leads of
over 40 ug/1 (which is considered evidence of excessive exposure.) Several
recent studies suggest that the frequency of intellectual and psychological
impairment is increased among children overexposed to lead who were not
thought to have had overt clinical lead poisoning (14, 15, 16, 17). With the
widespread prevalence of undue exposure to lead in children, its serious
potential sequelae, and studies suggesting increased lead absorption in chil-
dren (chronic brain or kidney damage, as well as acute brain damage); it
would seem wise at this time to continue to limit the lead in water to as low
a level as practicable. Data from the Community Water Supply Study and
other sources indicate that a lead concentration of 0.05 mg/1 or less can be
attained in most drinking water supplies. Experience indicates that less than
four precent of the water samples analyzed exceed the 0.05 mg/1 limit and
the large majority of these are due to stability (corrosion) problems not due
to naturally occurring lead content in the raw waters.
REFERENCES
1. Kehoe, R.A., The Harben Lectures 1960. The Metabolism of Lead in Man in Health
and Disease. Lecture 1. The Normal Metabolism of Lead. Lecture 2. The Metabolism
of Lead Under Abnormal Conditions. Lecture 3: Present Hygienic Problems Relat-
ing to The Absorption of Lead. J. Roy. Inst. Pub. Health 24: 81 (1960).
2. Goodman, L.S. and Gilman, A. The Pharmacological Basis of Therapeutics. The
MacMillan Co., London and Toronto, pp. 977-982 (1970).
3. Hardy, H.L., Lead. Symposium on Environmental Lead Contamination. PHS
#1440, December 13-15, (1965).
4. Jain, S., O'Brien, B., Fotheringill, R., Morgan, H.V. and Geddes, A.M., Lead
Poisoning Presenting as Infectious Disease. The Practioner, 205: 784 (1970).
5. Hernberg, S., Nikkanen, J., Melling, G. and Lilius, H. A-aminolevulinic Acid
Dehydrase as a Measure of Lead Exposure. Arch. Environ. Health 21: 140 (1970).
6. de Bruin, A. and Hoolboom, H., Early Signs of Lead-exposure. A Comparative
Study of Laboratory Tests. Brit. J. Industr. Med. 22, 203 (1967).
7. Westerman, M.P., Pfitzer, E., Ellis, L.D., and Jensen, W.N. Concentrations of Lead
in Bone in Plumbism. New Eng. J. Med. 273: 1246 (1965).
8. Chisolm, J J., Jr. Disturbances in The Biosynthesis of Heme in Lead Intoxication.
J. Pediat. 64, 174 (1964).
9. Cramer, K. and Selander, D., Studies in Lead Poisoning, Comparison of Differ-
ent Laboratory Tests. Brit. J. Industr. Med. 22, 311 (1965).
10. Selander, S., Cramer, L. and Hallberg, L. Studies in Lead Poisoning: Oral Ther-
apy with Penicillamine. Relationship Between Lead in Blood and Other Laboratory
Tests. Brit. J. Industr. Med. 23 : 282 (1966).
11. Airborne Lead in Perspective. The Committee on Biological Effects of Atmo-
spheric Pollutants. National Research Council, National Academy of Sciences.
Washington, D.C. (1972).
12. Byers, R.K. Lead Poisoning. Review of The Literature and Report on 45 Cases.
Pediatrics 23 : 585 (1959).
13. Mellins, R.B., and Jenkins, C.C. Epidemialogical and Psychological Study of Lead
Poisoning in Children. J. Am. Med. Assn. 158: 15 (1955).
14. Moncrieff, A.A., Koumides, O.P. and Clayton, B.E. Lead Poisoning in Children.
72
-------
APPENDIX A—DRINKING WATER REGULATIONS
Arch. Dis. Child. 39: 1-13 (1964).
15. David, 0., Clark, J., and Voeller K., Lead and Hyperactivity. Lancet 2 : 900 (1972).
16. de la Burde, B., and Choate,. M.S., Jr., Does Asymptomatic Leade Exposure in
Children Have Latent Sequelae? J. Pediat., 81: 1088 (1972).
17. Pueachel, S.M., Kopito, L,, and Schwachman, H. Children with An Increased Lead
Burden: A Screening and Follow-up Study. J. Am. Med. Assn. 222: 462 (1972).
18. Millar, J.A., Battistini, V., Cumming, R.L.C., Carswell, F., and Goldberg, A. Lead
and -atninolaevulimc Acid Dehydraae Levels in Mentally Retarded Children and
in Lead-poisoned Suckling Rata. Lancet, 2: 695 (1970).
19. Henderson, D.A. Follow-up of Casea of Plumbism in Children. Aust. Ann. Med.
3, 219 (1954).
20. Betts, P.R., Astley, R., Raine, D.N. Lead Intoxication in Children in Birmingham,
British Med. J. J: 402 (1973).
21. Selye, H., Tuchwever, B., and Bertok, L. Effect of Lead Acetate on the Suscep-
tibility of Rats to Bacterial Endotoxins. J. Bacteriol. 91: 884 (1966).
22. Hemphill, F.E., Kaeberle, M.A.,. and Buck, W.B. Lead Suppression of Mouse Re-
sistance to Salmonella Typhimurium. Science 127: 1031 (1971).
23. Gainer, J. H. Effects of Metals on Viral Infections in Mice. Env. Health Persp.
: 98-999 (June 1973).
24. Holper, K., Trejo, R.A., Brettschneider, L-, DiLuzio, N.R. Enhancement of Endo-
toxin Shock in The Lead-sensitized Subhuman Primate, Surg. Gynecol., Obstr.
136: 594 (1973).
25. Kehoe, R.A., Cholak, Hubbard, D.M., Bambach, K., McNary, R.R. and Story, R.V.
Experimental Studies on the Ingestion of Lead Compounds. J. Industr. Hyg. Tox-
icol. 22: 381 (1940).
26. Schroeder, H.A. and Balassa, J J. Abnormal Trace Elements in Man: Lead. J.
Chron. Diseases 14 : 408 (1961).
27. Kehoe, R.A. Under What Circumstances i« Ingestion of Lead Dangerous. Sym-
posium on Environmental Lead Contamination. (PHS #1440), (December 13-15,
1965).
28. Hardy, H.L., Chamberlain, R.I., Maloof, C.C., Boylen, G.W., and Howell, M.C.,
Lead as An Environmental Poison, Clin. Pharmacol. 12 : 982 (1971).
29- Schroeder, M.A. and Tipton, J.M., The Human Body Burden of Lead. Arch. En-
viron. Health 17 : 965 (1958).
30* Barry, P.S.I. and Mossman, D.B: Lead Concentrations in Human Tissues. British
Indus. Med, 27 : 339 (1970).
31. Medical Aspects of Childhood Lead Poisoning. HSMHA Health Repts. 86: 140
(1971.)
32. Goyer, R.A., Moore, J.F. and Kregman, M.R. Lead Dosage and the Role of the
Intranuclear Inclusion Body. Arch. Environ. Health 20 : 705 (1970).
33. Blokker, P.C. A Literature Survey of Some Health Aspects of Lead Emissions
from Gasoline Engines. Atmospheric Environ. 6: 1 (1972).
34. Hofreuter, D.H., et al. The Public Health Significance of Atmospheric Lead. Arch.
Environ. Health 3: 82 (1961).
35. Anonymous. Lead in the Environment and Its Effect on Humans, State of Califor-
nia Public Health Department. (1967).
36. Thomas, H.V., Milmore-, B.K., Heidbreder, G.A. and Kogan, B.A. Blood Lead of
Persons Living Near Freeways Arch. Environ. Health IS: 695 (1967).
37. Hammond, P.B. Lead Poisoning: An Old Problem with a New Dimension Easaya
in Toxicol. 1: 115 (1969).
38. Anonymous, Survey of Lead in The Atmosphere of Three Urban Communities,
U.S. Public Health Service Publication 999-AP-12, (1965).
73
-------
DRINKING WATER REGULATIONS
39. Tola, S., et al. Occupational Lead Exposure in Finland. II. Service Stations and
Garages. Work Environ. Health 9: 102 (1965).
40. Fiarey, F.S. and Gray, J.W. Soil Lead and Pediatric Lead Poisoning in Charles-
ton, S.C., J. South Carolina Med. Assn. 66: 79 (1970).
41. Evaluation of Certain Food Additives and Of the Contaminants Mercury, Lead
and Cadmium. Sixteenth Report of The Joint FAO/WHO Expert Committee on
Food Additives, Geneva, April 4-12, 1972. Published by FAO and WHO, Rome
(1972).
42. Needleman, H.L., and Scanlon, J., Getting the Lead Out. New Engl. J. Med. 288:
466 (1973).
43. Hunt, W.F., Jr., Pinkerton, C., McNulty, 0., et al. A Study in Trace Element Pol-
lution of Air in Seventy-seven Midwestern Cities. Trace Substances in Environ-
mental Health IV. University of Missoure Press, D.D. Hemphill (Ed.) Columbia,
pp. 56-68 (1971).
44. Petrova, A., Dalakmanski, Y., and Bakalov, D. Study of Contamination of the
Atmosphere in Injurious Road Transport and Industrial Products. J. Hyg. Edpi-
demio. Microbiol. Immunol. (Praha) 10 : 383 (1966).
45. Bazell, RJ. Leadpoisoning: Combating the Threat From The Air. Science 174:
574 (1971).
46. Alexander, F.W., Delves, H.T., and Clayton, B.E. The Uptake and Excretion by
Children of Lead and Other Contaminants. Proceedings of the International Sym-
posium of Environmetal Health Aspects of Lead. Luxembourg Commission of the
European Communities, Amsterdam, October 2-6, 1973 pp. 319-331 (1973).
47. Lead: Airborne Lead in Perspective. National Academy of Sciences, Washington,
D.C. (1972).
48. Grollman, A., and Grollman, F.F. Pharmacology and Therapeutics. 7th Ed. Lea
and Febigerer, Philadelphia (1970).
49. Lin-Fu, J.S. Undue Absorption of Lead Among Children—A new Look at an Old
Problem. New EngL J. Med. 286 702 (1972).
50. Scanlon, J. Human Fetal Hazards from Environmental Pollution with Certain Non-
essential Trace Elements. Clin. Pediatr. 11: 135 (1972).
51. Chatterjee, P. and Gettman, J.H., Lead Poisoning: Subculture as a Facilitating
Agent? Am. J. Clin. Nutr. 25 : 324 (1972).
52. Angle, CR. and Mclatire, M.S. Lead Poisoning During Pregnancy, Am. J. Dis,
Child 103: 436 (1964).
53. McCabe, LJn Symons, J.M., Lee, R.D., and Robeck, G.G., Survey of Community
Water Supply Systems. 62: 670 (1970).
54. Crawford, M.D. and Morris, J.N. Lead in Drinking Water. Lancet : 1087 (18,
1967).
55. Hem, J.D. and Durum, W.H. Solubility and Occurrence of Lead in Surface Water,
65,562 (1973).
56. King, B.C. Maximum Daily Intake of Lead Without Excessive Body Lead Burden
in Children. Am. J. Dis. Child. 122 : 337 (1971).
57. Lin-Fu, J.S., Vulnerability of Children to Lead Exposure and Toxicity. New Eng.
J. Med. 289: 1229 (1973).
58. Barltrop, D., Sources and Significance of Environmental Lead for Children. Pro-
ceedings of the International Symposium on Environmental Health Aspects of
Lead. Amsterdam, October 2-6, 1972. Luxembourg Commission of the European
Communities, pp. 675-681 (May 1973).
59. Murthy, G.R. and Rhea, U.S. Cadmium, Copper, Iron, Lead, Manganese and Zinc
in Evaporated Milk Infant Products, and Human Milk J. of Dairy Sci. 54: 1001
(1971).
74
-------
APPENDIX A—DRINKING WATER REGULATIONS
60. Berman, E., and McKiel, K. Is that Toothpaste Safe? Arch. Environ. Health 25:
64 (1972),
61. Shapiro, I.M., Cohen, G.H., and Needleman, H.L The Presence of Lead in Tooth-
paste. J. Am. Dent. Assn. 86: 394 (1973).
62. Joselow, M.N., Lead Content of Printed Media. Am. J, Pub. Health (in press).
63. Lourie, R,S., Pica and Poisoning. Am J. Orthopsychiatry 41: 697 (1971).
64. Cilsinin, J., Estimates of the Nature and Extent of Lead Paint Poisoning in The
United States (NBS TN-746) Dept. of Commerce, National Bureau of Standards,
Washington, D.C. (1972).
75
-------
DRINKING WATER REGULATIONS
MERCURY
Environmental exposure of the population to mercury and its compounds
poses an unwarranted threat to man's health. Since conditions indicate an
increasing possibility that mercurials may be present in drinking water,
there is a need for a guideline that will protect the health of the water con*
sumer.
Mercury is distributed throughout the environment. And as a result of
industrial and agricultural applications, large increases in concentrations
above natural levels in water, soils, and air may occur in localized areas
around chlor-alkali manufacturing plants and industrial processes involving
the use of mercurial catalysts, and from the use of slimicides primarily in
the paper-pulp industry and mercurial seed treatment.
Mercury is used in the metallic form, as inorganic mercurous (mono-
valent) and mercuric (divalent) salts, and in combination with organic
molecules (viz. alkyl, alkoxyalkyl, and aryl).
The presence of mercury in fresh and sea water was demonstrated more
than 50 years ago (1-4). In early studies in Germany, Stock (5, 6) found
mercury in tap water, springs, rain water, and beer. In all water, the con-
centration of mercury was consistently less than one ug/l; however, beer
occasionally contained up to 15 ug/l. A recent survey (7) demonstrated
that most U.S. streams and rivers contain 0.1 ug of dissolved mercury or
less per liter.
Presently the concentration of mercury in air is ill-defined for lack of
analytical data. In one study (8) the concentration of mercury contained
in particulates in the atmosphere of 2 U.S. cities was measured and ranged
from 0.03 to 0.21 ug/m3. One review (9) cited values up to 41 ug/rn3 of
particulate mercury in one U.S. metropolitan area.
Outside of occupational exposure, food, particularly fish, is the greatest
contributor to body burden of mercury. In 1967 a limited study of mercury
residues in foods waB conducted, involving 6 classes of foods. The results
indicated levels of mercury in the order of 2 to 50 ug/kg. The Atomic En-
ergy Commission sampled various foods for mercury in its tri-city study
and reported levels between 10 and 70 ug of of mercury per kg of meats,
fruits, and vegetables. In 1970, it was discovered that several types of fresh
and salt water fish contained mercury (mostly in the alkyl form) in excess of
the FDA guideline of 0.5 ppm (500 ug/kg). Mercury in bottom sediments
had been converted by micro-organisms to the alkyl form, entered the food
chains, and had accumulated in the higher members of the chains. Game
birds were also discovered to have high levels of mercury in their tissues,
persumably from the ingestion of mercury-treated seeds or of smaller ani-
mals that had ingested such seeds. The Food and Drug Administration has
established a guideline of 0.5 ppm for the maximum allowable concentra*
76
-------
APPENDIX A—DRINKING WATER REGULATIONS
tion of mercury in fish for human consumption", but for all other food-
stuffs, no tolerances have been established.
Mercury poisoning may be acute or chronic. Generally mercurous salts
are less soluble than mercuris salts and are consequently less toxic acutely.
Acute intoxication is usually the result of suicidal or accidental exposure.
For man the fatal dose of mercuric salts ranges from 20 mg to 3 g. The
acute syndrome consists of an initial phase referable to local effects (viz.
pharyngitis, gastroenteritis, vomiting, and bloody diarrhea) followed later
by symptoms of systemic poisoning (viz. anuria with uremia, stomatitis,
ulcerative-hemorrhagic colitis, nephritis, hepatitis, and circulatory col-
lapse) (10).
Acute intoxication from the inhalation of mercury vapor or dusts leads
to the typical symptoms of mercury poisoning coincident with lesions of the
mucous membranes of the respiratory tract which may ultimately develop
into bronchitis and bronchopneumonia. Inhalation of mercury in concen-
trations of 1,200 to 8,500 ug/m3 results in acute intoxication (10). In severe
cases, aigns of delayed neurotoxic effects, such as muscular tremors and
psychic disturbances, are observed. The Threshold Limit Value for all forms
°f mercury except alkyl is 0,05 mg/m3 in the U.S. (11).
Chronic mercury poisoning results from exposure to small amounts of
mercury over extended periods of time. Chronic poisoning from inorganic
mercurials has been most often associated with industrial exposure, whereas
that from the organic derivatives has been the result of accidents or environ-
mental contamination.
Workers continually exposed to inorganic mercury are particularly sus-
ceptible to chronic mercurialism. Usually the absorption of a single large
dose by such individuals is sufficient to precipitate the chronic disease that
is characterized mainly by central nervous systems toxicity (10, 12, 13).
Initially, non-specific effects, such as headaches, giddiness, and reduction
in the power of perception, are observed. Fine tremors gradually develop
Primarily in the hands and are intensified when a particular movement is
hegun. In prolonged and severe intoxication, fine tremor is interspersed
with coarse, almost choreatic, movements. Excessive salivation, aften ac-
companied by a . metallic taste and stomatitis, is common. As the illness
progresses, nervous restlessness (erethismus mercurialis) appears and is
characterized by psychic and emotional distress and in some cases hysteria.
Although the kidney is less frequently affected in this type of poisoning,
chronic nephrosis is occasionally observed.
Several of the compounds used in agriculture and industry (such as al-
koxyalkyls and aryls) can be grouped, on the basis of their effects on man,
with inorganic mercury to which the former compounds are usually
metabolized.
Alkyl compounds are the derivatives of mercury most toxic to man, pro-
ducing illness from the ingestion of only a few milligrams (21, 24).
77
-------
DRINKING WATER REGULATIONS
Chronic alkyl mercury poisoning, also known as Minamata Disease, is an
insidious form of mercurialism whose onset may appear after only a Jew
weeks of exposure or may not appear until after a few years of exposure.
Poisoning by those agents is characterized mainly by major neurological
symptoms and leads to permanent damage or death. The clinical features
in children and adults include numbness and tingling of the extremities,
incoordination, loss of vision and hearing, and intellectual deterioration.
Autopsy of the clinical cases reveals severe brain damage throughout the
cortex and cerebellum. There is evidence to suggest that compensatory mech-
anisms of the nervous system can delay recognition of the disease even when
partial brain damage exists.
Several episodes of alkyl mercury poisoning have been recorded. As early
as 1865, two chemists became ill and died as a result of inhaling vapors of
ethyl mercury (14). One of the largest outbreaks occurred in a village near
Minamata Bay, Japan, from 1953 through 1960. At least 121 children and
adults were affected (of whom 46 died) by eating fish containing high con*
centrations of methyl mercury (15). Of the population affected, 23 infants
and children developed a cerebral palsy-like disease which was referred
to as Congenital (or Fetal) Minamata Disease. Similarly, in 1964 and 1965,
the disease was reported among 47 persons, 6 of whom died, in Niigata,
Japan. Hunter et al (16) reported 4 cases of industrial intoxication from
handling of these agents. In Guatemala, Iraq, Pakistan, and the United
States, the human consumption of grain treated with alkyl mercurials for
seed purposes has led to the poisoning of more than 450 persons, some of
whom died (17-20).
The congenital (fetal) disease observed in Minamata and Niigata em-
phasize the devastating and insidious nature of these agents. Of particular
significance are the facts that (1) the affected children had not eaten con*
taminated fish and shellfish, and (2) the mothers apparently were not af-
fected although they had consumed some contaminated food. Exposure of
the fetus to mercury via the placenta and/or the mother's milk is believed
to be the etiologic basis for this disease, thus indicating the greater suscep-
tibility of infants to alkyl mercury.
Absorption is a factor important in determining the toxicity of alkyl mer-
curials. Berglund and Berlin (21) estimated that methyl mercury is
absorbed at more than a 90% rate via gastro-intestinal tract as compared
with 2% mercuric ion (22). In addition, methyl mercury crosses the pla-
centa into the fetus and achieves a 30% higher concentration in fetal
erythrocytes than in maternal red blood cells (23). However, the fetal
plasma concentration of mercury is lower than that of the mother. The rate
of uptake of methal mercury into the fetal brain is as yet unknown. Alkyl
mercury can cross the blood-brain barrier more easily than other mer-
curials, so that brain levels of mercury are much higher after a dose of alkyl
mercury than after a corresponding dose of any other mercurial.
78
-------
APPENDIX A—DRINKING WATER REGULATIONS
Excretion is of equal importance in determining the health hazard. Unlike
inorganic mercury, alkyl mercury is excreted mainly in the feces. After
exposure to methyl mercury, approximately 4% of the dose is excreted
within the first few days, and about 1% per day thereafter (24). The
biological half-life of methyl mercury in man is approximately 70 days.
Safe levels of ingested mercury can be estimated from data presented in
'Methyl Mercury in Fish" (24). From epidemiological evidence, the lowest
whole-brood concentration of methyl mercury associated with toxic symp-
toms is 0.2 ug/g. This blood concentration can be compared to 60 ug Hg/g
hair. These values, in turn, correspond to prolonged, continuous exposure
at approximately 0.3 mg Hg/70 kg/day. By using a safety factor of 10, the
maximum dietary intake should be 0.03 mg Hg/person/day (30 ug/70
leg/day). Although the safety factor is computed for adults, limiting inges-
tion by children to 30 ug Hg/day is believed to afford some, albeit smaller,
degree of safety. If exposure to mercury were from fish alone, the limit
would allow for a maximum daily consumption of 60 grams (420 g/week)
of fish containing 0.5 mg Hg/kg. In a given situation, if the total daily
mtake from all sources, air, water, and food, is approaching 30 ug/per-
son/day, the concentration of mercury and/or the consumption of certain
foods will have to be reduced if a safety factor of 10 is to be maintained.
Fortunately, since only a small fraction of the mercury in drinking water
18 in the alkyl form, the risk to health from waterborne mercury is not
nearly so great as is the risk from mercury in fish. Also fortunately, mer-
CUry in drinking water seldom exceeds 0.002 mg/1. Drinking water con-
taining mercury at the approval limit of 0.002 mg/1 will contribute a total
°f 4 ug Hg to the daily intake, and will contribute less than 4 ng methyl
mercury to the total intake. (Assuming that less than 0.1% of the mercury
m water is in the methyl mercury form.) Since the Regulations approval
limit is seldom exceeded in drinking water, the margin of safety gained
from the restricted intake of mercury in drinking water can be applied to
total intake with minimal economic impact.
REFERENCES
!• Proust, J.L On the Existence of Mercury in The Waters of The Ocean. /. Phyi.
49,153, 1799.
Garrigou, F. Sur la Presence du Mercure dans du Roeher. Compt, Rend. 84, 963-
965, 1877.
3- Willnt, E. Sur la Presence du Mercure dans les Esux de Saint-Nectaire. Compt.
Rend. 88, 1032, 1879.
4. Bardet, J. Etude Spectrographicque des Eauz Minerales Francaises. Compt. Rend.
157, 224-226, 1913.
5- Stock, A. and Cucuel, F. Die Verbreitung des Queckailbers. Naturwiasemchaften
22/24, 390-393,1934.
6- Stock, A. Die Mikroanalytische Bestimmung des Queckailbers and ihre Anwen-
dung auf Hygienische and Madizinische Fragen. Svensk Kem Tidskr 50, 242-250,
1938.
U.S. Geological Survey. Water Resources Review, July 70, p. 7.
79
-------
DRINKING WATER REGULATIONS
8. Cholak, J. The Nature of Atmospheric Pollution in a Number of Industrial Com-
munities. Proc. Natl. Air Pollution Symp., 2nd, Pasadena, California 1952.
9. National Air Pollution Control Administration, D.H.E.W. Preliminary Air Pol-
lution Survey of Mercury and Its Compounds, A Literature Review. NAPCA
Publication No. APTD 69-40, Raleigh, No. Carolina, p. 40.
10. Stokinger, H.E. "Mercury, Hg " in Industrial Hygiene and Toxicology, Vol. 2,
2nd ed., FA. Patty, Ed. (New York: Interscience, p. 1090, 1963).
11. Threshold Limit Values of Airborne Contaminants for 1970, Adopted by The
American Conference of Governmental Industrial Hygjenists.
12. Bidstrup, L.P. Toxicity of Mercury and Its Compounds, (New York: American
Elsevier Publishing Co., 1964).
13. Whitehead, K.P. Chronic Mercury Poisoning—Organic Mercury Componds. Ann.
Occup. Hyg 8, 85-89, 1965.
14. Greco, A.R. Elective Effects of Some Mercurial Compounds on Nervous System
Estimation of Mercury in Blood and Spinal Fluid of Animals Treated with Diethyl
Mercury and With Common Mercurial Compounds. Riv. Neurol. 3, 515-539, 1930.
15. Study Group of Minamata Disease. Minamata Disease. Kumamoto University,
Japan, 1968.
16. Hunter, D., Bomford, R.R., and Russell, D.S., Poisoning by Methyl Mercury Com-
pounds. Quart. J. Med. 9, 193-213, 1940.
17. Ordonez, J.V., Carrillo, J.A., Miranda, C.M., et al. Estudio Epidemiologico de Una
Enfermedad Considerada Como Encefalitis en la Region de Altos de Guatemala.
Bull. Pan Amer. Sanit. Bur. 60, 510-517, 1966.
18. J&lili, M.A., and Abbasi, A.H. Poisoning by Etheyl Mercury Toluene Sulphonalide.
Brit. J. Ind. Med. 18, 303-308, 1961.
19. Haw, I.U. Agrosan Poisoning in Man. Brit. Med. J. 1579-1582, 1963.
20. Likosky, W.H., Pierce, P.E., Hinman, A.H., et al. Organic Mercury Poisoning, New
Mexico. Presented at the Meeting of the American Academy of Neurology, Bal
Harbour, Fla., April 27-30,1970.
21. Berglund, F., and Berlin, M. Risk of Methylmercury Commutation in Men and
Mammals and the Relation Between Body Burden of Methyl-mercury and Toxic
Effects. In "Chemical Fallout" (M*W. Miller and G.C. Berg, etc.) (Springfield,
111.: Thomas Publishing Co., 1969), pp. 258-273.
22. Clarkson, T.W. Epidemiological Aspects of Lead and Mercury Contamination of
Food. Canadian Food and Drug Directorate Symposium, Ottawa, June 1970 (to be
published in Food and Cosmetic Toxicology in 1971).
23. Tejning, S. The Mercury Contents of Blood Corpuscles and in Blood Plasma in
Mothers and Their New-born Children. Report 704)5-20 from Dept. Occupational
Med., Univ. Hosp., S-221 85 Lund, Sweden, 1970.
24. Methyl Mercury in Fish, A Toxicologic-Epidemiologic Evaluation of Risks. Report
from An Expert Group. Nor. Hyg. Tidskr. Suppl. 4, 1971.
80
-------
APPENDIX A—DRINKING WATER REGULATIONS
NITRATE
Serious and occasionally fatal poisonings in infants have occurred fol-
lowing ingestion of well waters shown to contain nitrate (NO3) at concen-
trations greater than 10 mg/1 nitrate nitrogen. This has occurred with suf-
ficient frequency and widespread geographic distribution to compel recog-
nition of the hazard by assigning a limit to the concentration of nitrate in
drinking water at 10 mg/1 as nitrogen. This is about 45 mg/1 of the nitrate
ion.
Nitrate in drinking water was first associated in 1945 with a temporary
blood disorder in infants called methemoglobinemia (1). Since then, ap-
proximately 2000 cases of this disease have been reported from North
America and Europe, and about 7 to 8 percent of the infants died (2, 3, 4).
Evidence in support of the limit for nitrate is given in detail by Walton (2)
in a survey of the reported cases of nitrate poisoning of infants before 1951.
The survey shows that no cases of poisoning were reported when the water
contained less than 10 mg/1 nitrate nitrogen. More recent surveys (3, 4)
involving 467 and 249 cases tend to confirm these findings. Frequently,
however, water was sampled and analyzed retrospectively and therefore the
concentration of nitrate which caused illness was not really known. Many
infants have drunk water when the nitrate nitrogen was greater than 10
mg/1 without developing the disease. Many public water supplies in the
United States have levels of nitrate that routinely exceed the standard, but
only one case associated with a public water supply has been reported (5).
A basic knowledge of the development of the disease ¦ is essential to
understanding the rationale behind protective measures. The development
of methemoglobinemia, largely confined to infants less than three months
°ld, is dependent upon the bacterial conversion of the relatively innocuous
nitrate ion to nitrite. Nitrite then converts hemoglobin, the blood pigment
that carries oxygen from the lungs to the tissues, to methemoglobin. Be-
cause the altered pigment can no longer transport oxygen, the physiologic
effect of methemoglobinemia is that of oxygen deprivation, or suffocation.
The ingestion of nitrite directly would have a more immediate and direct
effect on the infant because the bacterial conversion step in the stomach
*ould be eliminated. Fortunately, nitrite rarely occurs in water in sig-
nificant amounts, but waters with nitrite nitrogen concentrations over 1
^g/l should not be used for infant feeding. Waters with a significant nitrite
concentration would usually be heavily polluted and would be unsatisfactory
°n a bacteriological basis as well.
There are several physiological and biochemical features of early infancy
lHat explain the susceptibility of the infant less than three months of age to
disorder. First, the infant's total fluid intake per body Weight is ap-
proximately three times that of an adult (6). In addition, the infant's
incompletely developed capability to secrete gastric acid allows the gastric
81
-------
DRINKING WATER REGULATIONS
pH to become high enough (pH of 5-7) to permit nitrate-reducing bacteria
to reside high in the gastrointestinal tract. In this location, the bacteria are
able to reduce the nitrate before it is absorbed into the circulation (7). To
further predispose the infant, the predominant form of hemoglobin at birth,
hemoglobin F (fetal hemoglobin), is more susceptible to methemoglobin
formation than the adult form of hemoglobin (hemoglobin A) (8). Finally,
there is decreased aotivity in the enzyme predominantly responsible for
the normal methemoglobin reduction (NADH-dependent methemoglobin
reductase) (9).
Winton reports on a study (10) where methemoglobin levels in blood
were measured on infants to determine subclinical effects. He indicates that
at intakes over 10 mg of nitrate ion per kilogram of body weight (2.2
mg/kg measured as nitrate nitrogen) the methemoglobin concentration is
slightly elevated over normal. The methemoglobin levels returned to normal
when the babies were changed to bottled water free of nitrate nitrogen. When
a baby is fed a dehydrated formula that is made with water that the mother
boils, (increasing the concentration), the intake of 2.2 mg N03-N/kilogram
can be reached if the water contains 10 mg/1 nitrogen. To determine if a
alight elevation of an infant's methemoglobin concentration has an adverse
health effect will require a large and elaborate study.
In some circumstances, which are not understood, the standard does not
have a safety factor. Cases of illness might occur, but for the usual situation
the limit of 10 mg/1 N03*N will protect the majority of infants. Older chil-
dren and adults do not seem to be affected, but the Russian literature reports
(11) elevated methemoglobin in school children where water concentrations
of N03-N were high, 182 mg/1.
Treatment methods to reduce the nitrate content of drinking water are
being developed and should be applied when they are ready if another
source of water cannot be used. If a water supply cannot maintain the
NO3-N concentration below the limit, diligent efforts must be made to
assure that the water is not used for infant feeding. Consumption of water
with a high concentration of NO3-N for as short a period as a day may
result in the occurrence of methemoglobinemia.
REFERENCES
1. Comly, H.H., "Cyanoaia in Infanta in Well Water," J, AM Med. Asm. 129:112-116
(1945).
2. Walton, G., "Survey of Literature Relating to Infant Methemoglobinemia Due to
Nitrate Contaminated Water." Am. J. Pub. Health, 41: 986-996.
3. Sattelmacher, P.G., "Methemoglobinemia from Nitrates in Drinking Water."
Schriftenreiche des Vereins fur Wasaer Boden and Lufthygiene. No. 21, 1962.
4. Simon, C., Mazke, M., Kay, H. and Mrowitz, G., "Uber Vorkommen, Pathogenea
and Moglichkeiten zur Propnylaxe der durcb Nitrit Veruaachten Methomoglo-
binamia." A. Kinderheilk. 91 x 124 (1964).
5. Vigil, Joseph, et al. "Nitrates in Municipal Water Supply Cause Methemoglo-
binemia in Infant," Public Health Reports, 80 (12) 1119-1121 (1965).
82
-------
APPENDIX A—DRINKING WATER REGULATIONS
Hansen, H.E. and Bennett, M.J. in Textbook of Pediatrics. Nelson, W.E., W.B.
Saunder* Company, 1964, p. 109.
7. Cornblath, M. and Hartmann, A.F., "Methemoglobinemia in Young Infanta,"
/. Pediat^ 33: 421-425 (1948).
8» BetJce, K., Kleihauer, E. and Lippa, M., "Vergleichende Untersuchugen uber die
Spontanoxydation von Nabelachnur und Erwachsenenhamoglobin." Ztschr. Kinderh
77:549 (1956).
9. Roaa, J.D, and Des Forges, J.F. "Reduction ol Methemoglobin by Erythrocyte# from
Cord Blood. Further Evidence of Deficient Enzyme Activity in Newborn Period."
Pediatrics, 23:218 (1959).
10. Winton, E.F., Tardiff, R.G., and McCabe, LJ. Nitrate in Drinking Water. J. Am.
Water Works Assn. 65:95-98 (1971).
11. Diakalenko, A.P. "Methemoglobinemia of Water-Nitrate Origin in Moldavian SSR",
hygiene and Sanitation 33:32-38 (1968).
83
-------
DRINKING WATER REGULATIONS
ORGANIC CHEMICALS
The Environmental Protection Agency's problem of how to deal with the
organics in drinking water is very complex. Several facta are undisputed:
1. Organics (synthetic and natural), some of which are produced during
the disinfection of water with chlorine, are present in all drinking waters to
some extent;
2. The organic compounds in raw source waters are from municipal and
industrial point source discharges and from urban and rural non-point
sources; the major portion of organics in most waters is of natural origin;
3. Most of the specific organic compounds in drinking water have not
been identified and analysis for many of them is very difficult;
4. Most of the identified organics in drinking water have not been
bioassayed;
5. Some of the organics that have been identified in drinking water in
¦mall quantities are toxicants, carcinogens, mutagens, and teratogens as
indicated by animal bioassay tests conducted at high dosages;
6. The effect on humans of long-term ingestion of very low levels (ng/1 to
mg/1) of organic chemicals in drinking water is not known, and the portion
of human exposure from drinking water versus the total exposure from all
sources (food, air) is seldom known although the drinking water portion is
usually considered to be small.
I. Some preliminary epidemiological studies have suggested a correlation
between cancer mortality and the concentration of certain organics in drink-
ing water but the conclusions are not firm;
8. With the passage of the Safe Drinking Water Act, Pub. L. 93—523,
Congress intended that at least some organic contaminants in drinking water
would be regulated;
9. Treatment processes are available for limiting the concentrations of
most known organic contaminants of concern;
10. Treatment for the control of organic compounds, other than those
that add taste and odor, is largely not practiced by water utilities in the
United States, although some organics are undoubtedly removed by con-
ventional treatment, which is commonly practiced;
II. Treatment for the control of organics would be an added expense and
an added operational burden for the water works industry.
Given these facts, a course of action is not clear. EPA is deeeply con-
cerned about the health of consumers of drinking water, but it does not wish
to regulate frivously without more knowledge of costs and benefits.
Only within the last few years have instrumentation and techniques
sophisticated enough to measure very small quantities of contaminants been
applied to drinking water. With the aid of modern analytic techniques, such
as gas chromatography and mass spectrometry, many types of organic chem*
icals have been detected in drinking water in various locations for the first
84
-------
APPENDIX A—DRINKING WATER REGULATIONS
time. The subsequent discoveries of chemical contaminants, including known
or suspected carcinogens, which may pose a threat to human health, con-
tributed to the passage of the Safe Drinking Water Act (SDWA) in Decern-
ber 1974.
Certain industrial, agricultural, and environmental practices have allowed
potentially harmful chemicals to enter the nation's drinking water. New com-
pounds such as various pesticides and other organic chemicals have been in-
troduced into the environment before full knowledge of their ultimate health
effects were known. In order to cope with these realities yet protect human
health to the maximum extent feasible, certain provisions were added to the
Public Health Service Act by Pub. L. 93—523 to allow for greater and
more comprehensive protection of public health from drinking water
contaminants.
Under the SDWA, EPA is required to prescribe national drinking water
regulations for contaminants that may adversely affect public health. Pur-
suant to section 1412(a) (1), EPA promulgated Interim Primary Drinking
Water Regulations (40 CFR, Subpart D, FR Vol. 40, No. 248, pp. 59566 to
59587, Wednesday, December 24, 1975) which become effective in June
1977. These are based on a review and updating of the 1962 Public Health
Service Standards and include Maximum Contaminant Levels (MCL's) for
microbiological and chemical contaminants (primarily selected inorganic
ions and organic pesticides) and turbidity (cloudiness in water). In addi-
tion, monitoring frequencies and public notification requirements for viola-
tions were established. National coverage was thereby expanded to approxi-
mately 40,000 community water systems and 200,000 other public water
systems. Maximum Contaminant Levels for natural and man-made radio-
activity were proposed in August 1975, promulgated in July 1976, and will
also become effective in June 1977.
Revised Primary Drinking Water Regulations are scheduled for proposal
in March 1977, followed by promulgation 6 months later; becoming ef-
fective 18 months thereafter (March 1979). These will either specify MCL's
or require the use of specific treatment techniques, which in the Administra-
tor's j udgment will prevent known or anticipated adverse effects on health to
the extent feasible. "Feasible" is defined in the SDWA as "use of the best
technology, treatment techniques and other means which the Administrator
finds are generally available (taking costs into consideration)."
Congress anticipated that organic chemicals would be dealt with pri-
marily in the Revised Primary Drinking Water Regulations because of the
paucity of data on the health effects of various organic chemicals, uncer-
tainties over appropriate treatment techniques, and the need for additional
information on the incidence of specific organic chemical in drinking water
supplies. Because the Interim Primary Drinking Water Regulations did not
contain Maximum Contaminant Levels for organic chemicals other than cer-
tain pesticides, EPA concurrently published Special Monitoring Regulations
85
-------
DRINKING WATER REGULATIONS
(40 CFR, Subpart E, Vol, 40, No. 248, pp. 59587-59588, Wednesday, De-
cember 24, 1975) pursuant to Sections 1445 and 1450 of the Act, that pro-
vide for a national evaluation of the presence in drinking water of approxi-
mately 20 specific organic chemicals and simultaneously attempt to correlate
their presence with several general organic measurement parameters.
In accordance with these Special Monitoring Regulations, CPA is currently
conducting an extensive year long National Organics Monitoring Survey
(NOMS) of drinking water supplies in 113 cities nationwide, which will
reflect long-term and seasonal variations and represent various types of
drinking water sources and treatment processes. Laboratory analyses will
be used to evaluate the extent and nature of organic chemical contamination
of drinking water, and to evaluate the validity of the several organic para-
meters as surrogates for measurement of potentially harmful organic
chemicals.
The National Academy of Sciences is currently conducting a major study
for EPA of the health effects related to contaminant levels in drinking water
of many potential toxicants including organic chemicals, as mandated by the
Safe Drinking Water Act. In this study the NAS will collect and evaluate
currently available published and unpublished information relating to the
toxicology of those substances in animals and humans and where possible,
where they believe sufficient data exists, make recommendations of "safe"
levels for humans. Among the factors the Academy will consider in this study
are: the margin of safety required to protect particularly susceptible seg-
ments of the population; the contributions of various routes of exposure in-
cluding water, air, food, and occupations; synergism among contaminants;
and the relative risk of different levels of exposure. The Academy will also
evaluate and report those contaminants that may pose a threat to human
health, but whose current level in drinking water cannot be determined. For
those contaminants, the Academy will recommend studies and test protocols
for future research. The project, initiated in June 1975, is scheduled for
completion by December 16,1976.
Based on the NAS report, EPA will publish;
(1) Recommended maximum contaminant levels (health goals) for sub-
stances which may have adverse effects on humans. These recommended
levels will be" set so that no known or anticipated adverse effects would
occur, allowing an adequate margin of safety. A list of contaminants which
may have adverse effect on health, but which cannot be accurately measured
in water, will also be published.
(2) Revised primary National Drinking Water Regulations. These will
specify MCL's or require the use of treatment techniques. MCL's will be
as close to the recommended levels for each contaminant as is feasible. Re-
quired treatment techniques for those substances which cannot be adequately
measured will reduce their concentrations to a level as close to the recom-
mended level as is feasible.
86
-------
APPENDIX A—DRINKING WATER REGULATIONS
The Organics Problem
Thus far, more than 300 specific organic chemicals have been identified in
various drinking water supplies in the United States. These compounds
result from such sources as industrial and municipal discharges, urban and
rural runoff, natural decomposition of vegetative and animal matter, as well
as water and sewage chlorination practices. Although compositions and con-
centrations vary from locality to locality and from time to time, the occur-
rence of organic compounds in tap water is universally acknowledged. The
human health effects of exposure to these compounds via drinking water are
®8 yet unclear. However, some of them have been shown to be carcinogenic
in animal tests and a few are known to be human carcinogens.
The majority of organic chemicals identified in drinking water have not
been examined for potential health effects. Even in the case of those with
recognized effects from studies at higher doses, the actual risk posed by in-
gesting very low concentrations over an extended period of time is not cur-
rently known. Some statistical correlations between water containing certain
organics and cancer incidences have been suggested in some very preliminary
studies. However, such correlations would not establish causality even if
they were statistically valid. Health effects research and epidemiological
studies involving organic chemical contamination of drinking water are un-
derway in an attempt to assess the effect on human health of exposure to
these substances from drinking water as well as the contribution of drink-
ing water to total human exposure.
Chloroform, one of the trihalomethanes, serves as one example of the or-
ganics problem with which EPA is dealing. Advanced analytical techniques
have facilitated the detection of chloroform in small amounts of drinking
water. The National Organics Reconnaissance Survey (NORS) in 1975 con-
firmed the widespread presence of several previously determined organics in
drinking water and, further, served to attribute the presence of chloroform
and related trihalomethanes to the chlorination disinfection process itself.
These results were subsequently supported by a further survey of 83 utilities
within EPA's Region V.
The range of the levels of chloroform found in those chlorinated water
surveys was from less than 1 microgram per liter to 366 micrograms per
liter; 20 micrograms per liter median. Chloroform seldom was detected in
the raw waters of those systems. The principal source of chloroform and
other trihalomethanes in drinking water is the chemical interaction of the
chlorine added for disinfection with the commonly present natural humic
substances found in raw water. The extent of trihalomethane formation
however, will vary depending upon season, contact time, water temperature,
PH, type and chemical composition of raw water, and treatment method-
ology.
To help assess the health risk, EPA in 1975 sought the advice of its Sci-
ence Advisory Board regarding potential carcinogenic or other adverse
87
-------
DRINKING WATER REGULATIONS
health effects resulting from exposure to organic compounds in drinking
water. Principal attention was directed to chloroform, carbon tetrachloride,
chloroethers, and benzene.
The Board prefaced its Report with the caveat that the chemicals thus far
identified in drinking water account for only a small fraction of the total
organic content. Thus, the possibility exists that additional substances of
equal or greater toxicological significance may be present. The Board also
expressed concern that future studies should take into account possible
synergistic effects of common combinations of contaminants. It recom-
mended that a complete analysis of the problem consider data from all
routes of exposure, such as dietary and occupation exposure, to these sub-
stances in addition to drinking water. Some of these additional sources of
exposure may pose a much greater potential intake and risk than the con-
sumption of drinking water.
The Report indicated that, in general, for all the compounds reviewed,
the carcinogenicity data and experimental designs were either inappropriate
or below the standard of current toxicological practice and protocols for
carcinogenicity testing. Additional well-designed experimental studies to de-
termine the carcinogenicity of lifetime exposures by ingestion were sorely
needed.
According to the Report, carbon tetrachloride, a demonstrated carcinogen
in laboratory studies, occurs in drinking water generally at much lower
levels and is much less widespread than chloroform and related trihalo-
methanes. Benzene has not been clearly established to be carcinogenic in
experimental animals, although epidemiological and clinical studies, largely
of occupational exposures, suggest that possibility. Certain haloetbers,
chloro-olefins, and polynuclear aromatic hydrocarbons have been demon-
strated to be carcinogenic in laboratory animals and have been identified in
some drinking waters.
The Report concluded that some human health risk probably does exist
from exposure through drinking water, although this risk is currently un-
quantifiable. The Report recommended that EPA seek ways to reduce ex-
posure to these compounds without increasing the risk of infectious disease
transmission.
In an early attempt to explore whether or not there is a relationship be-
tween water consumption and cancer, data obtained from the National Or-
ganics Reconnaissance Survey were compared with cancer mortality occur-
ring in populations served by these water utilities. One preliminary study
utilising data from 50 of the 80 water utilities samples indicated a statisti-
cally significant correlation between the cancer mortality for all anatomical
sites and both sexes combined in the years 1969-71, with the chloroform
concentration in the sample collected in spring 1975. Such a correlation
was not noted with total mortality or with the sum of the concentrations of
the four trihalomethanes in the drinking waters.
88
-------
APPENDIX A—DRINKING WATER REGULATIONS
In contrast to the above result, a similar epidemiological analysis of 43
cities from the Region V survey of 33 cities did not show any statistically
significant correlation between chloroform or trihalomethanes and cancer
mortality. Neither of these analyses attempted to correct for other variables
that are known to be related to cancer mortality, and which might have had
a fortuitous correlation with chloroform concentrations in water. Thus they
must be considered preliminary. These preliminary results do, however,
underline the need for more definitive analyses, which are now being
attempted.
The recently released National Cancer Institute (NCI) Report on the
bioassay of chloroform in rats and mice showed that chloroform caused can-
cers under the laboratory test conditions. EPA is very concerned with these
findings and has asked NAS to study the NCI findings and other data on the
carcinogenicity of chloroform as a part of its report to EPA under the Safe
Drinking Water Act.
Taking note of the NCI Report, the Food and Drug Administration has
banned the use of chloroform in human drugs, cosmetics and food packag-
ing. On April 6, Dr. Alexander M. Schmidt, Commissioner, stated:
The experiments on animal* by no means prove that chloroform causes cancer in
humans. The amount fed to the test animals exceeds, by far, the amount to which any
person could be exposed with present products, but the benefits of cholorform are
minimal and do not warrant any risk, however small.
Based on the information available at the time the Administrator of EPA
stated that the prudent course of action was to take steps to reduce exposure
to chloroform from drinking water wherever feasible by means that would
not increase the risk of microbiological contamination. On March 29, 1976,
EPA announced the institution of an experimental pilot cooperative chloro-
form reduction effort in which EPA would work through the states with a
number of water utilities experiencing high chloroform levels. The program
consists of carefully controlled modifications of existing water treatment
processes in 10 to 20 water utilities. To provide the supporting information,
a document titled "Interim Treatment Guide for the Control of Chloroform
and Other Trihalomethanes" has been prepared by and is available from
EPA's Water Supply Research Division in Cincinnati, Ohio. If successful
lhe effort could be expanded to include many more systems. This technical
assistance program will reduce human exposure to chloroform and other
chlorination by-products in the short run, while providing information to
support possible national regulations for organics is being developed.
Ongoing Research
In addition to the major NAS and NOMS studies in progress, research
efforts designed to identify sources, distributions, treatment techniques and
health effects of a variety of organic chemicals are being undertaken to find
answers to the following questions:
89
-------
DRINKING WATER REGULATIONS
1. What are the effects of commonly occurring organic compounds on
human health?
2. What analytical procedures should be used to monitor finished drink-
ing water to assure that any Primary Drinking Water Regulations dealing
with organics are met?
3. Because some of these organic compounds are formed during water
treatment, what changes in treatment practices are required to minimize the
formation of these compounds in treated water?
4. What treatment technology must be applied to reduce contaminant
levels to concentrations that may be specified in the Primary Drinking
Water Regulations? What is the cost of this technology?
This research will involve health effects and epidemiological studies, in*
vestigations of analytical methodology, as well as pilot plant and field
studies of organic removal unit processes.
The NAS and NOMS studies of drinking water contaminants with other
additional research efforts will provide an overview of the drinking water
problem essential in determining future national strategies. The results of all
these efforts in conjunction with public comment and advice should con*
tribute to the determination of whether an adequate basis exists, and if so,
provide that basis for establishing maximum contaminant levels for specific
organic contaminants that are found to be widespread, and/or for a general
organics parameter(s), and/or treatment requirements that may be incor-
porated into the Primary Drinking Water Regulations. This information will
enable the Administrator to determine appropriate health goals for these
contaminants and then after considering technological and economic feasi-
bility, to establish levels for National Primary Drinking Water Regulations.
However, although treatment technology development is processing rapidly,
significant new health effects information will probably not be available be-
fore regulatory decisions must be made because of the time required for
completion of animal feeding studies (usually 3 years).
Future Action
Although health effects research is underway, definitive relationships be-
tween human health effects of low level exposure to specific chemicals from
drinking water will be very difficult to establish, and such research requires
considerable time lags between its inception and conclusion. EPA feels that
the prudent action at this time is to consider the practically and feasibility
of the available control technologies which may be applied to reduce expo-
sure to many chemicals of unknown hazard and thus reduce the risks, what-
ever they may be, because of the following factors:
1. A large number of different chemicals have been found in drinking
water albeit in low levels; several are considered carcinogens, others may
have chronic toxic effects and more are likely to be found;
2. The large exposed population and the variable physiological suscepti-
bilities of the individuals;
90
-------
APPENDIX A—DRINKING WATER REGULATIONS
3. Statistical estimates of possible health effects, which although not defin-
itive, suggest that some level of risk may exist;
4. The complexities of possible health effects from interactions of the
many substances to which humans are exposed from a multitude of sources,
including drinking water.
According to the SDWA, Primary Drinking Water Regulations shall pro-
tect health to the extent feasible, using technology treatment techniques and
other means which EPA determines are generally available (taking cost in-
to consideration). In light of those considerations and the difficulty in ob-
taining the essential health data and quantifying risks in limited time, tech-
nologically and economically feasible solutions must be considered which
will reduce risks of exposure where necessary.
Possible Regulatory Options
Generally, organic chemicals in drinking water could be divided by
sources and type under the following headings:
1. Chemicals derived from natural sources (e.g. humus);
2. Contaminants introduced as a result of treatment technology {e.g. tri*
halomethanes);
3. Synthetic chemicals from point sources (e.g. industrial chemicals);
4. Chemicals from non-point sources (e.g. pesticides or aromatics).
Several categories of contaminants must be considered and therefore sev-
eral regulatory strategies may be necessary to address the problems fully. Ad-
ditonal complications are raised by local factors including raw water qual-
ity, size of the water system, financial and personnel limitations, as well as
the cost and availability of substances essential for treatment operation in-
cluding granular absorbents, such as activiated carbon, reactivation facili-
ties or disinfection chemicals.
The impact of any regulations for organics will be especially great on the
small public water systems; those serving between 25 and 1000 or 10,000
persons. The installation, operation, and maintenance of some fairly sophis-
ticated control processes and the monotoring requirements may result-in very
substantial per capita costs for small systems. EPA pointed out this problem
in the Interim Primary Drinking Water Regulations and is seeking means of
alleviating it. Fortunately many of those small systems utilize ground water
sources and some others may be able to switch to purer underground sources
which would not require extensive treatment. Since many ground waters are
already low in organics, they also would produce very little chloroform
(trihalomethanes) and minimal, if any, treatment for organics control would
be necessary in many cases. Thus, the following regulatory options would
likely impact primarily surface water supplies and shallow ground water
sources.
There are two basic regulatory philosophies possible within the SDWA;
(1) Set Maximum Contaminant Levels for chemicals, or (2) Establish
treatment technique requirements for substances which cannot be monitored
91
-------
DRINKING WATER REGULATIONS
feasibly. Within each category several actions are possible. These include,
for the MCL approach: (1) Establishing MCL's for each hazardous com-
pound, or (2) developing general indicators of organic contamination and
setting MCL's for these or (3) a combination of (1) and (2). Options with-
in the designated treatment technology category include: (1) Modification
of treatment and disinfection processes to eliminate specific contaminants
such as chloroform (this could include substitution of other disinfection
techniques for chlorine) or (2) requiring the use of a treatment technique
such as granular activated carbon (GAC) to remove almost all organic com-
pounds. These MCL and treatment options are not mutually exclusive, thus
more than one might be appropriate. Some of them relate specifically to
chloroform control and others are inclusive.
Establishment of MCL's for specific compounds or for a general organics
contamination indicator would designate the maximum amount of the sub-
stance which is permitted to be in drinking water. The standards would be
applicable in every public water system and periodic monitoring would be
required to assure compliance. If an MCL were exceeded, the water utility
would be required to notify the State and the water consumers and take
corrective action.
The MCL approach would result in consistent health protection of drink-
ing water throughout the nation. It offers flexibility by allowing each water
system to use any acceptable means to achieve the standard. These could in-
clude: use of alternative water sources, blending, or treatment methods
which could be optimized to be most cost effective in each specific case. In
general, monitoring costs would be dependent upon the number and types
of analyses required; and the problem is that many different substances
might have to be regulated. Monitoring costs tend to be sensitive economic
issues, particularly for small systems, where per capita expenditures may be
substantial.
Owing to the number of MCL's which might be necessary to regulate
organics in drinking water, and the feasibility of monitoring for such con-
taminants, it could also be appropriate to establish a treatment technique re-
quirement for organics in drinking water. Under this approach (which
could be phased-in according to system size), all public water systems would
essentially be required to apply the best treatment available for total or-
ganics or, for example, chloroform removal.
A system may obtain a variance (deferral) of an MCL if the system can-
not comply with an MCL, despite the application of the best technology
available, because of the poor quality of the raw water which is reasonably
available to the system. As with an exemption, the system must demonstrate
that the variance will not result in an unreasonable risk to public health. The
system must also comply with the MCL as expeditiously as practicable, in
accordance with a compliance schedule to be established after a public hear-
ing. It should be noted, however, that a variance from an MCL should not
92
-------
APPENDIX A—DRINKING WATER REGULATIONS
be granted unless the system has already installed the best technology avail-
able so as to at least minimize the contamination in the drinking water.
A system would not be required to comply with a prescribed treatment
technique if it can obtain a variance from the requirement under section
1415(a) (1) (B). A variance may be obtained if the system can demon-
strate to the satisfaction of a State (or EPA if a State does not have primary
enforcement responsibility) that such treatment technique is not necessary
to protect the health of persons because of the nature of the raw water source
of such system. In other words, if EPA prescribed GAC as a treatment tech-
nique for total organic contaminants, a system would not have to install
GAC if it could demonstrate to the satisfaction of the State that its finished
and/or raw water supply did not contain "harmful" quantities of total or*
ganics. This determination would presumably be based on federal and State
guidelines taking into account local raw water conditions.
Section 1416 of the Act provides for temporary exemptions from MCL's
or treatment techniques. Exemptions enable a public water system to remain
out of compliance with an MCL or treatment technique for a limited period
(up until 1981 under the interim regulations for most systems), subject to a
compliance schedule. In order to obtain an exemption, a public water system
must demonstrate to a State with primary enforcement responsibility (or
otherwise to EPA) that (1) it was in operation in June of 1977; (2) there
compelling reasons (e.g. economic or technical) for such an exemption;
and (3) the grant of such an exemption will not result in an unreasonable
risk to public health. Within one year of the grant of an exemption, a State
(or EPA) must hold a public hearing and establish a compliance schedule
to enable the system to meet the applicable requirements.
In short, a public water system may defer the impact of an MCL or treat-
ment technique upon a showing that such a deferral is necessary. However,
the duration of such a deferral is limited by Statute and the compliance
schedule established by the State (or EPA).
WPCA
In addition to the SDWA, various elements of the Federal Water Pollu-
tion Control Act (Pub. L. 92—500) impact on the quality of drinking water
sources; including control of effluents from point sources under sections
304, 307, 311 and others, non-point source controls, areawide waste treat-
ment management under section 208 and possible reporting requirements
under section 308. Use of Pub. L. 92—500 would prevent contamination of
certain water sources by some organic chemicals, and any reduction in or-
ganic load in raw water would help a water utility maintain good finished
^ater quality. Regulation sunder sections 304, 307, and 311 control the
quality of receiving waters or limit effluent discharges. Under section 308,
monitoring and reporting by dischargers can be required so that sources of
pollutants can be identified. Non-point sources of contamination are even
93
-------
DRINKING WATER REGULATIONS
more difficult to control and in situations where several sources of the same
contaminant existed, enforcement becomes more difficult.
Non-regulatory Options
A. Other short run actions, though not necessarily general regulatory op-
tions, need to be considered. Under the Safe Drinking Water Act, EPA has
the authority to take action to deal with an imminent and substantial en-
dangerment to human health involving a public water system. Unlike maxi-
mum contaminant levels or required treatment techniques, which cannot
take effect until June 24, 1977, the imminent hazard authority can be used
immediately. However, as a practical matter this authority could only be
used in a limited number of cases and does not appear to be appropriate for
dealing on a national basis with widespread problems.
B. An interim alternative specifically for chloroform reduction would
recommend the measurement of chloroform in finished water and offer
technical assistance to interested states and water utilities wanting to alter
their treatment procedure in order to lower chloroform concentrations, and
thus the risk from chloroform exposure. Some initial monitoring would be
necesary to determine which water utilities may need to alter their treatment
procedures. Since this would not be regulatory or mandatory, not all water
utilities which might need to take action will do so. Therefore, the health
risk reduction to the population would not be as great as it would be by
regulation. This was the interim approach outlined in the Administrator's
statement of March 29, described earlier in this notice.
C. Another interim alternative, short of establishment of MCL's or treat-
ment requirements, would be the issuance of regulations requiring monitor-
ing for many organic chemicals. This would produce a large data base from
which to develop future regulations*- increase awareness of the presence of
these contaminants, and point out the existence of potentially hazardous
substances where they were previously not suspected. This could result in
voluntary corrective actions, including the identification of sources so that
some of these would be controlled.
D. Lastly, a choice may be to recommend that no change in current water
treatment practices be made for the time being. Taking no regulatory action
at this time would avoid impacting water utilities with treatment require-
ments that may be changed in the near future. Also, not taking regulatory
action until additional data becomes available may be reasonable. However,
the negative aspect of this action would be that no change in water works
practice means no change in the current organic levels in finished water,
and no reduction of potential health risks. However, it should be noted that
the Agency has been challenged in the U.S. Court of Appeals for the District
of Columbia, in part because more extensive organic standards were not con-
tained in the Interim Primary Regulations of December 197S.
94
-------
APPENDIX A—DRINKING WATER REGULATIONS
Advantages and Disadvantages of Various Options
I. Maximum Contaminant Level Options. A. Establish MCVs for specific
organic chemicals. Based on nationwide distribution and health effects data,
MCL's might be established for many specific organic substances. Because
°f limited health data now available, a major factor in many cases would be
feasibility, based on economics and practicality of attaining lower risk expo-
sure levels. Thus far, MCL's for 6 pesticides have been established in the
Interim Primary Regulations. Others are currently being considered for
regulation and more information is being gathered in the current NOMS
Program. The acceptable MCL's would undoubtedly be quite low (mg/1 or
ug/1 level and below) thus both sophisticated monitoring (gas chromato-
graphy/mass spectrometry, (GC/MS)), and treatment methods would be
necessary. Since a large number of chemicals would be candidates for
MCL's, monitoring would probably be frequent and costly. A large number
of systems would probably require some kind of treatment; variances and
exemptions are possible under the Act but would only temporarily delay ac-
tion, The most likely means of achieving the standards would involve use of
less polluted source water or adsorbants.
Example, MCL for Chloroform (Trihalomethanes). Setting an MCL pro-
vides a legal requirement for a standard to be met on a nationwide basis by
public water systems and would require periodic monitoring along with
public notification if an MCL is being exceeded. The means of achieving
'he MCL would be the prerogative of the individual water system. These
c°uld include: treatment process modifications; switching to a raw water
8°urce which contains less of the precursor compounds (e.g. groundwater);
U8*ng a disinfectant other than chlorine (e.g. ozone, chlorine dioxide, chlora-
roines); use of adsorbants to remove either precursor substances (more
likely) or to remove chloroform (less likely). In some cases the action might
b© needed only intermittently (e.g. seasonally). Local conditions, including
^Qnomics and available personnel, could determine which approach is the
m°st practical.
If it were determined that the MCL approach is appropriate, a set of three
Possibilities related to chloroform are presented below as examples. A sim-
ilar approach could be applied to some other compounds. The several control
'©vela could also be applied consecutively in a phased approach starting
the less stringent levels and reducing them over some period of time
aa widespread compliance became more possible.
a- Establish Interim Levels to Cover Worst Case Situations: e.g. Chloro-
form, 100 ug/1; A small percentage of water utilities, mostly on surface
s°urces, would be affected by a regulation at these levels. Such an MCL
c°uld be imposed under Amended Interim Primary Regulations, then re-
duced to the maximum extent feasible under the Revised Primary Drinking
^ater Regulations. This might be cost-effective since in many cases, only
Modest or seasonal modifications would be required to meet the standard*.
95
-------
DRINKING WATER REGULATIONS
This could include adjustment of chlorination practices, use of alternative
disinfectants, or blending. Use of adsorbants for treatment would increase,
but longer than optimal periods between reactivation (regeneration)
could be employed because chloroform levels below 100 ug/1 might be main-
tainable for several months in many localities. Considerable reduction of
exposure would result at least seasonally for a fairly large population group.
b. Establish Levels at the Wintertime Median Found in the EPA National
Survey: e.g. Chloroform, 20 ug: A very large number of water systems
could be affected and considerable treatment would be necessary in many
cases, at least seasonally. Granular activated carbon (GAC) or other ad-
sorbants or alternative disinfectants would be necessary in many cases. A
much broader population segment would be consuming water of consider-
ably improved quality. Some phasing would undoubtedly be necessary, re-
sulting in issuance of many variances or exemptions until widespread com-
pliance could be achieved. Considerable increases in demand for ad-
sorbants, ozone, chlorine dioxide, ammonia feeders, reactivation facilities
and engineering service would result.
c. Establish Very Low Limit Levels: e.g. Chloroform, 5 ug/1 or less: Vir-
tually every surface water and many ground water systems would be af-
fected and adsorbants or alternate disinfectants would be needed for treat-
ment. Extensive phasing would be necessary, therefore, variances and ex-
emptions would be extensively used. Demand for new equipment and chem-
icals and engineering services would be intense for several years. If GAC
were used, reactivation would probably be required in many systems on
monthly or shorter schedule and consumers would be receiving water of
very high quality with respect to many chemicals as well as chloroform.
B. Establish MCUs for general organic contaminant indicators. Because
of the probably multitude of organic contaminants in drinking water, the
difficulties in toxicologically distinguishing between many of them at the
low levels generally found in drinking water, and the impracticability and
costs of monitoring and enforcing standards for tens or hundreds of in-
dividual contaminants, MCL's for groups of compounds or general organic
indicators should be considered. This is analogous to the use of coliform
bacteria as the indicator of microbiological contamination in water. These
general parameters might consist of standards for groups such as polynu-
clear aromatics, or nitrosamines, or element analyses such as Non-Purgeable
Total Organic Carbon (NPTOC), Total Organic Carbon (TOC), Total Or-
ganic Halogen (TOH), or Total Organic Nitrogen (TON).
Since a general indicator cannot distinguish individual compounds, some
relationship should exist between the indicator's value and the levels of toxic
compounds in the water, although the general organics indicator might not
be as sensitive as the most sophisticated single compound analyses. The in-
dicator could also be used as a trigger to indicate the need for more detailed
analyses.
96
-------
APPENDIX A—DRINKING WATER REGULATIONS
NPTOC is probably not sensitive to low level pollution from synthetic
chemicals (pesticides and other non-humus type compounds). Total Organic
Halogen (TOH) is used somewhat in Europe and it may be an acceptable
indicator of the many halogenated industrial and pesticide compounds and
halogenated trihalomethanes precursors. Total Organic Nitrogen (TON)
may be an acceptable indicator of nitrogenous compounds, some of which
may be precursors to nitrosamine formation. Ultraviolet absorption and
fluorescence have also been suggested as possibilities. The Organics-Carbon
Adsorbable (O-CA) test was suggested in the proposed Interim Primary
Drinking Water Regulations but was rejected in the promulgated regula-
tions. A common problem with these general indicators is the cost and avail-
ability of apparatus which is sensitive in low analytic ranges (sub mg/1).
NPTOC analysis is the most highly developed and TOC and TAH develop-
ment work is in progress. The NOMS is expected to provide data on several
of these general indicators. Neither NPTOC, O-CA, UV or fluorescence cor-
related well with chloroform concentrations in the NORS.
Monitoring could probably be less frequent than for individual com-
pounds and cost per analysis would be relatively small (less than $10 if per-
formed externally), however instrument costs could be substantial ($6,000
to $10,000 each). Since the resulting numerical value is non-specific, addi-
tional analytical data might be necessary if the indicator value is exceeded.
Because of the insensitivity and non-specific nature of general organic
indicators, selection of MCL's based on a direct health relationship is diffi-
cult, except by utilizing the principle (similar to the coliform indicator for
microbiological contamination) that the lower the level of total prganics,
the smaller the possibility of adverse effects.
By analogy to the chloroform MCL discussion, MCL's for a general or-
ganic indicator, for example NPTOC, could be selected from several possible
levels (e.g. 5 mg/1 or 1 mg/l or 0.1 mg/l), and a phased reduction could
be applied.
That approach has at least two problems: (1) NPTOC does not measure
v°latile compounds such as chloroform and (2) most utilities could be af-
fected ultimately, and many variance and exemption requests would have
to be processed. The second problem would be considerably alleviated if a
reasonable phase-in schedule were employed. The ultimate benefit would
k® that drinking water of high quality, considering both health risk and
e'thetics (taste and appearance), would result.
C. Combination of MCL's for specific compounds and general organic
indicators. Many water supplies that are known to be free of industrial or
human waste discharge contamination contain a high concentration of a
general organics indicator (e.g. NPTOC) because of the presence of large
founts of natural substances such at humus. Conversely, some waters con-
taminated with potentially hazardous chemicals at the microgram per liter
kvel might have a low NPTOC at the milligram per liter level.
97
-------
DRINKING WATER REGULATIONS
At least an initial possible solution might be to categorize water systems
based on contamination type (e.g. natural or synthetic) and to establish
MCL's combining both general indicators and appropriate specific chem-
icals. In this way, those water supplies contaminated with substances of
greatest concern for which MCL's would be written would be required to
take action first. Some element of phasing would be introduced, such that
the high NPTOC (or other general indicator) and low synthetic organic
contamined water sources would be affected later if adjustments in the
standards were deemed appropriate. However, this approach assumes that
the naturally occurring substances in water are normally less hazardous
than the others, which may be true in general, but much more analytical
and health effects work must be performed to determine if that indeed is
the case. In addition, the definition and application of the distinction be-
tween natural and synthetic chemicals would be difficult in practice.
II. Designated Treatment Technology to Control Either Specific Contamin-
ants (e.g. Trihalomethanes) or Total Organics.
Monitoring for a number of organic MCL's might be infeasible, and
moreover, the MCL approach might not encompass all possible components
of the problems. The SDWA allows EPA to establish treatment techniques
requirements if it is not feasible to monitor for a given contaminant. Thus,
a treatment technique requirement would prescribe one or more available
technologies that public water systems must apply instead of meeting par-
ticular MCL's,
For example, methods are available to analyze for trihalomethanes in
water, however, other products of chlorination may be much more difficult
to quantify. The formation of chloroform can be avoided or reduced in
water by certain chlorination process modifications, use of absorbants such
as GAC prior to chlorination or by using an alternate disinfectant such as
ozone or chlorine dioxide instead of chlorine. Unless an absorbent was be-
ing used, the concentrations of other organic contaminants would not be
materially affected, except for the by-product of reaction with the
disinfectants.
A treatment regulation for control of total organics would probably re-
quire the use of an adsorbant. The operation of the process would probably
be monitored by the breakthrough of some general organics parameter
(e.g. NPTOC), or of some indicator chemical (e.g. chloroform). Such a
technology requirement could be applied to public water systems in a phased
manner based on treatment plant size, A schedule could be selected such
that utilities of greater than 100 MGD could be affected initially in amended
Interim Primary Regulations and smaller systems could be included later
on a prescribed schedule (e.g. 100 MGD by June 1977, 50 MGD by June
1978,10 MGD by June 1979 etc.).
Treatment would not necessarily have to be in place on the effective date
of the regulation. States with primary enforcement responsibility or EPA
98
-------
APPENDIX A—DRINKING WATER REGULATIONS
could grant the variances if the water system could demonstrate that its wa-
ter supply did not contain harmful quantities of organics. Conditions for
granting variances probably would involve a survey of discharges into the
source water and detailed organic analyses, and so enable a State to make an
essentially "case by case" determination. Thus, in an area frequently sprayed
by a particular pesticide or subjected to particular discharges these in-
dividual problems could be considered.
Exemptions for a limited period could also be granted from treatment
techniques upon a showing of necessity (e.g. to install equipment or to raise
the necessary capital), and that the exemption would not result in an undue
risk to health.
Processing variance or exemption requests is an administrative burden
under either an MCL or treatment technique approach. Somewhat fewer
than 700 water utilities have an average flow of 10 MGD or greater. Of
these, nearly 300 use ground water as the source. By phasing in a treatment
technique requirement for plants over 10 MGD before 1980, the States or
EPA would be able to carefully process applications for variances and ex-
emptions and benefit the largest population segments initially, and it would
become more feasible for public water systems to construct or develop the
necessary technologies. Subsequently, smaller plants could also be required
to adopt a treatment technique for organics.
During the phase-in period, the smaller systems which were not yet af-
fected by the treatment requirement could be required to meet one or more
MCL limits (eg. chloroform). Thus some level of protection would be avail-
able immediately in all cases.
Several treatment technique possibilities involving both specific contami-
nants and total organics are described below. •
A. Modify the chlorination process. Chlorination is currently the principal
method of disinfection of water supplies and it is the major line of
defense against waterborne disease caused by bacterial and viral
contamination. EPA has been actively examining alterations in the
chlorination process to find ways of reducing the amount of trihalomethanes
that are produced. Although of questionable biocidal value, addition of am-
monia following chlorination is also a way that eliminates the chlorine that
would be available for further reaction with organic compounds.
It appears that changes in the point of application of chlorine can sig-
nificantly reduce the quantity of chlorine applied and the amounts of tri-
halomethanes and other chlorinated organics generated in some systems us-
ing filtration of source waters which contain the natural organics precursors.
For example, the common practice of prechlorination of raw surface water
to insure adequate disinfection is likely to produce greater quantities of tri-
halomethanes, compared to chlorination after the water has been coagulated
and settled, resulting in some chloroform precursor removal. For this
reason, EPA has been critically reviewing chlorination practices to see if
99
-------
DRINKING WATER REGULATIONS
simple modifications (such, as ceasing raw water chlorination in favor of
chlorination just prior to filtration) can be made that would minimize the
formation of chloroform yet still provide microbiologically safe drinking
water. Initial results have been promising.
The discontinuation of raw water chlorination would be easily accom-
plished by most water utilities, and it could be done at low cost since addi-
tional equipment would usually not be needed and chlorine use would de-
cline. Modifying the chlorination process could avoid causing a sudden de-
mand for water treatment chemicals and not further tax limited energy
resources. Discontinuation of raw water chlorination would not completely
eliminate chloroform from the finished water, so some continued risk from
chloroform exposure would exist.
Ceasing the disinfection of the raw water could possibly result in poorer
microbiological quality in the finished water, so increased microbiological
monitoring might be necessary. EPA is currently examining the practicality
of this approach in a limited number of water systems (vide supra Non
Regulatory Option B). Results will be released as they become available.
B. Use of alternate disinfectants. The principal source of chloroform in
drinking water is the chemical interaction of chlorine added for disinfection
with common humic substances formed from the nautral decomposition of
vegetation. One possible way of avoiding the formation of trihalomethanes
would be to substitute other disinfectants such as chlorine dioxide or ozone
for chlorine, where possible. Any action to control chloroform and other or-
ganics in drinking water must not increase risk of waterborne disease by re-
ducing the level of protection or by introducing other unknown risks (eg.
from chemical by-products of other disinfection processes).
a. Chlorine dioxide. Because of its oxidizing properties, chlorine dioxide
has been used to some extent for taste and odor control, but because of its
cost it is not widely used in water treatment practice for disinfection. Some
studies have shown that disinfection by this method is satisfactory and that
a residual can be maintained to insure against bacteriological contamina-
tion in the distribution system. The problem with using chlorine dioxide
now, is that our present knowledge is lacking regarding the products of its
interaction with organic chemicals in water and on the possible toxicity oi
the inorganic ions that it generates.
Installation of a unit for chlorine dioxide generation would not be partic-
ularly costly or complicated relative to chlorination. Disinfection cost should
average between 1 and 2 cents per 1000 gallons in larger systems, and about
3 cents per 100 gallons for a I MGD plant, when chlorine dioxide is gen-
erated from sodium chlorite.
b. Ozonation. Ozone i» a strong disinfectant but has the disadvantage of
not producing a disinfectant residual to carry throughout the water system.
Thus chlorine, chlorine dioxide or chloramines might have to be used fol-
lowing ozone. Ozonation of drinking water ia practiced in several hundred
100
-------
APPENDIX A—DRINKING WATER REGULATIONS
systems throughout the world. However, little is known about the by-pro-
ducts of ozonation of chemicals in water.
In general, both substitute treatment methods have the advantage of re-
ducing the health risk of exposure of chlorinated compounds. Most of the
disadvantages associated wtih implementing this option are due to a lack
of complete information as to the doses required for disinfection, the reac-
tions involved, possible toxic organic by-products of these reactions and the
operational experience needed. In addition there is the question of avail-
ability of equipment, chemicals, electric power and operators; particularly
with ozone.
Replacement of a chlorination unit with ozone would require installation
of ozonators, but the average cost for larger systems >10 MGD is not ex-
pected to exceed 1 cent per 1000 gallons. For a 1 MGD plant, ozone would
cost about 4 to 5 cents per 1000 gallons compared to 3 cents per 1000 gal-
lons for chlorination. If post-chlorination were necessary after ozonation the
cost would be additive.
C. Granular activated carbon to remove organic chemicals. The best
method yet developed for removing environmental organic contaminants
such as pesticides and aromatics from water is the use of adsorbants such
as granular activated carbon. GAC is also capable of removing trihalo-
methanes and their precursors. Installation and proper operation of GAC
would affect the concentration of a large number of chemicals in water.
The simplest, although perhaps not ultimately the most efficient, approach
applying GAC in systems already practicing filtration would be replacement
of the present media with GAC in the existing filter to a depth of at least 30
inches. Preliminary estimates indicate that approximtaely 10 tons of GAC
per MGD capacity would be needed. Chemical breakthrough rates and con-
sequent carbon reactivation frequencies have not yet been established in
large scale operations. Monitoring for chloroform breakthrough is a possi-
ble process control indicator, because chloroform is commonly present in
chlorinated water and rather weakly bound to GAC. Monitoring frequencies
would have to be at least weekly, particularly in the later stages of use near-
ing the reactivation time. If organic removal is to be maximized, the ad-
aorbant might require renewal when the NPTOC or TOC concenrtation in
the effluent exceeded 0.1 or 0.2 mg/1.
Deeper and/or countercurrent beds of GAC may be more cost effective in
the long run but time for redesign and construction would delay implemen-
tation. Since reactivation of GAC is essential to its operation, another limit-
ing factor in the short run is the almost complete lack of vailable reactiva-
tion facilities in appropriate locations.
More sophisticated operation and monitoring of GAC filtration would re-
quire personnel and apparatus not commonly available at this time. Studies
are not sufficiently advanced currently to determine the exact length of
operation before the activated carbon needs to be reactivated for a wide
101
-------
DRINKING WATER REGULATIONS
variety of chemicals. GAC can, however, lose its effectiveness for general
organic carbon removal after a few weeks as evidenced by an increase of
NPTOC values in the effluent with time, although some substances such as
polynuclear aromatics are effectively adsorbed. Current experience indicates
that the effective life of GAC for the removal of trihaloraethane precursors
is somewhat limited; a one month regeneration frequency has been generally
assumed.
Costs will vary widely depending on factors such as labor costs, carbon
costs, reactivation frequency, carbon loss due to attrition, and system size,
average production to design size ratio, and power costs. Therefore, it is
difficult to predict costs for specific systems. For a 1 MGD plant with 1
month reactivation frequency the cost might be more than 10 cents per
1000 gallons, but for large systems (over 100 MGD) between 5 and 7 cents
per 1000 gallons. This assumes that sand in existing filters is to be replaced
with GAC
Post treatment for larger systems is slightly less expensive, but installation
time would result in a substantial lag time before widespread compliance.
Cost-wiae the most serious problems appear to be with plants under 1 MGD.
The problem could be much more serious if conventional filtration units
would have to be constructed. Costs for smaller systems could be mitigated
somewhat if joint regeneration facilities could be used.
Some constraints may exist to prevent immediate and widespread instal-
lation of GAC treatment. Despite excess GAC production capacity available
at present, industry might not be able to supply enough GAC needed for
potable water treatment in the short run, if a rapid increase in demand
occurred. Questions of whether or not enough regeneration furnaces can be
produced quickly is a serious concern.
Some of the principal problems facing EPA in the control of the quality
of the Nation's drinking water are pressing the limits of current research
capabilities in health, science and technology. Regulatory decisionmaking is
further compounded by the dearth of definitive information, the lack of
agreement within the scientific community on many questions, and minimal
data on costs and impacts of changes in current technology at the national,
State, and local levels.
102
-------
APPENDIX A—DRINKING WATER REGULATIONS
PESTICIDES
A. Chlorinated Hydrocarbon Insecticides
The chlorinated hydrocarbons are one of the most important groups of
synthetic organic insecticides because of theiT wide use, great stability in
the environment, and toxicity to mammals and insects. When absorbed into
the body, some of the chlorinated hydrocarbons are not metabolized rapidly
but are stored in the fat.
As a general group of insecticides, the chlorinated hydrocarbons can be
absorbed into the body through the lungs, the gastro-intestinal tract, or the
skin. The symptoms of poisoning, regardless of the compound involved or
the route of entry, are similar but may vary in severity. Mild cases of
poisoning are characterized by headache, dizziness, gastro-intestinal dis-
turbances, numbness and weakneess of the extremities, apprehension, and
hyperirritability. In severe cases, there are muscular fasciculations spread-
ing from the head to the extremities, followed eventually by spasms involving
whole muscle groups, leading finally to convulsions and death from cardiac
or respiratory arrest. The severity of symptoms is related to the concentra-
tion of the insecticides in the nervous system, primarily the brain (1).
Criteria Based on Chronic Toxicity
Except as noted below, the approval limits (AL's) for chlorinated hydro-
carbons in drinking water have been calculated primarily on the basis of
the extrapolated human intake that would be equivalent to that causing min-
imal toxic effects in mammals (rats and dogs). Table I lists the levels of
several chlorinated hydrocarbons fed chronically to dogs and rats {2, 3, 4)
that produced minimal toxicity or no effects.
For comparison, the dietary levels are converted to mg/kg body
weight/day. Endrin and lindane had lower minimal effect/no-effect levels in
dogs than in rats; whereas, for toxaphene and methoxychlor the converse
was observed.
Human studies have also been conducted for methoxychlor, although they
were of short duration (8 weeks). The highest level tested for methoxychlor
was 2 mg/kg/day (5). No illness was reported in these subjects.
Such data from human and animal investigations may be used to derive
exposure standards, as for drinking water, by adjusting for factors that in-
fluence toxicity such as inter- and intra- species variability, length of ex-
posure, and extensiveness of the studies. To determine a "safe" exposure
level for man, conventionally a factor of 1/10 is applied to the data derived
from human exposure studies conducted longer than 2 months at which no
effects have been observed; whereas, a factor of 1/100 is applied to data
derived from human exposure studies conducted for 2 months or less as is
the case for the human methoxychlor data cited. A 1/100 factor is applied
to animal data when adequate human data are available for corroboration
and a factor of 1/500 is generally used on animal data when no adequate
103
-------
DRINKING WATER REGULATIONS
and comparable human data are available. The minimal effect levels of en-
drin, lindane, and toxaphene are adjusted by 1/500 since no adequate data
are available for comparison. These derived values are considered the max*
imum safe exposure levels from all sources. Since these values are expressed
as mg/kg/day, they are then readjusted for body weight to determine the
total quantity to which persons may be safely exposed.
Analysis of the maximum safe levels (mg/man/day) reveals that these
levels are not exactly the same when one species is compared with another.
The choice of a level on which to base an AL for water requires the selection
of the lowest value from animal experimentation, provided that the human
data are within the same order of mangnitude. Thus the human data should
substantiate the fact that man is no more sensitive to a particular agent than
is the rat or the dog.
To set a standard for a particular medium necessitates that account be
taken for exposure from other media. In case of the chlorinated hydro-
carbons, exposure is expected to occur mostly through the diet. Occasion-
ally, aerial sprays of these agents will result in their inhalation. Dietary in-
take of pesticide chemicals has been determined by the investigations of the
Food and Drug Administration from "market basket" samples of food and
water. Duggan and Corneliussen (6) report on this activity from 1964-1970.
The average dietary intakes (mg/man/day) are listed in Table I. Comparing
the intake from the diet with what are considered acceptable safe levels of
these pesticides, it is apparent that only traces of methoxychlor and toxa-
phene are present in the diet. Less than 10% of the maximum safe level of
endrin or lindane are ingested with the diet.
The AL's for chlorinated hydrocarbon insecticides reflect only a portion
of man's total exposure to the compounds. In general, 20% of the total ac-
ceptable intake is taken to be a reasonable apportionment to water. How-
ever, the AL for toxaphene was lowered because of organoleptic effects (7,
8) at concentrations above 0.005 mg/1.
The approval limits for the chlorinated hydrocarbon insecticides are listed
in Table I. These limits are meant to serve only in the event that these
chemicals are inadvertently present in the water. Deliberate addition of these
compounds is neither implied nor sanctioned.
104
-------
TABLE I. Derivation of Approval Limits (AL's) for Chlorinated Hydrocarbon Insecticides
Lowest Long-Term Levels
Calculated
Maximum Safe Levels
Intake from Diet
Water
With Minimal or
no Effects
ppm
mg/kg body
Safety
% of
% of
Recommended
Compound
Species
in diet
weight/day*
Factor (X)
mg/kg/day mg/man/dayk
mg/man/day (6) Safe Level Safe Level MAL (nig/1) *
Rat
5.0(3)
0.83
1/500
0.00166
0.1162
Endrin
Dog
1.0(3)
0.02
1/500
0.00004
0.0026*
0.00035
4.1
20
0.0002
Man
N.A.
N.A.
—
—
—
Rat
50.0(2)
8.3
1/500
0.0166
1.162
Lindane
Dog
15.0(2)
0.3
1/500
0.0006
0.042"
0.0035
8.3
20
0.004
Man
N.A.
N.A.
—
—
—
Rat
100.0(2)
17.0
1/100
0.17
11.9
Methoxychlor
Dog
4000.0(2)
80.0
1/100
0.3
21.0
T
T
20
0.1
Man
—
2.0(5)
1/100
0.02
1.4"
Rat
10.0(2)
1.7
1/500
0.0034
0.238*
T
T
20
(0.025)'
Toiapbene
Dog
400.02(2)
8.0
1/500
0.016
1.12
0.005*
Man
N.A.
N.A.
—
—
—
'Assume weight ol rat =^0.3 kg and of dog =10 kg; assume average daily food consumption of rat =0.05 kg and of dog
'Assume average weight of human adult =70 kg.
cAssume average daily intake of water for man — 2 liters.
*Chosen as basis on which to derive MAL.
'Adjusted for organoleptic effects.
'Calculated MAL in parentheses.
NA—no data available.
T--infrequent occurrence in trace tjuantilies.
-02 kg.
>
•B
PJ
2
O
X
>
a
38
fr*
z
w
St
o
>
H
M
»
M
O
e
£
H
**
O
z
(A
-------
DRINKING WATER REGULATIONS
Criteria Based on Potential Carcinogenicity
To establish AL's for Compounds such as DDT, aldrin, arid dieldrin, a
different method for deriving AL's must be used, since DDT, aldrin, and
dieldrin might represent a potential carcinogenic hazard to humans, based
on experiments with rats and mice. (9, 10, 11, 12). Aldrin is readily con*
verted to dieldrin by animals, soil microorganisms, and insects, and thus
the potential carcinogenicity of aldrin will be considered to be equivalent
to that of dieldrin (13).
It is recognized that scientists have yet to determine if there is any level
of exposure to chemical carcinogens that is completely free of risk of can-
cer. For the purpose of setting standards we will assume that the risk of
inducing cancer decreases with decreasing dose. Thus, the limits for these
possible carcinogens could be derived by estimating the health risk asso-
ciated with various concentrations and comparing these concentrations with
ambient levels to assess the attainability of the proposed limits with pres-
ently known means of technology.
Risk estimates at very low levels of exposure are subject to great un-
certainties. Extrapolation techniques such as the "one-hit" model and the
Mantel-Bryan use of the probit model (14) are being intensively reviewed
by several agencies of the federal government.
Aldrin-Dieldrin
Experiments carried out on mice (strain CF1) fed dieldrin in their daily
diet, at levels varying from 0.1 to 20 ppm during their normal life span,
resulted in significant increases in the incidence of liver tumors (11). The
results of this studp appear to be, at present, the most appropriate for cal-
culating the risk associated with a range of concentrations of dieldrin in
drinking water.
DDT:
Although earlier studies of the carcinogenic effect of DDT have yielded
generally negative results, three recent studies in experimental animals con-
flict with these previous findings. Using tumor-susceptible hybrid strains of
mice, Innes at al (15) produced significantly increased incidences of tumors
with the administration of large doses of DDT (46.4 mg/kg/day). In a
separate study in mice extending over five generations, a dietary level of
3 ppm of DDT produced a greater incidence of leukemia and malignancies
beginning with the F2 and F3 generations (16).
More recent information (12) on the effect of DDT on long-term ex-
posure in mice indicated a higher incidence of liver tumors in the treated
population. CF-1 minimal inbred mice were given technical DDT mixed
into the diet at the dose levels at 2, 10, 50 and 250 parts per million (ppm)
for the entire life span for two consecutive generations. Exposure to all
four levels of DDT resulted in a significant increase of liver tumors in
106
-------
APPENDIX A—DRINKING WATER REGULATIONS
males, this being most evident at the highest level used. In females, the
incidence of liver tumors was slightly increased following exposure to 250
ppm. In DDT-treated anijnals the liver tumors were observed at an earlier
age than in untreated controls. The age at death with liver tumors and the
incidence of liver tumors appear to be directly related to the dose of DDT
to which the mice were exposed. Four liver tumors, all occurring in DDT-
treated mice, gave metastases. Histologically, liver tumors were either well-
differentiated nodular growths, pressing but not infiltrating the surround-
ing parenchyma, or nodular growths in which the architecture of the liver
was obliterated showing glandular or trabecular patterns. The results of this
study appear to be, at present, the most appropriate to use as a basis for
extrapolating the risk associated with a range of concentrations of DDT
in drinking water.
Chlordane and Heptachior
Because recent evidence also implicates chlordane and heptachior as
potential carcinogens, establishment of limits for these pesticides must be
based on considerations similar to those for aldrin, dieldrin and DDT.
• » #
A national survey for aldrin, dieldrin and DDT in drinking water was
carried out during 1975. A total of 715 samples of raw and finished drink-
ink water were analyzed for the presence of aldrin, dieldrin, DDT, and
DDT metabolites. Dieldrin was found in 94 samples at concentration levels
of 4 ppt (minimum level of detection) to 10 ppt; 13 with levels of 11-20
ppt; 4 with levels of 21-29 ppt; and 6 with levels from 56-110 ppt. These 6
samples represented 3 raw and 3 finished waters from one location. Of
these 6 samples, 3 also contained aldrin with concentrations between 15-18
ppt. DDT at levels between 10-28 ppt was found in 6 other dieldrin-contain-
ing samples. DDT only at 15 and 32 ppt was found in 2 samples. Based on
the initial data, 30 "high potential" samples were selected and analyzed for
chlordane, heptachior and heptachior epoxide. None of them were found
above the detection limits of 5 ppt, 10 ppt and 5 ppt respectively. Note that
these ambient levels where measurable are approximately one ten thou-
sandths of the amounts that were employed in the animal tests described
above.
REFERENCES
1. Dale, W.E., Gaines, T.B., Hayes, W.M., Jr., and Pearce, G.W., Poisoning by DDT:
Relationship Between Clinical Signs and Concentrations in Rat Brain, Science
142: 1474 (1963).
2. Lehman, A.J., Summaries of Pesticide Toxicity. Association of Food and Drag
Officials of the U.S., Topeka, Kansas, 1965, pp. 1-40.
3. Treon, J.F., Cleveland, F.P., and Cappel, J.t Toxicity of Endrin for Laboratory
Animals, /, Agr. Food Chem 3, 842-848, 1955.
4. Unpublished Report o( Kettering Laboratory, University of Cincinnati, Cincinnati,
Ohio. Cited in "Critical Review of Lterature Pertaning to the Insecticide Endrin/'
107
-------
DRINKING WATER REGULATIONS
a dissertation for the Master's Degree at the University of Cincinnati by J. Cole,
1966.
5. Stein, A.A., Serrone, D.M., and Coulston, F., Safety Evaluation of Metboxychlor in
Human Volunteers. Toxic Appl. Pharmacol. 7 : 499, 1965.
6. Duggan, R.E. and Corneliussen, P.E. Dietary Intake of Pesticide Chemicals in
the United States (III), June 1968-April 1970. Pesticides Monotoring Journal 5
(4): 331-341, 1972.
7. Cohen, J.M., Rourke, G.A., and Woodward, R.L.: Effects of Fish Poisons on Water
Supplies. /. Amer. Water Works Assn. 53(1) : 49-62, 1961.
8. Sigworth, E.A., Identification and Removal of Herbicides and Pesticides. J. Amer.
Water Works Assn. 57(8) : 1016-1022, 1965.
9. Fitzhugh, O.G., Nelson, A.A., and Quaife, M.L. Chronic Oral Toxicity of Aldrin
and Dieldrin in Rats and Dogs. Fed. Cosm. Toxicol. 2:551, 1964.
10. Walker, A.I.T., Stevenson, D.E., Robinson, J., Thorpe, E., and Roberts, M. Phar-
macodynamics of Dieldrin (HEOD)-3: Two Year Oral Exposures of Rats and
Dogs. Toxicology and Applied Pharmacology, 15:345, 1969.
11. Walker, A.I.T., Thorpe, E., and Stevenson, D.E. The Toxicology of Dieldrin
(HEOD): Long-Term Oral Toxicity Experiments in Mice,- Fd. Cosmet. Toxicol
Vol. 11, pp. 415-432, 1972.
12. Tomatis, L., Turusov, V., Day, N., Charles, R.T. The Effect of Long-Term Ex-
posure to DDT on CF-1 Mice. Int. J. Cancer 10, 489-506, 1972.
13. Menzie, Calvin M. Metabolism of Pesticides Publ. by Bureau of Sport Fisheries
and Wildlife, SSR-Wildlife 127, Washington, D.C.; 24, 1969.
14. Mantel, N., and Bryan, W.R. "Safety" Testing of Carcinogenic Agents. /. Nat.
Cancer Inst. 27:455, 1961.
15. Innes, J.R.M., Ulland, B.M., Valerio, M.G., Petrucelli, L., Fishbein, L., Hart, E.R.,
Pallotta, A .J., Bates, R.R. Falk H.L., Gart, J.J., Klein, M., Mitchell, I., and Peters,
J. Bioassay of Pesticides and Industrial Chemicals for Tumorigenicity in Mice:
A Preliminary Note, J. Nat. Cancer Inst. 42 1101, 1969.
16. Tarjan, R. and Kemeny, T., Muitigeneration Studies on DDT in Mice. Fd. Cosmet
Toxicol. 7:215, 1969.
B. Chlorophenoxy Herbicides
Aquatic weeds have become substantial problems in the U.S. in recent
years, and chemical control of this vegetation has won wide acceptance.
Since waters to which applications of herbicides are made are sometimes
employed as raw water sources of drinking water, there is the possibility
that herbicides may enter potable source water. Consequently, a standard
is needed for the more extensively used herbicides so as to protect the
health of the water consumer.
Two widely used herbicides are 2, 4-D i2, 4-dichlorophenoxyacetic acid)
and 2, 4, 5-TP (silvex) [2-{2, 4, 5-trichlorophenoxy) propionic acid].
[A closely related compound, 2, 4, 5-T (2, 4, 5>trichlorophenoxyacetic acid)
had been extensively used at one time, but has been banned for major
aquatic uses.] Each of these compounds is formulated in a variety of salts
and esters that may have a marked difference in herbicidal properties, but
all of which are hydrolyzed rapidly to the corresponding acid in the body.
108
-------
APPENDIX A—DRINKING WATER REGULATIONS
The acute toxicity following oral administration to a number of experi-
mental animals is moderate. Studies (14) of the acute oral toxicity of the
chlorinated phenoxyalkyl acids indicate that there is approximately a three-
fold variation between the species of animals studied. It appears that acute
oral toxicity of the three compounds is of about the same magnitude within
each species (e.g., in the rat, an oral LD of about 500 mg/kg for each
agent).
The subacute oral toxicity of chlorophenoxy herbicides has been in-
vestigated in a number of species of experimental animals (1-6). The dog
was the most sensitive species studied and often displayed mild injury in
response to doses of 10 mg/kg/day for 90 days, and serious effects from
a dose of 20 mg/kg/day for 90 days. Lehman (6) reported that the no-
effect level of 2, 4-D is 50* mg/kg/day in the rat, and 8.0 mg/kg/day
in the dog.
Although 2, 4, 5-T has been banned for all aquatic uses there is con-
siderable interest as to why this action was taken, so for informational
purposes, a discussion of the toxicity of this herbicide is included. In a
study of various pesticides and related compounds for teratogenic effects,
Courtney, et al. (7) noted terata and embryotoxicity from 2, 4, 5*T. These
effects were evidenced by statistically increased proportions of litters affected
and of abnormal fetuses within the litters (notably, cleft palate and cystic
kidneys). Effects were noted in both mice and rats, although the rat ap-
peared to be more sensitive to this effect. A dosage of 21.5 mg/kg produced
no harmful effects in mice, while a level of 4.6 mg/kg caused minimal, but
statistically significant, effects in the rat. More recent work (8) has in-
dicated that a contaminant (2, 3, 7, 8-tetrachlorodibenzo-p-dioxin) which
was present at approximately 30 ppm in the 2, 4, 5-T formulation origin-
ally tested was highly toxic to experimental animals and produced fetal and
maternal toxicity at levels as low as 0.005 mg/kg. However, purified 2, 4,
5-T has also produced teratogenic effects in both hamsters and rats at rel-
atively high dosage rates (9). Current production samples of 2, 4, 5-T that
contain less than 1 ppm of dioxin did not produce embryotoxicity or terata
in rats at levels as high as 24 mg/kg/day (10).
The subacute and chronic toxicity of 2, 4, 5-TP has been studied in ex-
perimental animals (11). The results of 90-day feeding studies indicate that
the no-effect levels of the sodium and potassium salts of 2, 4, 5-TP are 2
mg/kg/day in rats and 13 mg/kg/day in dogs. In 2-year feeding studies
with these same salts, the no-effect levels were 2.6 mg/kg/day in rats and
0.9 mg/kg/day jn dogs.
Some data are available on the toxicity of 2, 4-D to man. A daily dosage
of 500 mg (about 7 mg/kg) produced no apparent ill effects in a volunteer
•In the March 14, 1975, issue of this document, this figure was erroneously written
at 0.5.
109
-------
DRINKING WATER REGULATIONS
over a 21-day period (12). When 2, 4-D was investigated as a possible
treatment fur disseminated coccidioidomycosis, the patient had no side
effects from IS intravenous doses during 33 days; each of the last 12 doses
in the series was 800 mg (about 15 mg/kg) or more, the last being 2000
mg (about 37 mg/kg) (13). A nineteenth and final dose of 3600 mg (67
mg/kg) produced mild symptoms.
The acute oral dose of 2, 4-D required to produce symptoms in man is
probably 3000 to 4000 mg (or about 45 to 60 mg/kg). A comparison of
other toxicity values for 2, 4, 5-TP indicates that the toxicity of these two
agents is of the same order of magnitude. Thus, in the absence of any
specific toxicologic data for 2, 4, 5-TP in man, it might be estimated that
the acute oral dose of 2, 4, 5-TP required to produce symptoms in man
would also be about 3000 to 4000 mg.
In addition to these specific data, the favorable record of use experience
of 2, 4-D is also pertinent. Sixty-three million pounds of 2, 4-D were pro-
duced in 1965 while there were no confirmed cases of occupational poison-
ing and few instances of any illness due to ingestions (14, 15). One case of
2,4-D poisoning in man has been reported by Berwick (16).
Berwick (16).
Table 1 displays the derivation of the approval limits for the two chlor-
ophenoxy herbicides most widely used. The long-term no-effect levels
(mg/kg/day) are listed for the rat and the dog. These values are adjusted
by 1/500 for 2, 4-D and 2, 4, 5-TP. The safe levels are then readjusted to
reflect total allowable intake per person. Since litle 2, 4-D or 2, 4, 5-TP are
expected to occur in foods, 20% of the safe exposure level can be reason-
ably allocated to water without jeopardizing the health of the consumer.
The approval limits for these herbicides are meant to serve in the event
that these chemicals inadvertently occur in the water. Deliberate addition of
these compounds to drinking water sources is neither implied not
sanctioned.
110
-------
Table I. Derivation of Approval Limits (AL) for Cklorophenoxy Herbicides
Lowest Long-Term
Levels with
Minimal or No Effects
Calculated Maximum Safe Levels
From all Sources of Exposure
Water
Compound
Species
mg/kg/day*
Safety
Factor (X)
mg/kg/day
mg/m an/da yk
% of
Safe Level
AL
(mg/1)*
Rat
50 (6)
1/500
0.1
7.0
2, 4-D
Dog
8.0 (6)
1/500
0.016
1.12*
20
0.1
2, 4, 5-TP
Rat
2.6 (12)
1/500
0.005
0.35
Dog
0.9 (12)
1/500
0.002
0.14*
20
0.01
%
u
pi
«
O
8
T
S
as
W
H
2
O
3
>
W
»
SO
H
O
"Assume weight of rat **0.3 kg and of dog =10 kg; assume average daily food consumption of rat =0.05 kg and of dog =0.2 kg.
kAssume average weight of human adult = 70 kg.
'Assume average daily intake of water for man -=2 liters.
*Chosen as basis on which to derive AL.
H
1-4
©
55
V
-------
DRINKING WATER REGULATIONS
REFERENCES
1. Hill, E.C. and Carlisle, H. Toxicity of 2, 4-DichIorophenoxyacetic Acid for Ex-
perimental Animals. J. Industr. Hyg. Toxicol. 29, 85-95, 1947.
2. Lehman, A J. Chemicals in Foods: A Report to The Association of Food and Drug
Officials on Current Development. Part II. Pesticides. Assoc. Food Drug Off. U.S.,
Quart. Bull. 15, 122-133, 1951.
3. Rowe, V.K. and Hymas, T.A. Summary of Toxicological Information on 2, 4-D and
2, 4, 5-T Type Herbicides and an Evaluation of the Hazards of Livestock Asso-
ciated with Their Use. Amer. J. Vet. Res. 15, 622-629, 1954.
4. Drill, V. A. and Hiratzfca, T. Toxicity of 2, 4-DichJorophenoxyacetic Acid and 2,
4, 5-Trichloropbenoxyacetic Acid: A Report of Their Acute and Chronic Toxicity
in Dogs. Arch. Industr. Hyg. Occup. Med. 7, 61-67, 1953.
5. Palmer, J.S. and Radeleff, R.D. The Toxicologic Effects of Certain Fungicides and
Herbicides on Sheep and Cattle. Ann N.Y, Acad. Sci. Ill, 729*736, 1964.
6. Lehman, A J. Summaries of Pesticide Toxicity. Association of Food and Drug
Officials of the U.S., Topeka, Kansas, 1965, pp. 13-14.
7. Courtney, K.D., Gaylor, D.W., Hogan, M.D., and Falk, H.L. Teratogenic Evaluation
of 2, 4, 5-T. Science 168, 864, 1970.
8. Courtney, K.D., and Moore, J.A. Terratology Studies with 2, 4, 5-Trichloropheno-
xyacetic Acid and 2, 3, 7, 8-Tetrachlorodibenzo-p-dioxin. Toxicology and Applied
Pharmacology 20, 396, 1971.
9. Collins, T.F.X. and Williams, C.H. Teratogenic Studies with 2, 4, 5-T and 2, 4-D
in Hamsters. Bull, of Environmental Contamination and Toxicology 6 (6) :559-
567, 1971.
10. Emerson, J.L., Thompson, D.J., Gerbig, C.G., and Robinson, V,B. Teratogenic Study
of 2, 4, 5-Trichlorophenoxy Acetic Acid in the Rat. To*. Apt, Pharmacol. 17,
311, 1970.
11. Mullioon, W.R, Some Toxicological Aspects of Silvex. Paper Presented at South-
ern Weed Conference, Jacksonville, Fla., 1966.
12. Krauc, as cited by Mitchess, J.W., Hogson, R.E., and Gaetjens, C.F. Tolerance of
Farm Animals to Feed Containing 2, 4-Dichlorophenoxyacetic Acid. J. Animal. Sci.
5, 226-232, 1946.
13. Seabury, J.H. Toxicity of 2, 4-DichIorophenoxyacetic Acid for Man and Dog. Arch.
Envir. Health 7, 202-209, 1963.
14. Hayes, WJ., Jr. Clinical Handbook on Economic Poisons. PHS Pub, No. 476, U.S.
Government Printing Office, Washington, D.C., revised 1963.
15. Nielson, K., Kaempe, B., and Jensen-Holm, J. Fatal Poisoning in Man by 2,
4-Dichlorophenoxyacetic Acid (2, 4-D); Determination of the Agent in Forensic
Materials. Acta Pharmacol. Tax. 22. 224-234, 1965.
16. Berwick, P., 2, 4-Dichlorophenoxyacetic Acid Poisoning in Man. LA.M.A, 214 (6):
114-117, 1970.
112
-------
APPENDIX A—DRINKING WATER REGULATIONS
SELENIUM
The 1962 Drinking Water Standards Committee lowered the limit for
selenium in drinking water primarily out of concern over the possible car-
cinogenic properties of the element. Data supporting the carcinogenicity of
selenium has not been forthcoming, and more recent findings concerning
the nutritional requirement for selenium has required a comprehensive
review of the data available concerning the toxicity of selenium and its
compounds.
The controversy over the present limits of selenium acceptable in the en-
vironment is largely the result of the demonstration by Schwartz and Foltz
(2) that the element was an integral part of "factor 3/' recognized for some
time as essential in animal nutrition. While definite evidence is still lack-
ing for a nutritional requirement for selenium in man, certain cases of pro-
tein-resistant kwashiorkor have been shown to be responsive to adminis-
tration of the element (3).
Consideration of a maximal concentration of selenium allowable in drink-
ing water is further complicated by the many secondary factors known to
affect both the efficacy of selenium in alleviating deficiency syndromes and
the intakes associated with toxicity. The chemical form of selenium (4), the
protein content of the diet (5), the source of dietary protein (6), the pres-
ence of other trace elements (1, 7, 8), and the vitamin E intake (9, 10, 11)
all affect the beneficial and/or adverse effects of selenium in experimental
animals. The fact that these interactions are not simple is illustrated by the
comments of Frost (1) on the well-known antagonism of arsenic in selenium
toxicity (1, 7, 8, 12). He has found that arsenic in drinking water accentu-
ates the toxicity of selenium in drinking water in contrast to the protective
effect of arsenic seen when selenium was administered via the diet. Conse-
quently, when considering "safe" levels of selenium in drinking water, con-
sideration must also be given to the variability in these other factors which
are certain to occur in any given population.
The current limit of 0.01 mg/liter of selenium in drinking water is based
on the total selenium content. No systematic investigation of the forms of
selenium in drinking water sources with excessive concentrations has ever
been carried out. Since elemental selenium must be oxidized to selenite or
selenate before it has appreciable solubility in water (13), one would predict
that these would be the principal inorganic forms that occur in water. Or-
ganic forms of selenium occur in seleniferous soils and have sufficient mo-
bility in an aqueous environment to be preferentially absorbed over selenate
in certain plants (14). However, the extent to which these compounds might
occur in source waters is essentially unknown.
There is considerable difficulty involved in determining what the required
level and toxic levels of selenium intake in humans might be. The basic
problem is that dietary selenium includes an unknown variety of selenium
compounds in varying mixtures. Toxicologic examination of plant sources of
113
-------
DRINKING WATER REGULATIONS
selenium has revealed that selenium present in seleniferous grains is more
toxic than inorganic selenium added to the diet (16), Although there is a
fairly extensive literature on industrial exposures to selenium (see Cerwenka
and Cooper, 1961 (17), and Cooper, 1967 (18) for reviews of this subject),
the results do not apply well to environmental exposures since the only
studies that made an attempt to document systemic absorption involved
elemental selenium (19). Elemental selenium is virtually non-toxic to plants
and animals that have been shown to be very sensitive to the water soluble
forms of selenium.
Only one documented case of human selenium toxicity for a water source
uncomplicated with selenium in the diet has been reported (21), Members
of an Indian family developed loss of hair, weakened nails, and listlessness
after only 3 months' exposure to well-water containing 9 mg/1. The children
in the family showed increased mental alertness after use of water from the
seleniferous well was discontinued, as evidenced by better work in school
(22).
Smith and co-workers (23, 24) reported the results of their studies deal-
ing with human exposure to high environmental selenium concentrations in
the 1930*s. They reported a high incidence of gastrointestional problems,
bad teeth, and an icteriod skin color in seleniferous areas. The individuals
exhibiting these symptoms had urinary selenium levels of 0.2-1,98 ug/liter
as compared to the 0.0*0.15 ug/liter that Glover (19) indicates to be the
normal range. The gastrointestinal disturbances and the icteriod discolora-
tion of the skin apparently have their counterparts in the anorexia (23)
and bilirubinemia (7), respectively, in rats fed selenium. The effect of
selenium on teeth has had some marginal documentation in rats (26); and
has been supported by Hadjimarkos (27) and refuted by Cadell and Cous-
ins (28) in epidemiologic studies.
From urinary concentrations of selenium, Smith and Westfall (24) esti-
mated that the individuals displaying these symptoms were ingesting 0.01
to 0.10 mg/kg/day, and possibly as much as 0.20 mg/kg/day. For the
70 kg man, this would amount to a daily intake of 700 to 7000 ug/day.
Smith (24, 29) also presented the range of selenium concentrations found
in various food classes in the areas in which the field studies had been con-
ducted. With the use of the table provided in Dietary Levels of Households
in the {7.S., Spring 1965 (U.S.D.A. Agri. Res. Service), calculations from
these data result in a range of intake of 600-6300 ug/day, very close to the
estimates made from urinary concentrations of selenium. These intakes of
selenium correspond in the main with the levels producing adverse effects
in other mammalian species. Tinsley et al. (25) found that an intake of
0.125 mg/kg/day adversely affected early growth in rats. 1.1 mg/kg, ad-
ministered twice weekly (ca. 0.3 mg/kg/day), has been found to adversely
affect growth and to increase mortality in Hereford steers (30). Mortality
in ewes was increased at 0.825 mgAg/day. The steers were administered
114
-------
APPENDIX A—DRINKING WATER REGULATIONS
sodium selenite; the ewes sodium selenate. Although these levels are
slightly higher than those reported for the human exposures, it must be
remembered that the parameters measured would not be acceptable either
in terms of severity or incidence in the human population.
Few studies have been performed to specifically examine the toxicity of
selenium administered in drinking water. Pletnikova (31) found the rabbit
to be very sensitive to selenium as selenite. Ten ug/1 in drinking water re-
sulted in a 40% reduction in the elimination of bromosulphalein by the
liver. Since no apparent consideration was given to the selenium content of
the diet of these animals, the meaning of this result in terms of liver func-
tion is obscure. If the sole intake of selenium were from the water in these
studies, the controls had to be deficient and the experimental group mar-
ginal, at best, in terms of the dietary requirement for selenium. The dura-
tion of the study was 7y» months. Schroeder (32) has indicated that intake
of selenite from drinking water is more toxic than when mixed with food.
However, this suggestion was not based on a direct experimental compari-
son. Rosenfeld and Beath (33) studied the effects of sodium selenate in
drinking water on reproduction in rats. Selenium concentrations of 2.5
mg/1 reduced the number of young reared by the second generation of
mothers, and 7.5 mg/l prevented reproduction in females.
Early work (34), using both naturally occurring, and a selenide salt,
indicated the formation of adenomas and low-grade non-metastasizing hepa-
tic cell carcinomas in 11 of 53 rats surviving 18 months of diets containing
selenium. Harr et al. (24), in a much more extensive study using selenite
and selenate salts, found no evidence of neoplasms that could be attributed
to the addition of these selenium compounds to the diet at 0.5 • 16 ppm.
Volganev and Tschenkes (35) negated their earlier results, which had indi-
cated that 4.3 mg/1 selenium as selenite in the diet gave rise to tumors, but
had not used proper controls. It should be noted that these studies are not a
direct negation of the earlier studies implicating selenium as a carcinogen,
since entirely different compounds of selenium were used in the early work.
Consequently, the possibility that other compounds of selenium, besides
selenite, possess carcinogenic properties cannot be strictly ruled out. The
carcinogenic properties of selenium are further complicated by recent re-
ports of the effectiveness of selenium, 1 mg/I (as selenite), in reducing pa-
pillomas induced by various chemicals in mice (36).
Any consideration of a maximum allowable concentration of selenium
must include the evidence that the element is an essential dietary require-
ment. A range of 0.04 to 0.10 mg/1 in the diet is considered adequate to
protect animals from the various manifestations of selenium deficiency
(10, 37, 38). Using the recent data on Morris and Levander (39), an esti-
mate of the present average daily intake of selenium by the American pop-
ulation may be calculated. This figure approximates 200 ug/day and some
variation around this figure would be anticipated primarily as the result
115
-------
DRINKING WATER REGULATIONS
of individual preferences, particularly in meats. Since no deficiency diseases
of selenium have been reported to date in the U.S., it may be assumed that
200 ug/day of selenium is nutritionally adequate.
Signs of selenium toxicity have been seen at an estimated level of se-
lenium intake of 0.7-7 mg/day according to the data of Smith et al. (23, 24).
At the present limit on selenium content of drinking water, water would
increase the basal 200 ug/day intake of selenium by only 10%, if one as-
sumes a 2-liter ingestion of water per day. This results in a minimum safety
factor of 3, considering the lower end of the range of selenium intakes
that have been associated with minor toxic effects in man. In view of the
relative scarcity of data directly applicable to the apparent small margin of
safety brought about by selenium contained in the diet, selenium concentra*
tions above 0.01 mg/liter shall not be permitted in the drinking water.
REFERENCES
1. Frost, Douglas V, (1967) Significance of The Symposium. In Symposium; Selenium
in Biomedicine, O.H. Muth. Ed. AVI Publishing Co., Inc., West port, Conn. p. 7-26.
2. Schwant, K. and Foltz, C.M., Selenium Aa An Integral Part of Factor 3 Against
Dietary Necrotic Liver Degeneration. J. Am. Cbem. Soc. 79:3292.
3. Hopkins, L.L., Jr., and Majaj, A.S. (1967) Selenium in Human Nutrition in Sym-
posium; Selenium in Biomedicine. O.H. Muth, Ed. AVI Publishing Co., Inc., West-
port, Conn. p. 203-214,
4. Schware, K. and Fvega, A. (1969) Biological Patterny of Organic Selenium Com-
pounds. I. Aliphatic Moneseleno and Diseleno Dicarboxific Acids. J. Biol. Chem.
244, 2103-2110.
5. Smith, M.I. (1939) The Influence of Diet on The Chronic Toxicity of Selenium.
Public Health Report HJ.S.) 54, 1441-1453.
6. Levander, O.A., Young, MX. and Meeks, S.A. (1970) Studies on The Binding of
Selenium by Liver Homogenatea From Rats Fed Diets Containing Either Casein or
Casein Plus Unseed Oil Meal. Toxicol, and Appl. Pharmacol. 16, 79-87.
7. Haiverson, A.W., Tsay, Ding-Tsair, Triebevaaser, K.C. and Whitehead, E.I. (1970)
Development of Hemolytic Anemia in Rat* Fed Selenite. Toxicol, and Appl,
Pharmacol. 17, 151-159.
8. Levander, O.A. and Bauman, C.A. (1966) Selenium Metabolism VI. Effect of Ar-
senic on The Excretion of Selenium in the Bile. Toxicol, and Appl. Pharmacol.
9, 106-115.
9. Levander. O.A. and Morris, V.C. (1970) Interactions of Methionine, Vitamin E,
and Antioxidants in Selenium Toxicity in the Rat. J. Nutrition 100:111 -1118.
10. Schwarz, K. (1960) Factor 3, Selenium, and Vitamin E. Nutrition Reviews. IS,
193-197.
11. Sondegaard, Ebbe. (1967) Selenium and Vitamin E. Interrelationships In Sym-
posium: Selenium in Biomedicine AVI Publishing Co., Inc., Westport, Conn. O.H.
Muth Ed. pp. 365-381.
12. Moxon, A.L, DuBois, K.P. and Potter, R.L. (1941) The Toxicity of Optically In-
active d, and 1-Selenium Cystine. J, Pharm. and Exp. Ther. 72, 184-195.
13. Lakin, Hubert W. and Davidson, David F. (1967) The Relation of The Geo-
chemistry of Selenium to Its Occurrence in Soils. In Symposium: Selenium in
Biomedicine p. 27-56.
14. Hamilton, John W. and Beath, O.A. (1964) Amount and Chemical Form of Selen-
ium in Vegetable Plants, Agr. and Food Chem, 12, 371-374.
116
-------
APPENDIX A—DRINKING WATER REGULATIONS
15. Olson, O.E. (1967) Soil, Plant, Animal Cycling of Excessive Levels of Selenium.
In Symposium: Selenium in Biomedicine (AVI Publishing Co., Inc., Westport,
Conn.) O.H. Muth, Ed. pp. 297-312.
16. Franke and Potter (1935) J. Nutr. 10, 213.
17. Cerwenka, Edward, A., Jr. and Cooper, W. Charles (1961) Toxicology of Selen-
ium on Tellurium and Their Compounds. Arch. Environ. Hlth. 3, 71-82.
18. Cooper, W. Charles (1967) Selenuim Toxicity in Man. In Symposium: Selenium in
Biomedicine. (AVI Publishing Co., Inc., Westport, Coaa.) O.H. Much. Ed. pp.
185-199.
19. Glover, J.R. (1967) Selenium in Human Urine: A Tentative Maximum Allowable
Concentration for Industrial and Rural Populations. Ann Occup. Hyg. 10, 3*14.
20. Schwarz, K. and Foitz, C.M. (1958) Factor 3 Activity of Selenium Compounds.
J. Biol. Chem. 233, 245.
21. Beath, O.A. (1962) Selenium Poisons Indians. Science News Letter 81, 254.
22. Rosenfeld, L and Beath, O.A. (1964) Selenium, Geobotany, Biochemistry, Toxic-
icity and Nutrition. Academic Press, N. Y. and London.
23. Smith, M.I., Franke, K.W. and Westfall, B.B. (1936) The Selenium Problem in
Relation to Public Health. Public Health Reports. (U.S.) 51, 1496-1505.
24. Smith, M.I. and Westfall, B.B. (1937) Further Field Studies on The Selenium
Problem in Relation to Public Health. Public Health Report (U.S.) 52, 1375-1384.
25. Tinsley, I.J., Harr, J.R., Bone, J.F., Weswig, P.H. and Yamamoto, R.S. (1967)
Selenium Toxicity in Rats I. Growth and Longevity. In Symposium: Selenium in
Biomedicine (AVI Publishing Co., Inc., Westport, Conn.) O.H. Muth, Ed. pp.
141-152.
26. Wheatcraft, M.G., English, J.A. and Schlack, C.A. (1951) Effects of Selenium on
The Incidence of Dental Caries in White Rats. J. Dental Res. 30, 523-524.
27. Hadjimarkos, D.M. (1965) Effect of Selenium on Dental Caries. Arch. Environ.
Health 10, 893-899.
28. Cadell, P.B. and Cousins, F.B. (I960) Urinary Selenium and Dental Caries Nature
185, 863.
29. Smith, M.I. (1941) Chronic Endemic Selenium Poisoning, J.A.M.A. 116, 562-567.
30. Magg, D.D. and Glenn, M.W. (1967) Toxicity of Selenium: Farm Animals. In
Symposium: Selenium in Biomedicine (AVI Publishing Co., Inc., Westport, Cona.)
O.H. Muth, Ed. pp. 127-140.
31. Pletnikova, LP. (1970 Biological Effect and Safe Concentration of Selenium in
Drinking Water. Hygiene and Sanitation 35, 176-181.
32. Schroeder, Henry A. (1967) Effects of Selenate, Selenite and Tellurite on The
Growth and Early Survival of Mice and Rats. J. Nutri. 92, 334, 338.
33. Rosenfeid, I. and Beath, O.A. (1954) Effect of Selenium on Reproduction in
Rats. Proc. Soc. Expl Bio.) and Medi. 87, 295-299.
34. Fitzhugh, O.G., Nelson, A.A. and Bliss. C.I. (1944) The Chronic Oral Toxicity of
Selenium. J. Pharmacol. 80, 289*299.
35. Volganev, M,N. and Tschenkes, L.A. (1967) Further Studies in Tissue Changes
Associated With Sodium Selenate. In Symposium: Selenium in Biomedicine (AVI
Publishing Co., Inc., Westport, Conn.) O.H. Muth, Ed. pp. 179-184.
36. Shamberger, R.J. (1970) Relationship of Selenium to Cancer I. Inhibitory Effect
of Selenium on Carcinogensis J. Natl. Cancer Inst. 44, 931-936.
37. Nesheim, M.C. and Scott, M.L. (1961) Nutritional Effects of Selenium Compounds
in Chicks and Turkeys. Fed. Proc. 20, 674-678.
38. Oldfield, J.E., Schubert, J.R., and Muth, O.H. 11963) Implications of Selenium in
Large Animal Nutrition. J. Agr, Food Chem. II, 388-390.
39. Morris, V.C, and Levander, O.A. (1970) Selenium Content of Foods J. Nutri. 100,
1383-1388.
117
-------
DRINKING WATER REGULATIONS
SILVER
The need to set a water standard for silver (Ag) arises from its inten-
tional addition to waters as a disinfectant. The chief effect of silver in the
body is cosmetic. It consists of a permanent blue-grey discoloration of the
skin, eyes, and mucous membranes which is unsightly and disturbing to the
observer as well as to the victim. The amount of colloidal silver required to
produce this condition (argyria, argyrosis), and to serve as a basis of deter-
mining the water standard, in not known, however, but the amount of silver
from injected Ag-arsphenamine, which produces argyria is precisely known.
This value is any amount greater than 1 gram of silver, 8g Ag-arsphenamine,
in an adult (X, 2).
From a review (2) of more than 200 cases of argyria, the following addi-
tional facts were derived. Most common salts of silver produce argyria when
ingested or injected in sufficient doses. There is a long-delayed appearance
of discoloration. No case has been uncovered that has resulted from an
idiosyncrasy to silver. There was, however, considerable variability in pre-
disposition to argyria; the cause of this is unknown, but individuals concur-
rently receiving bismuth medication developed argyria more readily. Al-
though there is no evidence that gradual deposition of silver in the body
produces any significant alteration in physiologic function, authorities are
of the opinion that occasional mild systemic effects from silver may have
been overshadowed by the striking external changes. In this connection,
there is a report (3) of implanted silver amalgams resulting in localized
argyria restricted to the elastic fibers and capillaries. The histopathologic
reaction resembled a blue nevus simulating a neoplasm with filamentous
structures and globular masses. Silver affinity for elastic fibers had been
noted a half-century earlier (5).
A study (5) of the metabolism of silver from intragastric intake in the
rat, using radio-silver in carrier-free tracer amounts, showed absorption to
be less than 0.1-0.2 percent of the silver administered; but this evidence is
inconclusive because of the rapid elimination of silver when given in carrier-
free amounts. Further study indicated, however, that silver is primarily ex-
creted by the liver. This would be particularly true if the silver were in col-
loidal form. Silver in the body is transported chiefly by the blood stream
in which the plasma proteins and the red cells carry practically all of it in
extremely labile combinations. The half-time of small amounts of silver in
the blood stream of the rat was about I hour. A later report (6), using the
spcetrographic method on normal human blood, showed silver unmistakably
in the red blood cell and questionably in the red cell ghosts and in the
plasma. Once silver is fixed in the tissues, however, negligible excretion oc-
curs in the urine {7).
A study (8) of the toxicologic effects of silver added to drinking water
of rats at concentrations up to 1,000 ug/1 (nature of the silver salt unstated)
118
-------
APPENDIX A—DRINKING WATER REGULATIONS
showed pathologic changes in kidneys, liver, and spleen at 400, 700, and
1,000 ug/1, respectively.
A study (9) of the resorption of silver through human skin using radio-
silver Ag111 has shown none passing the dermal barrier from either solution
(2 percent AgN03) or ointment, within limits of experimental error (s2
percent). This would indicate no significant addition of silver to the body
from bathing waters treated with silver.
Uncertainty currently surrounds any evaluation of the amount of silver
introduced into the body when silver-treated water is used for culinary pur-
poses. It is reasonable to assume that vegetables belonging to the family
Brassicoceae, such as cabbage, turnips, cauliflower, and onions, would
combine with residual silver in the cooking water. The silver content of sev-
eral liters of water could thus be ingested.
Because of the evidence (7) that silver, once absorbed, is held indefinitely
in tissues, particularly the skin, without evident loss through usual channels
of elimination or reduction by transmigration to other body sites, and be*
cause of the probable high absorbability of silver bound to sulfur com-
ponents of food cooked in silver-containing waters [the intake for which
absorption was reported in 1940 to amount to 60-80 ug per day (10) ], the
concentration of silver in drinking water shall not exceed 0.05 mg/1.
REFERENCES
1. Hill, W.B., and Pillabury, D.M. Argyria. The PharmocoJogy of Silver. Baltimore,
Md., Williams and Wilkins, 1939, 172 pp.
2. Ibid, Argyria Inveatigation-Toxicologic Properties of Silver, Am. Silver Producers
Res. Proj. Report. Appendix II, (1957).
3. Bell, C.D., Cookey, D.B., and Nickel, W.R. Amalgam Tatoo-localized Argyria.
A.M.A. Arch Derm. Syph. 66: pp. 523*525 (1952).
4. Joseph, M., and Van Deventer, J.B. Atlas of Cutaneous Morbid Histology. W.T.
Kliner & Co., Chicago, 1906.
5. Scott, K.G., and Hamilton, J.G. The Metabolism of Silver in The Rat With Radio-
silver Used As Indicator. U. of Cal. Publ. in Pharm. 2: pp. 241-262 (1950).
6. Wyckoff, R.C., and Hunter. F.R. Spectrographs Analysis of Human Blood. Arch.
Biochem. 63: pp. 454460 (1956),
7. Aub. J.C. and Fairhall, LT. Excretion of Silver in Urine. J.A.M.A. 118: p. 319
(1942).
8. Just, J. and Szniolis, A. Germicidal Properties of Silver in Water. J. Am. Water
Works A., 28 : 492-506, April 1936.
9. Norgaard, 0. Investigations with Radio Ag Into the Resorption of Silver Through
Human Skin. Acts Dermatovener 34 : 4115-4119 (1945).
10. Kehoe, R.A., Cholak, J., and Story, R.V. Manganese, Lead, Tin, Copper and Silver
in Normal Biological Material. J. Nutr. 20 : 85-98 (1940).
119
-------
DRINKING WATER REGULATIONS
SODIUM
Man's intake of sodium is mostly influenced by the use of salt. Intake of
sodium chloride for American males is estimated to be 10 grams per day,
with a range of 4 to 24 grams (1). This would be a sodium intake of 1600
to 9600 mg per day. Intake of these amounts is considered by most to have
no adverse effect on normal individuals. Even Dahl, who has been one of
the strong advocates of the need for restricting salt intake, has felt that an
intake of 2000 mg of sodium could be allowed for an adult without a family
history of hyertension. Intake of sodium from hospital "house" diets has
been measured recently (2). The sodium content of a pool of 21 consecutive
meals that were seasoned by the chef or the dietitian from twenty selected
general hospitals was determined each quarter. The average sodium intake
per capita per day was 3625 - 971 (SD) milligrams. The intake could be
greatly changed between individuals who never add salt to the food at the
table and the individuals who always add salt even before tasting.
The taste threshold of sodium in water depends on several factors (3).
The predominant anion has an effect; the thresholds for sodium were
500 mg/1 from sodium chloride, 700 mg/1 from sodium nitrate, and 1000
mg/1 from sodium sulfate. A heavy salt user had a threshold of taste that
was 50 percent higher, and the taste was less detectable in cold water.
Six of 14 infants exposed to a sodium concentration of 21, 140 mg/1 died
when salt was mistakenly used for sugar in their formula (4). Sea water
would have about 10, 000 mg/1 of sodium.
Severe exacerbation of chronic congestive heart failure due to sodium in
water has been documented (3). One patient required hospitalization when
he changed his source of domestic water to one that had 4200 mg/1 sodium.
Another patient was readmitted at two-to-three-week intervals when using a
source of drinking water of 3500 mg/l sodium.
Sodium*restricted diets are used to control several disease conditions of
man. The rationale, complications, and practical aspects of their use were
reviewed by a committee on food and nutrition of the National Research
Council (5). Sodium-restrictive diets are essential in treating congestive
cardiac failure, hypertension, renal disease, cirrhosis of the liver, toxemias
of pregnancy, and Meniere's disease.
Hormone therapy with ACTH and cortisone is used for several diseases.
Sodium retention is one of the frequent metabolic consequences following
administration of these therapeutic agents, and sodium-restricted diets are
required, especially for long periods of treatment More recent medical text
books continue to point out the usefulness of sodium-restricted diets for
these several diseases where fluid retention is a problem (6).
When disease causes fluid retention in the body, with subsequent edema
and ascites, there is a diminished urinary excretion of sodium and of water.
If the sodium intake is restricted in these circumstances, further fluid re-
tention will usually not occur, and the excess water ingested will be excreted
120
-------
APPENDIX A— DRINKING WATER REGULATIONS
in the urine because the mechanisms that maintain the concentration of
sodium in the extracellular fluid do not permit the retention of water with-
out sodium.
Almost all foods contain some sodium, and it is difficult to provide a
nutritionally adequate diet without an intake of about 440 mg of sodium
per day from food; this intake would be from the naturally occurring
sodium in food with no salt added. The additional 60 mg that would in-
crease the intake to the widely used restricted diet of 500 mg per day must
account for all non-nutrition intake that occurs from drugs, water and inci-
dental intakes. A concentration of sodium in drinking water up to 20 mg
per liter is considered compatible with this diet. When the sodium content
exceeds 20 mg/1, the physician must take this into account to modify the
diet or prescribe that distilled water be used. Water utilities that distribute
water that exceeds 20 mg/1 must inform physicians of the sodium content of
the water so that the health of consumers can be protected. About 40 percent
of the water supplies are known to exceed 20 mg/1 and would be required
to keep physicians informed of the sodium concentration (7). Most of the
State health departments have made provision for determining the sodium
content of drinking water on a routine basis and are now informing physi-
cians in their jurisdiction (8). If change of source or a treatment change
such as softening occurs that will significantly increase the sodium con-
centration, the utility must be sure that all physicians that care for con-
sumers are aware of the impending change. Diets prescribing intakes of less
than 500 mg per day must use special foods such as milk with the sodium
reduced, or fruits that are naturally low in sodium.
It is not known how many persons are on sodium-restricted diets and to
what extent the sodium intake is restricted. To reduce edema or swelling,
the physician may prescribe a diuretic drug, a sodium-restricted diet, or a
combination of the two. Therapy, of course, depends on the patient's condi-
tion, but there are also regional differences that probably result from physi-
cian training. The American Heart Association (AHA) (9) feels that di-
uretics may allow for less need of very restricted diets and that di-
urectics are necessary for quick results in acute conditions. For long-term,
use, a sodium-restricted diet is simpler, safer, and more economical for the
patient. It is preferable, especially when a moderate or mild sodium-re-
stricted diet will effectively control the patient's hypertension and water re-
tention. Literature is provided to physicians by the AHA to distribute to
their patients explaining the sodium-restricted diets. These cover the "strict"
restriction • 500 mg. sodium, "moderate" restriction • 1000 mg sodium, and
the "mild" restricted diet-2400 to 4500 mg sodium. From 1958 through June
1971, there were 2,365,000 pieces of this literature distributed: 37% -500
mg; 34% ¦ 1000 mg; and 29% • "mild" (10). There are many ways a
physician can counsel his patients other than using this literature, so the
total distribution does not reflect the extent of the problem, but the pro-
121
-------
DRINKING WATEH REGULATIONS
portion of booklets distributed may provide an estimate of the portion of
diets that are prescribed. The "mild" restricted diet could require just cut-
ting down on the use of salt, and literature for the patient would not be as
necessary.
The AHA estimates that hypertension affects more the 2 million Amer-
icans, and in more than half of these cases put enough strain on the heart
to be responsible for the development of hypertensive heart disease (11).
Congestive heart failure is a sequelae of several forms of disease that damage
the heart and would affect some unknown portion of the 27 million persons
with cardiovascular disease. Thus, from 21 to 27 million Americans would
be concerned with sodium intake.
Toxemias of pregnancy are common complications of gestation and occur
in 6 to 7 percent of all pregnancies in the last trimester (12). Thus, about
230,000 women would be very concerned with sodium intake each year.
Other diseases are treated with restricted sodium intake, but no estimate can
be made on the number of people involved.
Questions about salt usage were asked on the ninth biennial examination
of the National Heart Institute's Framingham, Massachusetts Study (13).
The study population was free of coronary heart disease when the study
began in 1949 and now are over 45 years of age. There were 3,333
respondents. Forty-five percent, of the males and 30 percent of the females
reported that they add salt routinely to their food before tasting. But at the
other extreme, 9 percent of the men and 14 percent of the women avoid
salt intake. More of the people 60 and over avoid salt intake than the 45 to
59 population. It is not determined if the salt restriction was medically pre-
scribed nor how extensively the sodium intake was restricted.
It can be seen that a significant proportion of the population needs to
and is trying to curtail its sodium intake. The sodium content of drinking
water should not be significantly increased for frivolous reasons. This is
particularly true of locations where many of the people using the water
would be susceptible to adverse health effects, such as hospitals, nursing
homes, and retirement communities. The use of sodium hypochlorite for
disinfection, or sodium fluoride for control of tooth decay, would increase
the sodium content of drinking water but to an insignificant amount. The
use of sodium compounds for corrosion control might cause a significant
increase, and softening by either the base exchange or lime-soda ash process
would significantly increase the sodium contentof drinking water. For each
milligram per liter of hardness removed as calcium carbonate by the ex-
change process, the sodium content would be increased about one-half mg
per liter. The increase in excess lime softening would depend on the amount
of soda ash added. A study in North Carolina found that the sodium con-
tent of 30 private well-water supplies increased from 110 mg/1 to 269 mg/1
sodium on the average after softening (14). The sodium content of the
softened water was much higher shortly after the softener had been regen-
122
-------
APPENDIX A—DRINKING WATER REGULATIONS
erated than later in the cycle. A case has been reported where a replacement
element type softener was not flushed, and the drinking water had a sodium
content of 3,700 mg/1 when the unit was put back in service.
As a further deterrent to softening of water, it should be noted that there
is considerable evidence of an inverse relationship between water hardness
and certain cardiovascular diseases. Research in the area is being accele-
rated to determine cause and effect relationships. Until the full significance
of water hardness is known, and because of the increase in sodium content
of softened waters, utilities should carefully consider the consequences of
installing softening treatment.
All consumers could use the water for drinking if the sodium content was
kept below 20 mg per liter, but about 40 percent of the U.S. water supplies
have a natural or added sodium content above this concentration (7). Many
industrial wastes and runoff from deiced highways may increase the sodium
pollution of surface water (15). The problem is most acute when ground
water is polluted with sodium (16, 17) because it remains for a long time.
Removal of sodium from water requires processes being developed by the
Office of Saline Water (13) and are economically feasible only in certain
situations.
The person who is required to maintain a restricted sodium intake below
500 mg per day can use a water supply that contains 20 mg or less sodium
per liter. If the water supply contains more sodium, low sodium bottled
water or specially treated water will have to be used. In the moderately re-
stricted diet that allows for a consumption of 1000 mg sodium per day the
food intake is essentially the same, but the diet is liberalized to allow the
use of 1/4 teaspoon of salt, some regular bakery bread, and/or some salted
butter. If persons on the moderately restricted diet found it necessary to use
a water with a significant sodium content they could still maintain their
limited sodium intake with a water containing 270 mg/liter. This would re-
quire allocating all the liberalized intake to water (the original 20 mg/1 and
250 mg/1 more with two liter domestic use, drinking or cooking, per day).
High sodium in water causes some transfer of sodium to foods cooked in
such water (5).
It is essential that the sodium content of public water supplies be known
and this information be disseminated to physicians who have patients in the
service area. Thus, diets for those who must restrict their sodium intake can
be designed to allow for the sodium intake from the public water supply or
the persons can be advised to use other sources of drinking water. Special
efforts of public notification must be made for supplies that have very high
sodium content so that persons on the more restricted sodium intakes will
not be overly stressed if they occassionally use these water supplies.
The 1963 Sodium Survey (7) had the following percent distribution of
sodium concentration from 2100 public water supplies:
123
-------
DRINKING WATER REGULATIONS
Range of Sodium Ion
Concentration
Percent of Total
Samples
0- 19.9
20- 49.9
50 - 99.9
100 - 249.9
250 . 399.9
400 - 499.9
500 - 999.9
Over 1000
mg/1
%
58.2
19.0
9Z
8.7
3.6
0.5
0.7
0.1
While the question of a maximum contaminant level for sodium is still
under consideration by the National Academy of Sciences and others, no
specific level will be proposed for the Interim Primary Drinking Water
Regulations. The Environmental Protection Agency believes that the avail*
able data do not support any particular level for sodium in drinking water,
and that the regulation of sodium by a maximum contaminant level is a
relatively inflexible, very expensive means of dealing with a problem which
varies greatly from person to person.
1. Dahl, LJC. Possible Role of Salt Intake in The Development of Essential Hyper-
tension, From Essential Hypertension: An International Symposium. P. Cottier and
K.D. Bock, Berne (Eds.) Springer Verlag, Neidelberg pp. 53-65 (1960).
2. Bureau of Radiological Health. California State Department of Public Health, Esti-
mated Daily Intake of Radionuclides in California Diets, April-December 1969, and
January-June 1970. California State Department of Health, Radiological Health
Data, and Reports, 6250632, November 1970 (1970).
3. Elliott, G.B., and Alexander, E.A. Sodium from Drinking Water as An Unsuspected
Cause of Cardiac Decompensation. Circulation 23: 562 (1961).
4. Fin berg, L., Kiley, J., and Luttrei, C.N. Mass Accidental Salt Poisoning in Infancy.
Med. Assn. 184: 187 (1963). 187-190 (April 20,1963).
5. Food and Nutrition Board-NAS-NRC, Sodium-Restricted Diets, Publication 325,
National Research Council, Washington, D.C. (1954).
6. Wintrobe, M.M., Thorn, G.W., Adams, R.D., Bennett, I.L., Brauwald, E„ Isselbacher,
KJn and Petersdorf, R.G., (Eds.) Harrison's Principles of Internal Medicine,
(6th ed.) McGraw-Hill Book Co., New York. (1970).
7. White, J.M., Wingo, J.G., Alligood, L.M., Cooper, G.R., Gutridge, J„ Hydaker, W.,
Benack, R.T., Dening, J.W. and Taylor, F.B. Sodium Ion in Drinking Water I.
Properties, Analysis, and Occurence, Dietetic Asin., 50 : 32 (1967).
8. Review of State Sodium-in-Drinking-Watcr Activities. Bureau of Water Hygiene,
U.S. Public Health, Service, Washington, D.C. (1971).
9. Pollack, H. Note to The Physician (inserted with diet booklets) Your 500 mg.
Sodium Diet-Strict Sodium Restriction, Your 1000 mg. Sodium Diet • Moderate
Sodium Restriction, and Your Mild Sodium-Restricted Diet, American Heart
Association (1960).
10. Cook, LP. American Heart Aaan. Personal Communication (1971).
11. American Heart Assn. Heart Facts 1972. A.H.A., New York (1971).
12. Eastman, NJ. and Hellman, L.M. Williams Obstretics. (13th ed.) Appleton-Century.
Crofts, New York (1966).
13. Kannel, W.B. Personal Communication (1971).
REFERENCES
124
-------
APPENDIX A—DRINKING WATER REGULATIONS
14. Garrison, C.E., and Ader, O.L. Sodium in Drinking Water. Arch. Environ. Health,
13: 551 (1966).
15. Bubeck, R.C., Diment, W.H., Deck B.L., Baldwin, A.L, and Lipton, S.D. Runoff of
Deicing Salt: Effect on Irondequoit Bay, Rochester, New York. Science 172: 1128
(1971).
16. Joyer, B.F., and Sutcliffe, H. Jr. Sait-Water Contamination in Wells in the Sara-
Sands Area of Siesta Key, Saratoga County, Florida. JAWWA. 59: 1504 (1967).
17. Parka, W.W. Decontamination of Ground Water at Indian Hill. JAWWA. 5J: 644
(1959).
18. U.S. Department of the Interior. Saline Water Conversion Report for 1969-1970.
Government Printing Office, Wasiagton, D.C. (1970).
125
-------
DRINKING WATER REGULATIONS
SULFATE
The presence of sulfate ion in drinking water can result in a cathartic ef-
fect. Both sodium sulfate and magnesium sulfate are well-known laxatives.
The laxative dose for both Glauber salt ( NaoSO+'lOHoO) and Epsom salt
(MgS04*7H20) is about two grams. Two liters of water with about 300 mg/1
of sulfate derived from Glauber salt, or 390 mg/1 of sulfate from Epsom salt,
would provide this dose. Calcium sulfate is much less active in this respect.
This laxative effect is commonly noted by newcomers and casual users of
waters high in sulfates. One evidently becomes acclimated to use of these
waters in a relatively short time.
The North Dakota State Department of Health has collected information
on the laxative effects of water as related to mineral quality. This has been
obtained by having individuals submitting water samples for mineral an-
alyses complete a questionnaire that asks about the taste and odor of the
water, its laxative effect (particularly on those not accustomed to using it),
its effect on coffee, and its effect on potatoes cooked in it.
Peterson (1) and Moore (2) have analyzed part of the data collected,
particularly with regard to the laxative effect of the water.
Peterson found that, in general, the waters containing more than 750
mg/1 of sulfate showed a laxative effect and those with less than 600 mg/1
generally did not. If the water was high in magnesium, the effect was shown
at lower sulfate concenrations than if other cations were dominant. Moore
showed that laxative effects were experienced by the most sensitive persons,
not accustomed to the water, when magnesium was about 200 mg/1 and by
the average person when magnesium was 500-1,000 mg/1. Moore analyzed
the data as shown in Table 1. When sulfates plus magnesium exceed 1,000
mg/1, a majority of those who gave a definite reply indicated a laxative
effect.
Table 2 presents some data collected by Lockhart, Tucker and Merritt (3)
and Whipple (4) on the influence of sulfate on the taste of water and coffee.
Because of the milder taste of sulfate over chloride (5) (6) a taste standard
for sulfate would probably be in the 300-400 mg/1 range. The Peterson data
(1) and Table 1 (2), however, indicate that from 600 to 1000 mg/1 of sul-
fate has a laxative effect on a majority of users.
While a limit for sulfate may be included in Secondary Drinking Water
Regulations, on the basis of the effect of sulfate on water taste, no maximum
contaminant level is being proposed at this time. As noted above, a relatively
high concentration of sulfate in drinking water has little or no known effect
on regular users of the water, but transients using high sulfate water some*
times experience a laxative effect Whether this effect will occur, and its
severity, varies greatly with such factors as the level of sulfate in the water
being consumed and the level of sulfate to which the transient is accustomed.
Because of this great variability, the available data do not support
126
-------
APPENDIX A—DRINKING WATER REGULATIONS
Table 1
Effects
Percent
Determination
Range mg/1 Number
Present
of Yes
of Wells
Laxative
Not
Answers
in Range
Yea
No
Stated
*
Magnesium plus sulfate
0-200
51
9
34
8
21
200-500
45
7
27
11
21
500-1,000
56
11
38
17
28
1,000-1,500
36
18
10
8
64
1,500-2,000
14
6
4
4
60
2,000-3,000
2i
13
3
5
81
Over 3,000
14
5
1
8
83
Sulfate
0-200
56
10
36
10
22
200-500
47
9
28
10
24
500-1,000
56
13
26
17
33
1,000-1,500
34
16
10
8
62
1,500-2,000
16
9
4
3
69
2,000-3,000
20
9
3
8
75
Over 3,000
8
3
0
5
100
*This percentage is based only on the total of yes and no answers. It is probable that
a large proportion of the wells for which no statements were made were not regularly
used as water supplies.
Table 2. Data on the Influence of Sulfate Salts on the Taste of Water and Coffee
Threshold Concentration — rag/1
Median Range
Salt Salt Anion Salt Anion
Na»SO*
350
327
250-550
169-372 (4)
CaSO*
525
370
250-900
177-635 (4)
MgSO«
525
419
400-600
320-479 (4)
Average
MgSO«
500
400 (3)
the establishment of any given maximum contaminant level. The Environ-
mental Protection Agency recommends that the States institute monitoring
programs for sulfates, and that the transients be notified if the sulfate con-
tent of the water is high. Such notification should include an assessment of
the possible physiological effects of consumption of the water.
In the meantime, research is being undertaken to determine if the health
effects of sulfate in drinking water warrant further consideration. If data
we generated to support a maximum contaminant level, this level will be
proposed for inclusion in Revised Interim Primary Water Regulations.
REFERENCES
1- Peterson* N.L. Sulfates in Drinking Water. Official Bulletin North Dakota Water
and Sewage Works Conference. 18: (1951).
2. Moore, E,W. Physiological Effects of The Consumption of Saline Drinking Water.
Bulletin of Subcommittee on Water Supply, National Research Council, Jan. 10,
1952, Appendix B, pp. 223-227 (1952).
127
-------
DRINKING WATER REGULATIONS
3. Loclthart, E.E., Tucker, C.L., and Merritt, M.C. The Effect of Water Impurities
on The Flavor of Brewed Coffee. Food 20: 598 (1955).
4. Whipple, G.C., The Value of Pure Water. John Wiley, New York (1907).
5. Bruvold, W.H., and Gaffey, W.R., Evaluation Rating of Mineral Taste in Water, J.
Perceptual Motor Skills 28: 179 (1969).
6. Bruvold, W.H., and Gaffey, W.R., Rated Acceptability of Mineral Taste in Water.
II Combinatorial Effects of Ions on Quality and Action Tendency Ratings. J. Ap-
plied Psychol. 53 : 317 (1969).
128
-------
APPENDIX B—RADIONUCLIDES
Appendix B-Radionuclides
Introduction
The Safe Drinking Water Act directs the Administrator to set interim
primary standards for drinking water that "shall protect health to the extent
feasible, using technology, treatment techniques and other means, which the
Administrator determines are generally available (taking costs into con-
sideration)." The cost considerations referred to are limited to treatment
techniques and other means which are under the control of the water sup-
plier. The Agency believes that the establishment of maximum contaminant
levels for radioactivity (1) will protect health to the extent feasible and aid
achievement of the national goal of safe drinking water.
General Considerations
In determining maximum contaminant levels for radioactivity in drink-
ing water the Agency has given consideration to several important factors
including the diversity of sources causing radioactivity to be present in
drinking water. Radioactivity in public water systems may be broadly cat-
egorized as naturally occurring or man-made. Radium-226 is the most im-
portant of the naturally occurring radionuclides likely to occur in public
water systems. Although radium may occassionally be found in surface
water due to man's activities, it is usually found in ground water where
it is the result of geological conditions, not subject to prior control. In con*
trast to radium, man-made radioactivity is ubiquitous in surface water be-
cause of fallout radioactivity from nuclear weapons testing. In some
localities this radioactivity is increased by small releases from nuclear
facilities (such as nuclear power plants), hospitals, and scientific and in-
dustrial users of radioactive materials. The Agency recognizes that, for both
man-made and naturally occurring radioactivity* a wide range of both
controllable and uncontrollable sources can influence the concentration of
radioactivity in water served by public systems.
Variability in the quality of source waters is not unique for radioactive
contaminants; other containments in drinking water also differ widely in
their occurrence. Limits to protect public health can not be based on some
proven harmless intake of radioactive material. Rather, maximum contam-
inant levels for radioactivity are based on the assumption that there iis no
harmless level of dose from ionizing radiation and that any detrimental
effects on health produced by the radiation will be proportional to the dose
equivalent delivered by the radioactivity in drinking water.
The Agency recognizes that for the low doses and dose rates expected
from intakes of drinking water, the risk to an individual is small and that
the potential health effects associated with the risk are no different in the
types of diseases manifested spontaneously, representing in fact only small
potential increases in the normal incidences in these diseases. The Agency
also recognizes that the number of health effects caused by ionizing radia-
129
-------
DRINKING WATER REGULATIONS
tion at very low doses and dose rates is presently unknown and unlikely to
be quantified more precisely in the immediate future. Therefore, the En-
vironmental Protection Agency has adopted a prudent policy which assumes
that any dose of ionizing radiation may produce potential harmful effects
to human health and that the extent of such harm can be estimated from
effects that have been observed at higher doses and dose rates than are
likely to be encountered from environmental sources of radiation. Accept-
ance of this policy by the Agency cannot be based solely on the scientific
evidence but must include an operational judgment, for practical reasons,
in applying present knowledge to the establishment of standards. A more
detailed statement of this policy on the relationship between radiation dose
and effects is reprinted in Appendix I.
Depending on the circumstances of the exposure, risks from ionizing
radiation may or may not be accompanied by an offsetting benefit. In the
case of radium contaminated ground water there is no benefit, per se, from
the geological processes causing the radiocontamination. On the other hand,
man-made radioactivity in public water supply systems may be deliberate
due to man's use of nuclear energy to produce electric power, or to his use
of radionuclides in the diagnosis and treatment of diseases or research and
industrial applications. Balancing the risks and benefits from these activities
and specifying appropriate controls for the resultant liquid effluent waste
streams is required by other Federal statutes. The Administrator is limited
under the Safe Drinking Water Act to regulating the water supplier. How-
ever, the Interim Regulations for radioactivity take full account of. the fact
that control regulations established under authority of the Atomic Energy
Act as amended (PL 83-703) and Environmental Protection Standards pro-
mulgated under this Act by EPA (Radiation Protection for Nuclear Power
Reactors, 42FR2859, January 13, 1977 as well as Federal Radiation Council
Guides are intended to limit liquid radioactive discharges into surface
waters to the extent practicable.
In addition to man-made radioactivity in drinking water due to effluents
from nuclear facilities, surface waters may contain radioactive materials
from aerial effluent releases and from nuclear weapons testing. The residual
radioactivity in surface waters from fallout due to atmospheric nuclear
weapons testing is mainly strontium-90 and tritium, the former being the
more important in health considerations. Current data on the impact of
fallout strontium-90 on public water supplies is incomplete. However, the
available data indicate strontium-90 concentrations are about 1 pCi per
liter, corresponding to a dose equivalent to bone marrow of less than 0.5
mrem annually.* Tritium concentrations in surface water rarely exceed
1000 pCi per liter, corresponding to a dose equivalent of less than 0.2 mil-
lirem per year.
'Definition* of unit* and term* are given in the regulation*; dosimetry calculations in
Appendix IV.
130
-------
APPENDIX B—RADIONUCLIDES
Unplanned releases of radioactive materials are another source of pos-
sible contamination. It is not anticipated that the proposed maximum con-
taminant levels for radioactivity would apply to transient situations such
as might follow a major contaminating event. In accident situations it is
necessary to balance, on a case-by-case basis, the potential risk from radia-
tion exposure against the practicality and consequences of any remedial
measures taken to ameliorate that risk. In such situations Federal guidance
as promulgated in the Federal Register Notices of August 22, 1964 and May
22, 1965 will apply and the emergency plans of the States, as provided for
in Section 1413(A) (5) of the Safe Drinking Water Act should reflect this
Federal Guidance.
Radium in drinking water is primarily a problem of the smaller public
water systems. About 40 percent of the U.S. population is served by 243
regional systems supplying large metropolitan areas. Yet, most of the na-
tion's 40,000 community water systems serve less than 500 persons. In
general, the large regional systems utilize surface water which on the whole
contains very low concentrations of radium. Small supplies commonly use
ground water, water which in some cases may contain radium. Therefore,
the impact of maximum contaminent levels for radium is more likely to
fall on some small supply systems which generally have limited resources.
Although one of the intentions of the Safe Drinking Water Act is to en-
courage the regionalization of these small systems, the availability of local
resources for the control and monitoring of radioactivity has been of con-
cern to the Agency. This concern is balanced by the belief that the identi-
fication of an atypical radium concentration and the introduction of its
control is a direct benefit to the user population. This benefit is a reduction
in any health risks due to radium in drinking water.
Health Risks From Radionuclides in Drinking Water
Risk estimates from total body and to a lesser extent, partial body ex-
posure have been made using data published in the NAS-BEIR Report
(National Academy of Sciences Report of the Advisory Committee on the
Biological Effects of Ionizing Radiation) (2). Such estimates are based on
the likely conservative, but nevertheless prudent assumption that the radia-
tion effects are linearly proportional to the dose* and that the number of
cancers per rem that have been observed at high doses and dose rates is a
practical predictor of the effects per rem at the low doses and dose rates
encountered from environmental sources of radiation. The degree of con-
servatism in such an approach has not been documented but it is likely to
be less for ingested alpha particle emitting radionuclides than for those
man-made sources of radioactivity which decay by beta and gamma ray
emission.
'For the purpose of this statement "dose" means "dose equivalent" as defined in the
regulations.
131
-------
DRINKING WATER REGULATIONS
The NAS-BEIR risk estimates are for the U.S. population in the year
1967. For an exposed group having the same age distribution, the individual
risk of a fatal cancer from a lifetime total body dose rate of 4 mrem per
year ranges from about 0.4 to 2 x 10-6 per year depending on whether an
absolute or relative risk model is used. *The NAS-BEIR Committee does not
choose between these two models but their "most likely estimates" corres-
pond to an average of the absolute and relative risk estimate i.e., about twice
the absolute risk. For fatal cancer, an individual risk of 0.8 x 1O-0 per year
from a 4 mrem annual total body dose is believed to be a reasonable es-
timate of the annual risk from the lifetime ingestion of drinking water at
the maximum contaminant level for man-made beta and photon emitting
radioactivity. The risk from the ingestion of water containing lesser
amounts of radioactivity would be proportionately smaller.
The estimated total health risk from radiation exceeds that due to fatal
cancers alone. The NAS-BEIR committee projected that the incidence of
non-fatal cancers would be about the same as fatal cancers. The incidence
of genetic effects is more difficult to estimate; but the increase, expressed
over several generations, would be comparable to the increased incidence
of fatal cancer (2).
The estimated risks of a fatal cancer due to a lifetime exposure of ioniz-
ing radiation can be compared to the risk without additional radiation by
normalizing the NAS-BEIR data for the 1967 population in terms of a
single individual's exposure history. Based on U.S. Vital Statistics, (3) the
probability that an individual will die of cancer is about 0.19. This prob-
ability may be increased by 0.1% from a lifetime dose equivalent rate of
15 mrem per year. Maximum contaminant levels for man-made beta and
photon emitters limit the dose equivalent from the drinking water pathway
to 4 mrem per year, corresponding to a lifetime risk increase of 0.025% to
exposed groups.
For partial body irradiation, which is not uncommon for ingested radio-
nuclides since the* radioactivity may be largely concentrated in a particular
organ or group of organs, the estimated risk is somewhat less than for total
body exposure where all organs are irradiated. For example, the estimated
thyroid cancer incidence rate from the thyroid gland receiving 10 mrem
per year continuously ranges from about 0.5 to 1.3 per year per million ex-
posed persons (averaged over all age groups). Fatality due to thyroid
cancers varies with age, from nearly zero for children and young adults to
about 20 percent of the incidence for persons well past middle age. Although
it is noted that estimated fatalities from thyroid exposure are at least five
times less than that from whole body exposure, other factors bearing on
the health impact are significant.
• Absolute risk estimates are based on the reported number of cancer deaths per rad;
relative risk estimates, on the percentage increase in cancer mortality per rad.
132
-------
APPENDIX B—RADIONUCLIDES
The incidence in thyroid tissue of non-cancerous neoplasms, (benign
nodules), following radiation exposures is much higher than the incidence
of thyroid cancers, particularly in the young (2). Since the most likely
treatment for such nodules is severe, thyroidectomy, the medical conse-
quences are underestimated by a consideration of cancers only. In addition,
there is clinical evidence that the young appear to be particularly suscep-
tible to radiation induced cancer of the thyroid, perhaps by as much as a
factor of 10 (2, 3). While it is appropriate to calculate risks due to the
dose permitted by an ambient standard on the basis of the average risk
throughout life and not just childhood alone, as in the Interim Regulations,
the Agency recognizes a need for some conservatism where the major im-
pact of the allowed radiation may fall on a particular subgroup.
Radium locates primarily in bone where 80 to 85 percent of the retained
radium is deposited. However, other organs are also irradiated to a lesser
extent and the total health risk from radium ingestion has been estimated
by summing the dose and resultant risk from all organs, Appendix II. Risk
estimates derived from the BEIR Report (2) indicate that continuous con-
sumption of drinking water containing radiunv-226 or radium-228 at the
maximum contaminant level of 5 pCi/1 may cause between 0.7 and 3 can-
cers per year per million exposed persons. Almost all of these cancers would
probably be fatal. Although the maximum contaminant level for radium
is much nearer Federal Radiation Council guides than the limit for man-
made radioactivity, see below, the estimated risks from maximum contamin-
ant levels for radium and for man-made radioactivity are nearly the same.
It should be noted that these risk estimates apply only to the relatively small
proportion of the population exposed to radioactivity at the maximum con-
taminant level.
While it is incorrect to speak of safety factors in radiation standards,
since only in the complete absence of radiation can any effects be avoided
completely, some perspective may be gained by comparing the dose due
to drinking water at maximum contaminant levels to dose levels established
for population groups by the Federal Radiation Council (4). The radiation
protection guide for all sources of total body exposure except radiation
received for medical purposes and that due to natural background is 170
millirem per year. At this dose rate effects are not expected to be necessarily
non-existent but rather non-detectable, except perhaps by rigorous statistical
analysis involving a large exposed population. The annual dose allowed
by the proposed maximum contaminant levels for man-made radionuclides
is over forty times smaller (4 millirem vis-a-vis 170 millirem) for a single
exposure pathway, drinking water. Similarly, in the case of radium-226,
Federal Guides for total ingestion recommend that the daily intake not ex-
ceed 20 pCi,* twice that allowed by the maximum contaminant level, 5
pCi/1 and an intake of 2 liters per day.
* Upper limit of Range II (5).
133
-------
DRINKING WATER REGULATIONS
In addition to the maximum contaminant level for radium-226 and
radium-228 of 5 pCi/1, the Interim Regulations specify a maximum con-
dium-226.* A limit is placed on gross alpha particle activity rather than
dium-226.## A limit is placed on gross alpha particle activity rather than
each alpha particle emitting radionuclide individually since it is imprac-
tical at the present time to require identification of ail alpha particle emiting
radionuclides because of analytical costs.
The maximum contaminant level for gross alpha particle activity is based
on a consideration of the radiotoxicity of other alpha particle emitting con-
taminants relative to radium. The 15 pCi per liter gross alpha particle
limit (which includes radium-226) is based on the conservative assumption
that if the radium concentration is 5 pCi/I and the balance of the alpha
particle activity is due to the next most radiotoxic alpha particle emitting
chain, starting with lead-210, the total dose to bone would be equivalent to
less than 6 pCi/l of radium-226 (6).
As stated in Section 141.15(b) in the Interim Regulations, the maximum
contaminant level for gross alpha particle activity does not include any
uranium or radon that may be present in the sample. The Agency may con-
sider proposing maximum contaminant levels for these radionuclides at a
later date after determining the national need for such regulations, the cost
of water treatment to remove these radionuclides and their dosimetry and
potential for causing adverse health effects. It should be noted that the
maximum contaminant level for gross alpha particle activity includes man-
made as well as naturally occurring radioactive materials, Section
141.2 (m).
The Control of Radium in Public Water Systems
In contrast to man-made radioactivity, for which the environmental
impact it controlled by a number of regulatory agencies, the abatement of
radium radioactivity in drinking water has received little attention. There-
fore, radium contamination of drinking water is often of more concern from
a regulatory standpoint than that due to man-made radioactivity. Radium-
226 is distributed widely in the U.S., and is found frequently in ground
water, particularly in the midwestern and Rocky Mountain States. (In a
comparatively few cases radium-228, a beta emitter having a chain of
daughter radionuclides which decay by alpha particle emission, like radium-
226, is also present.) Unlike the situation for ground water, radium is in-
frequently found in any appreciable quantity in U.S. surface waters. In most
of the public supply systems utilizing surface water the radium content is
extremely low, less than 0.1 pCi per liter. In contrast to surface waters the
concentration of radium in ground waters used by public supply systems
can be appreciable, concentrations as large as 50 pCi per liter have been
* Radium-228 is a bets particle emitter.
134
-------
APPENDIX B—RADIONUCLIDES
reported and perhaps as many as 500 community water systems supply
water that exceeds 5 pCi per liters
Several remedial measures are applicable to radium control. In some
instances it should be possible to utilize surface or other ground water
sources containing less radium. Dilution with less radioactive waters is an
acceptable abatement technique for complying with the interim regulations.
Depending on the quality of the source water, such common water treat-
ment practice as coagulation may remove about 25% of the radium (7).
However, in some cases more rigorous treatments will be required to meet
the maximum contaminant level for radium-226 and radium-228. Radium
removal by means of conventional technology is feasible. A number of
public water systems currently remove radium as part of their water soften-
ing treatment processing. The most efficient and in many cases the most
economical treatment method for radium removal is based on the use of
zeolite as an ion exchange medium. In this process calcium and radium
are exchanged for sodium. The Agency is aware that if the mineral content
of the source water is high, the exchange of calcium with sodium could
result in a marked increase in the sodium content of the drinking water.
However, ingestion of sodium via drinking water in such cases would still
be lower than the normal dietary intake level. Even so, persons on low so-
dium diets should be informed of any significant changes in sodium con-
centration.
National Cost For Radium Removal
In order to estimate the total national cost to remove radium from all
public water systems it is necessary to know both the local concentration
of radium and the population served by each system. Such complete infor-
mation is not available since the majority of U.S. systems have not been
analyzed for radium. However, many systems have been radioassayed, par-
ticularly in the Midwest where radium contamination is encountered most
often. The estimated costs of radium removal, given below, are based on
a sample of public water systems identified by Straub in his search of the
relevant literature on radium contamination (8). Straub listed 306 com-
munity water systems serving radium-226 at a concentration of 0.5 pCi/l or
more. While his list is probably representative of the population size of
systems serving water at various radium concentrations, it is not of course
complete and contains some bias since radium assay has been extensive
only in areas known to have a potential for higher radium levels. A second
source of bias is that larger water systems are more likely to be selected
for study by public health authorities than small community systems serving
only a few persons. At best the sample of 306 systems represents a minimum
estimate of the total number of impacted systems. However, in view of the
extent of national monitoring that has occurred in recent years, it is doubt-
ful that the sample is low by an order of magnitude. For the purpose of this
135
-------
DRINKING WATER REGULATIONS
analysis, EPA has estimated Straub's sample represents about 30% of the
systems in the U.S. having radium concentrations greater than 0.5 pCi/1.
This may underestimate the number of supplies but probably overestimates
the population impacted because of the likely bias in the sample, as outlined
above. Since costs for radium removal are directly related to population,
the estimate of national costs developed below may be somewhat high.
The cost of achieving various control levels and the estimated health
benefits are shown in Table 1. It is seen that the total national cost for
radium removal increases rapidly with decreasing concentrations of radium
not only because of the increased marginal cost for treatment at low con-
centrations (Appendix III) but also because both the number of supply
systems impacted and the average population served becomes larger. The
Administrator believes that because of the limited data on the cost of
radium removal and the extent of radium contamination in community
water supplies currently available it would be unwise to prescribe radium
removal at concentrations lower than 5 pCi per liter. It should be noted,
however, that under the Safe Drinking Water Act of 1974 (PL 93-523),
States may set more stringent standards if they so desire.
Table 1. Annual National Coat and Health Savings
for Achieving Radium Control Limits
Estimated
Average
Average National Cost
Estimated Total
Control
Number of
Size o!
Cost Per to
Achieve
Number of Lives
Limit
Systems
Systems
Systems
Limit
Saved per yr.
pCi/1
#
Population
Thousands
Millions
#
dollars/yr.
dollars/yr.
9
240
4,200
6.0
1.4
0.6
8
300
5,400
8.0
2.4
1.1
7
370
5,000
92
3.4
1.6
6
450
7,450
12.4
5.6
2.5
*5
500
8,800
17.5
ZA
3.7
4
670
9,500
2U
14.
5.5
3
800
12.000
30.4
24.
8.2
2
860
12,100
41.6
36.
11
1
960
18.400
70.2
70.
15
0.5
1100
20,800
90.2
100.
20
Includes systems currently exceeding 10 pCi/1.
•Interim maximum contaminant level for radium.
At the maximum contaminant level selected it is estimated that as many
aa 500 community water systems may need to remove radium or utilize ad-
ditional source waters containing a lower radium concentration. If ion ex-
change were the method selected to lower radium concentrations the average
cost per supply would be $18,000 per year or about two dollars per person
served. The estimated cost effectiveness of radium removal to avoid a po-
tential fatal cancer is not high, mainly because only about one-half percent
of the treated water is consumed as drinking water. In some cases it may be
136
-------
APPENDIX B—RADIONUCLIDES
possible to minimize costs by not treating water used only for commercial
purposes.
The methodology used to estimate the marginal cost of ion exchange to
remove radium and the cost-effectiveness of radium removal to prevent
health effects is outlined in Appendix III. It must be understood that other
abatement measures such as dilution will have lower costs than those pre-
dicted in Appendix III and that the effects of radium removal in terms of
reducing the predicted excess cancer incidence is uncertain by at least a
factor of four. Therefore, the estimated cost effectiveness of radium removal
should not be given undue weight in evaluating the proposed maximum con-
taminant levels. However, the cost estimates are not affected by the uncer-
tainty in health effect models and have been used by EPA to project the
national cost of various control limits considered by the Agency in its se-
lection of a maximum contaminant level for radium.
Impact of Maximum, Contaminant Levels for Man-made Radionuclides
Though man-made radioactivity in public water systems is sometimes a
matter of concern it is important to recognize that unlike the case for ra-
dium, current ambient concentrations are less than the proposed limits be*
cause of regulatory concern for these radionuclides. Drinking water is not
a major pathway for exposure from nuclear power plants. The Agency has
reviewed all the Envionmental Impact Statements for power reactors cur-
rently available. Based on the design of these reactors the estimated total
body doses due to drinking water served by public water systems from
these facilities range from 0.00001 to 0.3 millirem per year with 90% of the
expected doses less than 0.04 millirem per year. The average total body dose
is 0.3 millirem per year. Thyroid doses are somewhat larger, ranging from
0.0003 to 0.8 millirem per year, with an average annual dose of 0.08
millirem per year.
Data on ambient levels in public water systems indicate that almost all
of the radioactivity in the aquatic environment is due to residual radioactiv-
ity from nuclear weapons testing. The historical trend of radioactivity in
the Great Lakes and in other waterways shows this source of radioactivity
is diminishing (9).
The maximum contaminant level for man-made radionuclides is expressed
in terms of the annual dose rate (millirem per year) from continuous in-
gestion. Specifying maximum contaminant levels in terms of radioactivity
concentration (pCi per liter) was considered but rejected in view of the
short length of time such limits would be appropriate, since presently avail-
able dose conversion factors for ingested radioactivity are obsolescent and
the ICRP is developing new dose models. When appropriate models for
doses due to environmental contamination become available, the Agency
will revise the Interim Regulations to permit the use of newer data. The con-
centrations yielding 4 millirem annually, given in Appendix IV, are based on
137
-------
DRINKING WATER REGULATIONS
NBS Handbook 69 as required by the Interim Regulations, 41 FR 133, p.
28402, July 9, 1976.
Monitoring for Radioactivity in Community Water Systems
The Agency has developed monitoring requirements for radioactivity with
two ends in view. Information must be available to the supplier so he can
control the quality of the water he serves. However, the cost of the monitor-
ing should not result in an undue economic burden in terms of other finan-
cial requirements for safe operation of the system. To an extent these are
conflicting requirements since more information can always be purchased
for more money. The Agency has tried to limit the monitoring to that which
is essential for determining compliance with maximum contaminant limits
under most conditions. As State capability for effective monitoring is aug-
mented, States are encouraged to introduce more rigorous monitoring of
particular supplies because of local knowledge of their potential for radio-
contamination. In addition Federal monitoring requirements for radioactiv-
ity are limited to community water systems as defined in Sectionl41.2 of the
Interim Regulations. Since the proposed limits are based on lifetime ex-
posure, any radiation risk to transient populations is minimal.
In general, the Interim Regulations call for quarterly sampling. In the
case of naturally occurring radioactivity it is often thought that a single
sample can be used to determine the average annual concentrations. This is
not the case for some ground water sources where the annual discharge cycle
of the aquifers has a pronounced effect on radium concentration. In such
cases, a single yearly grab sample could show a low concentration, result-
ing in the acceptance of water containing more than a maximum con-
taminant level. Conversely, an abnormally high level could lead to the insti-
tution of expensive control measures where the annual average concentra-
tion is really acceptable. Although sampling at monthly intervals might be
advisable in certain locations and situations (and should be required by the
State where necessary) the Agency believes quarterly sampling will be suf-
ficient to determine the average annual concentration in most cases. Where
the average annual concentration has been shown to be less than one-half
the relevant maximum contaminant level, a yearly sampling procedure is
permitted by the regulations.
In order to reduce monitoring costs, the Interim Regulations allow com-
posited samples to be radioassayed, usually at yearly intervals. In such cases
care must be taken to prevent the loss of activity by means of absorption on
container walls. Acidification with 1 milliliter of 16N HN03 per liter of
sample is a method suggested in "Interim Radiochemical Methodology for
Drinking Water" (10). In the case of iodine-131, hydrochloric rather than
nitric acid should be used for acidification and sodium bisulfite should be
added to the sample. In some cases State laboratoroies may prefer to count
quarterly samples rather than keep track of quarterly aliquots. If so, the
estimated costs given below will be exceeded. The increased cost is not large,
138
-------
APPENDIX B—RADIONUCLIDES
however, and quarterly measurements are recommended, particularly for the
monitoring of gross beta activity.
It should be noted that from the definition of "maximum contaminant
level" in the Interim Regulations, section 141.2(c), samples should be col-
lected from free flowing outlets, not at the source of supply water. Since in
some cases, several sources may contribute water to the system, samples
should be taken at representative points within the system so as to truly re-
flect the maximum concentration of radioactivity received by users. In cases
where more than one source is utilized, suppliers shall monitor source water,
in addition to water from a free flowing tap, when ordered by the State.
Although monitoring a typical community water system is relatively inex-
pensive, less than five dollars per year, the total national cost of monitoring
for radium-226, radium-228, and gross alpha particle activity is not trivial
because of the large number of supplies involved, 40,000. In order to min-
imize cost, the Agency is proposing that a water supplier initially obtain a
relatively low cost analysis of gross alpha particle activity. In most cases
this test will indicate that no significant activity is present and additonal
tests will not be required. However, when the gross alpha measurement indi-
cates the alpha particle activity may exceed 5 pCi per liter, a further test for
radium-226 is required.
Although not in the same decay chain, radium-228 sometimes accom-
panies radium-226. Only rarely, however, does the radium-228 concentra-
tion exceed that of radium-226. Therefore, a radium-228 analysis, which is
relatively expensive, is only required when the radium-226 concentration ex-
ceeds 3 pCi per liter. In localities where radium-228 may be present in
drinking water, it is recommended that the State require radium-226 and/or
radium-228 analyses when the gross alpha particle activity exceeds 2 pCi/1.
The Interim Regulations require sampling and measurement at quarterly
intervals where the limits are exceeded so that the water supplier can follow
the variation of radium concentration through the yearly cycle and thereby
institute remedial measures, such as additional dilution or treatment, during
periods when concentrations are unusually high. Monitoring at quarterly in-
tervals shall be continued until the annual average concentration no longer
exceeds the maximum contaminant level or until a monitoring schedule as
a condition to a variance, exemption or enforcement action shall become
effective.
Monitoring Costs for Radium and Alpha Particle Activity
Estimated monitoring costs are based on the assumption that 40,000 com-
munity water systems will initially monitor for gross alpha particle activ-
ity as required by the regulations. If a composite of quarterly collected
samples is assayed to minimize analytical expenses the cost for initial survey
will be $400,000, Table 2, which lists estimated monitoring costs. The
Agency recognizes that the Interim Regulations impose a national program
to determine once and for all which community water systems require fur-
139
-------
DRINKING WATER REGULATIONS
ther testing for radium contamination. In order to ameliorate the financial
impact of this requirement, the Interim Regulations allow samples to be
collected over a three year interval and the substitution of measurements
made one year previous to the effective date of the regulations. The Agency
considered the possibility of using geological information in selecting which
systems should be tested for radium contamination. The poor predictive
value shown in the past by such information, however, indicates such a pro-
cedure could fail to identify systems which exceed the maximum con-
taminant levels.
Estimated National Costs for Monitorinc Radioactivity
in All Community Water Systems-
Initial Annual Cost
Survey ' succeding year)
Dollars Dollars per Year
Public water system# serving more than
100,000 persona 15,000 4,000
Community systems potentially impacted
by nuclear facilitea 20,000 20,000
Gross alpha particle activity in all
community water systems 400,000 100,000
Radium-226 and radium-228 133,000 60,000
Estimated totals 568,000 184,000
'Based on an estimated 40,000 community water systems including an estimated 60
systems impacted by nuclear facilities. The estimates of initial cost are high since
States are permitted to substitute equivalent data.
Cost estimates for radium-226 and radium-228 analyses are based on the
assumption that, nationally, ten percent of the approximately 35,000 systems
using ground water will exceed the screening level for gross alpha activity
and therefore require further testing. The Agency recognizes that in some
States a much higher percentage of the systems will require radium analyses
and that these costs will be distributed very unevenly. Of the 3500 systems
analyzing for radium it is assumed that about 700 will also be required to
assay for radium-228, Table I.
After the initial survey, a subsequent gross alpha particle anlysis is re-
quired every four years both for those systems utilizing surface water and
for those using ground water. Nationwide total annual cost in succeeding
years is estimated as $184,000, based on estimated assay costs of $10 for
gross alpha activity, $30 for radium-226, by the precipitation method and an
additional $15 if a subsequent radium-228 analysis is required.
Hie annual cost for radium assay in succeeding years is difficult to esti-
mate because it is highly dependent on the findings of the initial survey. For
the present the Agency has assumed that 500 systems will continue radium-
226 monitoring on a quarterly basis. This is the number of systems thought
to exceed the maximum contaminant limit, Table 1. The frequency at which
140
-------
APPENDIX B—RADIONUCLIDES
these 500 systems are monitored will be reduced as they come into compli-
ance with maximum contaminant levels.
The cost estimates shown in Table 2 do not make allowance for the cost
saving that will be realized by those States which use data already collected.
Monitoring Costs for Man-made Radioactivity
National monitoring costs for man-made radioactivity are smaller than
for natural radioactivity but costs for analysis of individual samples are
somewhat greater, Table 3.
Table 3.—Estimated Assay Costs for Man-made Radionuclides
$ Costs per sample
Gross beta activity 10
Tritium 20
Stxontium-90 30
fodine-131 60
Strontium-89 30
Ceaium-134 30
Except for community water systems directly impacted by nuclear facili-
ties, only an estimated 243 systems serving more than 100,000 persons and
utilizing surface water are required to monitor for man-made radioactivity.
Since monitoring for gross beta particle, tritium and strontium-90 activity is
required, the initial survey cost will be $15,000 and the annual cost for re-
survey every four years is $4,000.
The Administrator is allowing wide discretion to the States in determining
where quarterly monitoring in the vicinity of nuclear facilities will be re-
quired. Community water systems near nuclear facilities other than power
reactors and support facilities for the Uranium Fuel Cycle may be monitored
for man-made radionuclides at the option of the State. In some local situa-
tions a State may want to consider monitoring for contamination from waste
storage areas, and large experimental facilities and medical centers. Mon-
itoring is not expected at all community water systems within an impacted
water shed but only in those systems most likely to be contaminated.
At present about 40 nuclear power reactors have a potential for introduc-
ing man-made radioactivity into community water systems. The estimated
annual national cost for monitoring potentially impacted community water
systems is $20,000 based on the assumption that 60 community water sys-
tems may require assay. This cost will increase, of course, as the number of
nuclear facilities increases. The annual cost to an impacted system is esti-
mated as $330 per year.
REFERENCES
1. "National Interim Primary Drinking Water Regulations • RadioactivityFederal
Register. 41 Fit 133, p. 28402, July 9, 1976.
2. "The Effects on Populations of Exposure to Low Levels of Ionizing Radiation,"
Division of Medical Sciences, National Academy of Sciences, National Research
Council, November 1972, Washington, D. C.
141
-------
DRINKING WATER REGULATIONS
3. "The Evaluation of the Risks from Radiation," ICRP Publication 8, Pergamon
Press, New York, N. Y. 1966.
4. "Radiation Protection Guides for Federal Agencies," Federal Radiation Council,
Federal Register, 26fr 9057, September 26, 1961.
5. "Background Material for the Development of Radiation Protection Standards,"
Federal Radiation Council, Report #2, U. S. Department of Health, Education and
Welfare, USPHS, Washington, D. C., September 1961.
6. "Maximum Permissible Body Burdens and Maximum Permissible Concentrations of
Radionuclides in Air and Water for Occupational Exposure," NBS Handbook 69,
Department of Commerce, revised 1963.
7. "Costs of Radium Removal from Potable Water Supplies," Singley, J.E., et. al.,
Office of Research and Development, U. S. EPA, to be published.
8. Report to U. S. Environmental Protection Agency, "Radium-226 and Water Sup-
plies," by Conrad P. Straub, Ph.D., Director, Environmental Health and Research
Training Center, University of Minnesota.
9. Health and Safety Laboratory Environmental Quarterly, HASL-294, Energy Re-
search and Development Administration, New York, N.Y.
10. "Interim Radiochemical Methodology for Drinking Water,** EPA-600/4-75-008,
Environmental Monitoring and Support Laboratory, Office of Research and De-
velopment, USEPA, Cincinnati, Ohio, September 1975.
142
-------
APPENDIX B—RADIONUCLIDES
APPENDIX I
EPA Policy Statement on Relationship Between Radiation Dose
and Effect, March 3, 1975
The actions taken by the Environmental Protection Agency to protect pub-
lic health and the environment require that the impacts of contaminants
in the environment or released into the environment be prudently examined.
When these contaminants are radioactive materials and ionizing radiation,
the most important impacts are those ultimately affecting human health.
Therefore, the Agency believes that the public interest is best served by the
Agency providing its best scientific estimates of such impacts in terms of
potential ill health.
To provide such estimates, it is necessary that judgements be made which
relate the presence of ionizing radiation or radioactive materials in the en-
vironment, i.e., potential exposure, to the intake of radioactive materials in
the body, to the absorption of energy from the ionizing radiation of differ-
ent qualities, and finally to the potential effects on human health. In many
situations the levels of ionizing radiation or radioactive materials in the en-
vironment may be measured directly, but the determination of resultant
radiation doses to humans and their susceptible tissues is generally derived
from pathway and metabolic models and calculations of energy absorbed.
It is also necessary to formulate the relationships between radiation dose
and effects; relationships derived primarly from human epidemiological
studies but also reflective of extensive research utilizing animals and other
biological systems.
Although much is known about radiation dose-effect relationships at high
levels of dose, a great deal of uncertainty exists when high level dose-effect
relationships are extrapolated to lower levels of dose, particularly when
given at low dose rates. These uncertainties in the relationships between dose
received and effect produced are recognized to relate, among many factors, to
differences in quality and type of radiation, total dose, dose distribution,
dose rate, and radiosensitivity, including repair mechanisms, sex, variations
in age, organ, and state of health. These factors involve complex mechan*
ims of interaction among biological, chemical, and physical systems, the
study of which is part of the continuing endeavor to acquire new scientific
knowledge.
Because of these many uncertainties, it is necessary to rely upon the con-
sidered judgments of experts on the biological effects of ionizing radiation.
These findings are well-documented in publications by the United Nations
Scientific Committee on the Effects of Atomic Radiation (UNSCEAR), the
National Academy of Sciences (NAS), the International Commission on
Radiological Protection (ICRP), and the National Council on Radiation
Protection and Measurements (NCRP), and have been used by the Agency
in formulating a policy on relationship between radiation dose and effect
143
-------
DRINKING WATER REGULATIONS
It is the present policy of the Environmental Protection Agency to assume
a linear, nonthreshold relationship between the magnitude of the radiation
dose received at environmental levels of exposure and ill health produced
as a means to estimate the potential health impact of actions it takes in de-
veloping radiation protection as expressed in criteria, guides or standards.
This policy is adopted in conformity with the generally accepted assumption
that there is some potential ill health attributable to any exposure to ionizing
radiation and that the magnitude of this potential ill health is directly pro-
portional to the magniture of the dose received.
In adopting this general policy, the Agency recognizes the inherent un-
certainties that exist in estimating health impact at the low levels of ex-
posure and exposure rates expected to be present in the environment due
to human activities, and that at these levels the actual health impact will not
be distinguishable from natural occurrences of ill health, either statistically
or in the forms of ill health present. Also, at these very low levels, mean-
ingful epidemiological studies to prove or disprove this relationship are dif-
ficult, if not practically impossible, to conduct. However, whenever new in-
formation is forthcoming, this policy will be reviewed and updated as neces-
sary.
It is to be emphasized that this policy has been established for the purpose
of estimating the potential human health impact of Agency actions regarding
radiation protection, and that such estimates do not necessarily constitute
identifiable health consequences. Further, the Agency implementation of this
policy to estimate potential human health effects presupposes the premise
that, for the same dose, potential radiation effects in other constituents of
the biosphere will be no greater. It is generally accepted that such constitu-
ents are no more radiosensitive than humans. The Agency believes the policy
to be a prudent one.
In estimating potential health effects it is important to recognize that the
exposures to be usually experienced by the public will be annual doses that
are small fractions of natural background radiation to at most a few times
this level. Within the U. S. the natural background radiation dose equivalent
varies geographically between 40 to 300 mrem per year. Over such a rela-
tively small range of dose, any deviations from dose-effect linearity would
not be expected to significantly affect actions taken by the Agency, unless
a dose-effect threshold exists.
While the utilization of a linear, nonthreshold relationship is useful as a
generally applicable policy for assessment of radiation effects, it is also
EPA's policy in specific situations to utilize the best available detailed
scientific knowledge in estimating health impact when such information is
available for specific types of radiation, conditions of exposure, and recip-
ients of the exposure. In such situations, estimates may or may not be based
on the assumptions of linearity and a nonthreshold dose. In any case, the
144
-------
APPENDIX B—RADIONUCLIDES
assumptions will be slated explicitly in any EPA radiation protection
actions.
The linear hypothesis by itself precludes the development of acceptable
levels of risk based solely on health considerations. Therefore, in establish-
ing radiation protection positions, the Agency will weigh not only the health
impact, but also social, economic and other considerations associated with
the activities addressed.
145
-------
DRINKING WATER REGULATIONS
APPENDIX H
Risk to Health from Internal Emitters
A. The Dose and Health Risk from Radium Ingestion
The Federal Radiation Council has also recommended radium-226 in-
gestion limits for the general population and stated that such limits should
be based on environmental studies not the models used to establish occupa-
tional dose limits (1). The FRC ingestion limit is based on the assumption
that the skeletal radium-226 burden does not exceed 50 times the daily ra-
dium intake. This assumed relationship between ingestion and body burden
agrees quite well with the measurements of skeletal body burdens and ra-
dium ingestion data reported by the U. N. Scientific Committee on the Ef-
fects of Atomic Radiation (2). By comparing Tables 9 and 10 in reference
(2) it is seen that the skeletal burden is about forty times the estimated
daily radium-226 intake.
The FRC limit on radium ingestion is 20 pCi per day.* After continuous
ingestion at this limit the skeletal body burden is 1000 pCi. Ingestion of 2
titers of drinking water per day containing radium-226 at a maximum con-
taminant level of 5 pCi per liter would result in a skeletal burden of 500 pCi.
In order to estimate potential health effects from radium ingestion, it is
necessary to express the dose equivalent from this body burden in terms of
the ICRP dose model which was used in the dose estimates made in the NAS
BEIR Report (3). The ICRP model predicts an average dose to bone of
about 30 rem per year from a body burden of 100,000 pCi (2). A body bur-
den of 500 pCi would therefore cause an average dose of 150 mrem per
year.
The NAS BEIR Report (Table 3-2) gives the rate of absolute risk from
bone cancer as four percent of all non leukemia type cancers (3). For a life-
time risk plateau and continuous lifetime exposure (Table 3-1 in reference
3) the number of bone cancers per year is 3 per 10® man-rem per year, esti-
mated on the basis of absolute risk.
Relative risk, the number of cancers expected on the basis of their per-
cent increase in an irradiated population, is also estimated in the BEIR for
total body exposure, Table 3-1. The NAS-BEIR committee risk report does
not give a breakdown by cancer site of the relative risk per rem. Assuming
that bone cancers are fouT percent of the relative risk from total body ex-
posure, excluding leukemia as before, the relative risk oi bone cancer is
about 17 per year per 10® man-rem per year.
Bone cancer is not the only risk from radium ingestion. About 15 percent
of the radium is deposited in soft tissue where bone marrow is the primary
tissue at risk. Doses to soft tissue relative to those in bone from ingested
radium have been calculated in reference 2, Table 9. The risk to these tissues
from radium ingestion has been calculated by weighing the risk estimates
'Range II, averaged over a suitable sample (1).
146
-------
APPENDIX B—RADIONUCLIDES
for leukemia (and other cancers) given in the NAS-BEIR Report, by the
appropriate organ dose. The total absolute risk due to bone and soft tissue
cancers is 60 percent larger than that from bone cancer alone; the relative
risk, 16 percent greater. Therefore, the annual rate of total cancers from
ingesting radium ranges from 4.8 (3 x 1.6) to 20 (17 x 1.16) per million
man-rem/year depending on whether an absolute or relative risk model is
used.
Combining these estimates of the annual risk of total cancer with the
ICRP dose to bone, 0.15 rem per year, from the ingestion of 10 pCi of ra-
dium-226 per day yields the range of estimated health effects from radium
ingestion, 0.7 to 3 cancers per year, per million exposed persons. Almost all
of any induced cancers would be fatal. Bone cancer fatality is estimated at
nearly % percent, that for leukemia is much higher.
Given the assumption that radiation damage occurs at incremental doses
greater than those due to external background radiation, the total health im-
pact from a public water supply system can be estimated on the basis of the
total dose received by the population it serves. This aggregate dose can be
calculated by multiplying the number of persons served by the average dose
received by a reference man consuming two liters of drinking water per day.
Based on the geometric mean of the individual risk discussed above, a ra-
dium concentration of 5 pCi per liter in a water system serving 1,000,000
persons could result in an estimated health impact o! 1.5 fatalities per year
or about 3 x 10-7 per person per year for each pCi per liter of radium-226
or radium-228 in the drinking water. As is shown in Appendix III, this
number can be used to estimate the marginal cost effectiveness of radium
control in public water systems to prevent cancer. However, it must be kept
in mind that the risk estimates are uncertain by a factor of four or more.
B. The Relative Health Risk of Radium-228 as Compared to Radium-226
Unfortunately, guidance on the body burden from chronic radium-228 in-
gestion was not provided by the Federal Radiation Council in their discus-
sion of radium-226 limits. Because Handbook 69, which is based on 1959
ICRP dose models (4), gives a maximum permissible concentration in water
for radium-228 that is three times greater than for radium-226, many per-
sons have concluded that these two isotopes are not equally toxic. However,
more recent data (particularly that in the 1972 UNSCEAR report (2) and
the 1972 ICRP Report (4) on alkaline earth metabolism) indicates that
radium-228 is at least as toxic as radium-226.
There are two major difficulties with the old ICRP model. It assumes for
radium-226 an effective half-life in bone of 1.6 x 104 days (44 years) and
because of the shorter physical half-life of radium-228 an effective half-life
of 2.1 x 103 days (5.8 years) for radium-228. Therefore, using the old ICRP
model, on the basis of effective half-life the body burden due radium-226
would be 7.6 times greater than that calculated for radium-228 for equal
daily intakes of each.
147
-------
DRINKING WATER REGULATIONS
The recent report from the ICRP Committee II task group on alkaline
earth metabolism shows that the old ICRP bone model overestimated the
effective half-life of radium-226 and that 17.1 years, not 44, is currently the
best estimate of the half-time for radium retention (5).* On this basis the
effective half-life of radium-228 (physical half-life 5.75 years) (5) is 4.3
years, assuming the half-time of radium-226 retention is a reasonable esti-
mate of the biological half-life of radium. In light of this new information,
the body burden from chronic radium-226 ingestion is about four, times
greater than that from radium-228, not 7.6 times greater as predicted by the
old ICRP model.
The old ICRP model also underestimates the effective energy delivered to
bone from a given body burden. The old ICRP model assumes that 50 per-
cent of the radon-220 (physical half-life 55 sees) produced in the radium-
228 decay chain escapes from bone as compared to an assumed 70 percent
escape of the radon-222 (physical half-life 3.8 days) produced in the ra-
dium-226 decay chain. Speculation on this point is unnecessary. The MIT
Radioactivity Center has measured the escape of this short half-life radon-
220 from bone and found it to be about one to two percent (6).
Since almost all of the radon-220 decay products are retained in bone,
the effective energy per disintegration of radium-228 in bone is about 330
MEV, not 190 MEV as given by the old ICRP #2 model. The effective en-
ergy for radium-226 in the old ICRP model is 110 MEV, a factor of three
less than that for radium-228.
The average dose to bone due to continuous radium ingestion (based on
an expontential retention function) is proportional to the effective half-life
and effective energy;
for radium-226 this product is 17.1 years x 110 MEV = 1880.
for radium-228 this product is 4.3 years x 330 MEV = 1420.
which indicates that even on the basis of a single exponential retention model,
as used in reference (4) these two radionuclides give approximately the same
dose per unit activity ingested.
Actually, a simple exponential retention model is not a very good approx-
imation of radium retention in man and the more sophisticated model based
on studies in humans that were not available in 1959 (5) is currently being
considered by ICRP Committee II.
This new ICRP model on alkaline earth metabolism, indicates that for
equal intakes the 50 year dose to bone surfaces from radium-228 is signifi-
cantly greater than that from radium-226. Experimental data given in the
1972 UNSCEAR report supports this viewpoint (2). In the United States
the average daily ingestion of radium-226 and radium-228 is about equal.
Table 10 in reference 2. Table 9 in reference 2 shows that the dose to bone
*n.b. that since the old ICRP model vu used to calculate both radium doaet sad
health effects this change does not change the risk estimates given is II-A.
148
-------
APPENDIX B—RADIONUCLIDES
surfaces, calculated on the basis of measured skeletal body burdens of ra-
dium-226 and radium-228, is greater for radium-228 than for radium-226.
Since radium carcinogenicity is associated with the dose to bone surfaces
(7), it is likely that radium-228 is more of a health risk than radium-226.
Experimental findings in dogs bear this out. The measured relative bi-
ological effectiveness of radium-228 is over twice as great as radium-226
when death by osteosarcomas is used as an end point (8). Though the car-
cinogencity of radium-228 relative to radium-226 may not be as great in
man as in dogs, it is prudent to assume chronically ingested radium-228 is
at least as dangerous as radium-226.
REFERENCES
1. "Background Material for the Development of Radiation Protection Standards,"
Federal Radiation Council, Report #2, U. S. Department of Health, Education and
Welfare, USPHS, Washington, D.C., September 1961.
2. "Ionizing Radiation Levels and Effects," Vol. I, United Nations Publication
E.72.IX.17, 1972, New York, N. Y.
3. "The Effects on Populations of Exposure to Low Levels of Ionizing Radiation," Divi-
sion of Medical Sciences, National Academy of Sciences, National Research Council,
November 1972, Washington, D. C.
4. Report of Committee II on Permisaable Dose for Internal Radiation, ICRP Publica-
tion 2 (1959), Pergamon Press, New York, N. Y.
5. 44Alkaline Earth Metabolism in Adult Man/' ICRP Publication 20, 1972, Pergamon
Press, New York, N. Y.
6. Evans, R. D., "Radium and Mesotborium Poisoning and Dosimetry and Instrumen-
tation Techniques in Applied Radioactivity," MIT-9S2-3, 1966, Division of Technical
Information, ORNL, Oak Ridge, Tennessee.
7. "A Review of the Radiosensitivity of the Tissues in Bone," ICRP Publication 11,
1968, Pergamon Press, New York, N. Y.
8. Dougherty, T.F. and Mays, C.W., "Bone Cancer Induced by Internally Deposited
Emitter* in Beagles," Radiation Induced Cancer, IAEA-SM-118/3, 1969, Interna-
tional Atomic Energy Agency, Vienna, Austria.
149
-------
DRINKING WATER REGULATIONS
APPENDIX IE
The Cost and Cost-Effectivene&s of Radium Removal
The United States Environmental Protection Agency planning guide for
water use provides estimates of the amount of water used per day by various
population groups (1). Per capita water consumption increases with com-
munity size because of industrial and commercial usage. In this cost analysis
a water use of 100 gallons per person day is assumed. This may be some-
what high since mainly small community systems, serving less than 10,000
persons, would be impacted by the proposed regulations.
"Selecting a Softening Process," by Frank 0. Wood, has served as the
Agency's primary reference for assessing the cost of zeolite treatment to re-
move radium(2). Wood surveyed a representative sample of community
water systems to determine their construction and operating costs for water
softening in order to compare the economics of lime-soda ash softening with
treatment by ion exchange. Zeolite ion exchange was the lower cost opera-
tion for public water systems serving fewer than about 50,000 persons and
therefore is applicable to all systems which may require radium abatement.
Wood's report shows that while the cost per 1000 gallons increases slightly
with system capacity, per 1000 gallons is a conservative average value for
systems supplying less than 1 million gallons per day. Because plants exam-
ined by Wood had been built over a period of several years, he normalized
costs in terms of the 1967 wholesale price index to place them on an equal
chronology basis. For this analysis Wood's estimates have been updated to
1975 by means of the "Sewage Treatment Plant Construction Cost Index,"
prepared by the United States Environmental Protection Agency Office of
Water Programs Operations. From 1967 to January 1975 the index in-
creased by about 90%. Therefore, for the cost analysis for radium removal
the Agency has assumed a treatment cost of 15tf per 1000 gallons. It should
be noted that these costs include amortization of capital costs over a 20 year
period as well as chemical costs for regeneration of the zeolite system. Labor
costs for equipment operation are not included since these costs were too
small to be included in Wood's analysis; the equipment is essentially
automatic.*
Usually only a fraction of the supply water need be treated since the mix-
ing of treated and untreated water is an acceptable abatement procedure.
The fraction of water treated, F, to achieve a given radium concentration is
calculated as follows:
r„ 1 — C«/C„
e
where CQ is the radium concentration in untreated water, C, is the average
radium concentration in treated and untreated waters and e is the efficiency
of radium removal.,
'Recently completed studies indicate that addition of labor costs would increase the
treatment cost by about 2( per 1000 gallons (3).
150
-------
APPENDIX B—RADIONUCLIDES
The efficiency at which radium is removed from water by a zeolite ion
exchange column is very high, approaching 99% for a newly charged
column and falling to around 90% just before breakthrough in a spent
column. The results listed below are based on an estimated overall removal
efficiency of 97 percent.
The volume of water that must be treated per person year to reduce the
radium concentration from (n) pCi/1 to (n-1) pCi/1 is shown in Table III-l
along with the annual marginal cost per pCi/1 removed to treat this volume
of water. Costs are based on 15l per 1000 gallons, as outlined above. For
concentrations greater than 5 pCi/l the annual per capita cost ranges from
about 60 cents to 90 cents per pCi/1 removed depending on the initial
concentration.
Each decrement of the average annual concentration of radium by 1 pCi/1,
corresponds to an estimated health savings of approximately 3 x 10-7 ex*
cess cancers averted per year, Appendix II-A. Dividing this number by the
annual expenditure required to obtain a given concentration yields the esti-
mated marginal costs per cancer averted shown in Table III-l. The mar-
ginal cost increases slowly as the radium concentration is decreased until
at about 2-3 pCi per liter the cost per estimated excess cancer averted in-
creases more rapidly due to the larger fraction of the water needing treat-
ment to achieve smaller concentrations.
Table III-l—The Marginal Cost-Effectiveness of Radium Removal*
Initial
Volume of Water
Annual Cost
Marginal Cost to
Radium
Treated Per Person
Per Person to
Prevent One
Concentration
Year
Remove One
Cancer
pCi per liter
(pCi/l)
(1000 gallons)
(dollars)
(millions of dollars)
10
3.8
0.57
1.88
9
4.2
0.63
2.09
8
4.7
0.71
235
7
52
0.78
2.61
6
6.3
0.94
3.14
5
7 S
1.13
3.77
4
9.4
1.41
4.71
3
12.6
1.88
6.28
2
18.9
2.82
9.41
1
36.5
5.48
18.83
•by zeolite ion exchange
REFERENCES
1. Manual of Individual Water Supply Systems, EPA-430-9-74-007, U. S. Environmental
Protection Agency, 1974, Superintendent of Document!, U. S. Government Printing
Office, Waabington, D.C. 20402.
2. Wood, Frank 0., "Selecting a Softening Process," Journal AWWA pp. 820-824,
December 1972.
3. "Costa of Radium Removal from Potable Water Supplies," to be published.
151
-------
DRINKING WATER REGULATIONS
APPENDIX IV
Dosimetric Calculations for Man-Made Radioactivity
A. Calculations Based on NBS Handbook 69
The dose rate from radioactivity in drinking water is calculated on the
basis of a 2 liter daily* intake. Except for tritium and strontium-90, see be-
low, the concentrations of man-made radionuclides causing 4 millirem per
year have been calculated using the data in NBS Handbook 69 (1) and
are tabulated in Table IV-2A and IV-2B. The dose models used in preparing
Handbook 69 are outlined in reference 2. Maximum Contaminant Levels are
defined in terms of the annual dose equivalent to the total body or any in-
ternal organ. Handbook 69 lists the critical organ for each radionuclide.
Often the total body is listed as the critical organ. The 168 hour maximum
permissible concentrations for ingestion in Handbook 69 are not calculated
on the basis of the same annual dose to each critical organ as in the Interim
Regulations, rather different organ doses are permitted by occupational ra-
diation protection limits (ORL), Table IV-1.
Table IV-1. Occupational Radiation Limits
(ORL)
Critical Organ ORL (rents)
Total body 5
Gonads 5
Thyroid 30
Bone 29.1 (a)
Other Organs 15
(a) Based on the alpha energy deposited in bone by 0.1 pCi of radium-226.
The maximum permissible concentrations for a 168 hour week, MPC, in
Handbook 69, assume ingestion at 2.2 liters per day and are in units of pCi
per cc. The various numerical factors can be combined to find C4, the con-
centration causing 4 mrem per year from 2 liters daily ingestion of drink-
ing water as follows:
C4 «¦ 4.4 x 10® x MPC pCi per liter
ORL
Critical organs are identified by boldface type in Handbook 69 so that an
appropriate ORL can be selected from Table IV-1.
To illustrate, a sample calculation, taken from page 24 of Handbook 69
is given:
*Tbe recent ICRP publication #23, "Report of the Task Croup on Reference Man,"
(3) gives the total daily water intake as 3 liters, 1.95 liters by fluid intake, the bal-
ance by food and food oxidation. Almost all of the fluid intake is from tap water and
water baaed drinks (Page 360).
152
-------
APPENDIX B—RADIONUCLIDES
Radionuclide
BerylIium-7 MPC (168 hours) 0.02 uCi/cm3
Listed critical organ GI(LLI) gastrointestional tract
(lower large intestine)
r a a 0-02 pCi
C4 = 4.4x10« x -J?- -£j- = 5867 pCi/1
= 6000 pCi/1
Rounding is appropriate since the values in Handbook 69 are given to one
significant figure.
Calculation of the dose resulting from the ingestion of drinking water
containing a known mixture of radionuclides is straightforward. Let A,
B, ... be the concentrations, in pCi per liter, of isotopes a, b, ... in the
water and let C
-------
DRINKING WATER REGULATIONS
dose. An example is tritium where two or three exponentials may be needed
to describe the dose-time relationship of ingested tritium (4). Some in-
vestigators have estimated that following chronic ingestion organically
bound tritium may increase the dose by a factor of 1.4 to 1.5 over that
predicted by Handbook 69 (5). Such estimates are too high because or-
ganically bound tritium irradiates the total body mass, and not just the
mass of body water, as assumed irr the model used in Handbook 69 (2).
Consideration of the daily intake of hydrogen and water shows that the
tritium concentration (specific activity) in any organ is no greater than
120% of the tritium concentration in body water. The concentration of
tritium in body water following chronic ingestion is T/3 where T is daily
intake of tritium in pCi and the total water intake, including that in food,
is 3 liters per day (3). Water content by weight of any organ does not
exceed 80 percent (4). Therefore, equilibrium concentration of tritium in
any organ due to its water content, can not exceed 0.8 T/3 = .267 T pCi/kg.
Because of organically bound hydrogen an organ's hydrogen (and
tritium) content is greater than that due to water alone. The daily hydrogen
intake is .35. kg per day (3) and, since no organ contains more than 11
percent hydrogen by weight (4), the maximum tritium concentration in
any organ following chronic ingestion is .11 T/.35 — .314 T pCi/kg. The
specific activity of tritium in any organ due to bound and unbound hy-
drogen exceeds that due to its water content alone by the ratio .314/.267 —
1.18. Therefore, the dose to any organ due to organically bound tritium
exceeds the dose to body water, given in Handbook 69, by no more than
about twenty percent.
The Agency is aware that the ICRP is developing new tritium dose
models more suitable for environmental sources of tritium exposure than
the model used in Handbook 69. Until these models are published and rec-
ommended by the Agency, the maximum contaminant level for tritium is
calculated on the basis of 80 percent of the value calculated using NBS
Handbook 69.* For tritium in drinking water:
C« = 0.8 x 4.4 x 10fl x = 21,120 pCi/1
= 20,000 pCi/1
The maximum contaminant level for strontium-90 in the Interim Regula-
tions is based on the dose model used by the Federal Radiation Council
(FRC) to predict the dose to bone marrow (6). According to the FRC
model a continuous daily intake of 200 pCi per day of strontium-90 will
result in a body burden of 50 pCi per gram of calcium in bone. The annual
•n.b. In accordance with current guidance to Federal agenciet, a quality factor of
1.7, u in Handbook 69, ia uaed in this calculation.
154
-------
APPENDIX B—RADIONUCLIDES
dose rate to bone marrow from this body burden would be 50 mrem per
year (7). Therefore, continuous ingestion of 16 pCi per day would result
in 4 mrem per year, the limit for man-made radionuclides in drinking water.
For two liters ingestion of water per day:
C, = 16 P2C' = 8 pCi/1
C. Concentrations yielding an Annual Dose of 4 Millirem
Tables IV-2A and IV-2B give C4 the annual average concentrations for
man-made radionuclides which are assumed to yield an annual dose of 4
millirem to the indicated organ. Table IV-2A comprises those nuclides
having half-lives greater than one day. Table IV-2B contains shorter half-
life radionuclides not expected to appear in drinking water supplies. In-
gestion at a rate of 2.0 liters per day is assumed. The values shown were
calculated from the Maximum Permissible Concentrations listed in Hand-
book 69 (1) as outlined above.
Table IV-2A. Annual Average Concentrations Yielding 4 Millirem per Year for a Two
Liter Daily Intake
(Half-life greater than 24 hours)
Radionuclide
Critical Organ
c4
(pCi/1)
Tritium
Total Body
20,000
~Be"
GI (LU)
6,000
6C14
Fat
2,000
UNa22
Total Body
400
lSp32
Bone
30
I6g35
Testis
500
17Q36
Total Body
700
20Ca«
Bone
10
20Cai7
Bone
80
«Sc«
GI (LU)
1,000
GI (LLI)
300
21Sc+8
GI (LU)
80
23V48
GI (LLI)
90
2*Cr«
GI (LU)
6.000
25Mn"
GI (LLI)
90
25M034
GI (LU)
300
2«Fesa
Spleen
2,000
2flpeS9
GI (LLI)
200
27Co"
GI (LU)
1,000
2TC0M
GI (LLI)
9,000
2?Coao
GI (LLI)
100
2SNi5»
Bone
300
28^83
Bone
50
30Zn«5
Liver
300
32G«71
GI (LLI)
6,000
38A.TS
GI (LU)
1,000
GI (LLI)
100
GI (LU)
60
33 At"
GI (LU)
200
155
-------
DRINKING WATER REGULATIONS
34Se"
Kidney
900
35Br82
GI (LLI)
100
37Rb80
Total Body
600
37Rb87
Pancreas
300
38gr85
GI (SI)
21,000
38Sr89
Bone
20
38gr80
Bone Marrow (FRC)
80
38gr00
Bone Marrow (FRC)
8
39Y«0
GI (LLI)
60
36Y91
GI (LLI)
90
40^3
GI (LLI)
2,000
~oZjO#
GI (LLI)
200
41Nb93»
GI (LLI)
1,000
*1Nbd6
GI(LLl)
300
«Mo»»
Kidney
600
43TC06
GI(LLI)
300
~3TC07»
GI(LLI)
1,000
43*rc97
GI(LLI)
6,000
~3Tca0
GI(LLI)
900
«Ru<"
GI(LLI)
1,000
^Ru103
Gil LLI)
200
44RU1(*
GI(LLI)
30
45RJ,106
GI(LLI)
300
~apjioa
GI(LLI)
900
lapdioe
GI(LLI)
300
47AglOS
GI(LLI)
300
47^gllO«
GI(LLI)
90
4TAgltl
GI(LLI)
100
48Cd»09
GI(LLI)
600
48Cdl""
GI(LLI)
90
48CdWB
GI(LLI)
90
4»InlJB
GI(LLI)
300
MSnlW
GI(LLI)
300
50SnX2B
GI(LLI)
60
sisb^s
GI(LLI)
90
51Sb12*
GKLLI)
60
«Sb"s
GI(LLI)
300
52Te125»
Kidney
600
52TC127™
Kidney
200
52Te127
GKLLI)
900
52Te129»
GKLLI)
90
52Tel29
GI(S)
2,000
62Xe131ra
GKLLI)
200
52Te132
GKLLI)
90
631126
Thyroid
3
531120
Thyroid
1
5SJ131
Thyroid
3
B5Q131
Total Body
20,000
»»C«18*
GI(S)
20,000
5BC#135
Total Body
900
88CS136
Total Body
800
06(2(137
Total Body
200
568*131
GKLLI)
600
156
-------
APPENDIX B—RADIONUCLIDES
OOfiaHO
GKLLI)
90
57 La140
Gl(LLI)
60
58Ce141
GKLLI)
300
5S0ei«
GKLLI)
100
06pr143
GKLLI)
100
GKLLI)
3,000
7»Ptl»3
Kidney
3,000
78ptl07
GKLLI)
300
70AU10«
GKLLI)
600
70Au198
GKLLI)
100
81T12M
GKLLI)
300
82pb203
GKLLI)
1,000
83Bi20fl
GKLLI)
100
8jfBi207
GKLLI)
200
ttlpa233
GKLLI)
300
Table IV • 2B
Annual Average Concentrations Yielding 4 Miilirem
per Year for a Two Liter Daily Intake
(Half-life less than 24 hours)
Radionuclide
Critical Organ
(pCi/1)
0pi8
GKSI)
2,000
i*Siw
GKS)
3,000
17(338
GKS)
1,000
157
-------
DRINKING WATER REGULATIONS
19R42
GI(S)
900
28MnS6
GI(LLI)
300
2TCo58»
GKLLI)
300
28^66
GI(LLI)
300
2»cua4
GKLLI)
900
30Zn
GKLLI)
60
49In115™
GKULI)
1,000
S3|132
Thyroid
90
931133
Thyroid
10
53J134
Thyroid
100
53J135
Thyroid
30
50C8134">
Total Body
80
S0pr142
GI(LLl)
90
60[yfdl*9
GKLLI)
900
6SEU1»2
GKLLI)
200
MGdWO
GKLLI)
200
ft6])yl65
GKLLI)
1,000
<»Erm
GKULI)
300
74^187
GKLLI)
200
76ReX88
GKLLI)
200
7«081M«»
GKLLI)
9,000
771,194
GKLLI)
90
78pt197»
GKULI)
3,000
81TJ203
GKLLI)
300
158
-------
APPENDIX B—RADIONUCLIDES
REFERENCES
1. "Maximum Permissible Body Burdens and Maximum Permissable Concentrations
of Radionuclides in Air and Water for Occupational Exposure," NBS Handbook 69,
Department of Commerce, revised 1963.
2. Report of Committee II on Permissible Dose for Internal Radiation, ICRP Publi-
cation 2 (1959), Pergamon Press, New York, N. Y.
3. Report of the Task Group on Reference Man, ICRP Publication 23, 1975, Pergamon
Press, New York, N. Y.
4. Snyder, W, S., Fish, B. R., Bernard, S. R., Ford, M. R. and Muir, J. R., "Urinary
Excretion of Tritium Following Exposure of Man to HTO-A Two-Exponential Model,
"Physics in Medicine and Biology, Vol. 13, p. 547, 1968.
5. Evans, A. G., "New Dose Estimates from Chonric Tritium Exposures," Health
Physics, Vol. 16, pp. 57-63, 1969.
6. "Background Material for the Development of Radiation Protection Standards,"
Federal Radiation Council, Report #2, U,S> Department of Health, Education and
Welfare, USPHS, Washington, D. C., September 1961.
7. "Estimates and Evaluation of Fallout in the United States from Nuclear Weapons
Testing Conducted through 1962", Federal Radiation Council, Report #4, U.S.
Department of Health, Education and Welfare, USPHS, Washington, D. C., May
1963.
159
a3S3Q0698403
------- |