Environm'enTa, Protection               EPA-600/6-81-001
                Agency                        January 1981
4>EPA        Research  and

                OF CARBARYL
                Prepared for
                Prepared by

                Office of Health and
                Environmental Assessment
                Washington DC 20460

                                   TECHNICAL REPORT DATA
                           (I'h-asc read Instructions on the reverse before completing)
 . flfPORT NO.
-zacaiicii IOTORMAIION
                                                           3. RECIPIENT'S ACCESSION-NO.
  Preliminary Report on the Mrtagenicity of
                                                           5. REPORT DATE
                                                             January 1981
                                  6. PERFORMING ORGANIZATION CODE
                                                           8. PERFORMING ORGANIZATION REPORT NO.
 Vicki Vaughan-Dellarco,  Ph.D.
  Reproductive Effects Assessment Group
  Office of Health and Environmental Assessment
  Office of Research and  Development
  U.  S.  EPA, Washington,  D.C.   20460
                                  10. PROGRAM ELEMENT NO.

                                    ABX E1A
                                  11. CONTRACT/GRANT NO.

  Special Pesticide Review Division
  Office of Pesticide  Programs
  Office of Pesticide  and Toxic Substances
  U.S.  EPA, Washington,  D.C.   20460
                                  13. TYPE OF REPORT AND PERIOD COVERED
                                    Response Report
                                  14. SPONSORING AGENCY CODE
16-ABSTRC§fbaryl  has  been reported to cause  point  mutations in bacteria,
   Drosophila,  and mammalian cells in vitro;  unscheduled DNA snythesis  in  human
   cells in culture;  and chromosome effects  (including spindle effects)  in
   plants, mammalian  cells in vitro, and  animals.   Although there are
   inadequacies  in these studies, the results when considered together  are
   strongly suggestive that carbaryl may  have the intrinsic ability  to  act as a
   mutagen.   It  should be noted that because  carbaryl  appears to act as  a  weak
   mutagen in the  experimental test systems,  it  is likely to act as  a weak
   mutagen in humans  as well.  To cause heritable effects in humans, however,
   carbaryl and/or an active metabolite(s)  must  reach the germinal tissue.
   Evidence that carbaryl reaches the mammalian  gonad is considered  suggestive.
   Adverse gonadal effects, e.g., abnormal  sperm morphology (Degraeve et al.
   1976), reduction  in the number of spermatogonia and spermatozoa in the
   seminiferous tubules, (Kitagawa et at.  1977)  reduced sperm motility
   (Shtenberg and  Rybakova 1968), and decreased  fertilization index  (Collins  et
   al. 1971)  have  been reported  in rodents  exposed to carbaryl.  In  addition,
   abnormal sperm  head morphology has been  reported in workers with  known
   exposure to carbaryl  (Wyrobek et al. 1980).  Therefore, carbaryl  may have  the
   potential  to act  as a germ-cell mutagen.
                                KEY WORDS AND DOCUMENT ANALYSIS
                     b.IDENTIFIERS/OPEN ENDED TERMS  C. COSATI Field/Group

  NTIS Release  to Public
                     19. SECURITY CLASS (This Report)
                                            21. NO. OF PAGES

                                              20. SECURITY CLASS (This page)

                                                                         22. PRICE
EPA Form 3220-1 (9-73)


                             ON THE MUTAGENICITY  OF

                                                       Peter E.  Voytek,  Ph.D.
                                                       Di rector
Prepared by

Vicki Vaughan-Dellarco, Ph.D,
Participating Members

John R. Fowle III, Ph.D,
Ernest Jackson, M.S.
Donna K. Kuroda, Ph.D.
David Mann, B.A.
Carol N. Sakai, Ph.D.


Purpose 	   1
Overview	1
Review of Experimental  Evidence on the Mutagenicity  of Carbaryl  	   3
    Evidence Concerning Gene Mutations  	   3
          Mammalian Cells in Culture
    Evidence Concerning Chromosomal  Effects 	  17
           Mammalian Cells in Culture
           In Vivo Cytogenetic Studies
           "Dominant Lethal Assay in Rodents
   Evidence Indicative of Primary DNA Damage  	  27
   Evidence Concerning Whether the Active Form of Carbaryl
    Reaches or Affects Germinal Tissue	30
Review of Experimental  Evidence on the Mutagenicity  of Nitrosocarbaryl.  .  35
Conclusion	37
References	39

    The Reproductive Effects Assessment Group (REAG)  conducted  an  extensive
review of published and unpublished data regarding the mutagenicity  of
carbaryl (Sevin) at the request of the Special  Pesticide  Review Division  of
the Environmental Protection Agency's Office of Pesticide Programs as part of
their Rebuttable Presumption Against Registration (RPAR).  The  acceptability
of the test results was determined and the  mutagenic  potential  of  carbaryl was

    Mutagenic chemicals are recognized as posing a potential  risk  to human
health because of their ability to cause heritable changes in genes  and
chromosomes.  Such germline changes, for example, can lead to spontaneous
abortions, birth defects, or the accumulation of deleterious mutations  in the
human gene pool.  In addition, somatic mutations may be involved in  the
etiology of cancer.
    The primary objective of a mutagenicity risk assessment is to  determine
the potential of a chemical to cause heritable germline effects in man
(Environmental Protection Agency's "Proposed Guidelines for Mutagenicity Risk
.Assessment," Federal Register 45:74984, November 1980).  Positive  responses
have been reported in three gene mutation test systems (bacteria,  Drosophila,
and mammalian cells in culture) using carbaryl (McCann, 1980 personal
communication, Cook et al. 1977, Rashid 1978, Egert and Greim 1976,  Brzheskii
1972, Ahmed et al. 1977a).  In addition, the results of cytogenetic tests
suggest  that carbaryl may induce chromosomal aberrations in mammalian cells  in
culture  (Ishidate and Odashima 1977, Kazarnovskaya and Vasilos 1977); carbaryl

h,as been shown to cause spindle effects in plants (Wuu and Grant 1966,  Amer
1965, Amer et al. 1971, Amer and Farah 1968,  Brankovan 1972)  and may cause
spindle effects in rodents (Vasilos et al. 1975a, b)  and in human cells j_n_
vitro (Vasilos et al.  1972, Shpirt 1975);  and carbaryl  has been  shown to  cause
unscheduled DNA synthesis, which is indicative of primary DMA damage, in
cultured human cells (Ahmed et al. 1977b).  Although  there are inadequacies in
the available studies, the results, when considered together, are strongly
suggestive that carbaryl may act as a mutagen.
    To cause heritable effects in humans,  however, a  chemical with Intrinsic
mutagenicity must reach the germinal tissue.   Evidence that carbaryl  and/or
its active metabolites reaches mammalian gonads is suggestive.  Adverse
gonadal effects, e.g., abnormal sperm morphology (Degraeve et al. 1976),
reduction in the number of spermatogonia and spermatozoa in the seminiferous
tubules, (Kitagawa et al. 1977) reduced sperm motility (Shtenberg and Rybakova
1968), and decreased fertilization index (Collins et al. 1971) have been
reported in rodents exposed to carbaryl.  In addition, abnormal  sperm head
morphology has been reported in workers with known exposure to carbaryl
(Wyrobek et al. 1980).  Therefore, carbaryl  may have  the potential to act as a
germ cell mutagen.
    Further experiments are needed, however, to better characterize the
mutagenic risk of carbaryl (e.g., gene mutation tests in Drosophila or
mammalian cells in vitro describing a dose-response,  and data demonstrating
alkylation or radiolabeling in germ cell DNA of intact animals).  It should be
noted that carbaryl is not a potent mutagen in the reported studies, and
probably acts as a weak mutagen only.
     It is well-established that nitrosocarbaryl, a nitroso derivative of
carbaryl, is a potent mutagen  in bacteria.  A dose-related increase in mutants

above the spontaneous mutation frequency has been observed (Blevins  et al.
1977, Elespuru et al. 1974, Marshall  et al.  1976, Seller 1977,  Greim et al.
1977, and Uchiyama et al.  1975).   Also, some investigators have shown that
nitrosocarbaryl can cause  primary DNA damage as manifested by the  ability to
strongly induce mitotic gene conversion in yeast (Siebert and Eisenbrand 1974)
and by the detection of DNA strand breaks and unscheduled DNA synthesis in
cultured human cells (Regan et al. 1976). Although,  there is no available
evidence regarding the ability of nitrosocarbaryl to  reach or affect the
mammalian gonads, nitrosocarbaryl is  intrinsically mutagenic and thus likely
to cause somatic mutations in humans  which may be involved in the  etiology of
cancer.  Some studies indicate that nitrosocarbaryl  given orally to  rats
results in the development of carcinomas in  the forestomach (Eisenbrand et al.
1976, Lijinsky and Taylor  1976).   Because the in vivo formation of this
compound in the human stomach from nitrites  and carbaryl  in the diet is
conceivable (Rickard 1979), nitrosocarbaryl  should be included  as  a  potential
parameter in the toxicological consideration of carbaryl.

(Studies summarized in Tables 1 and 2)
    Gene mutations are defined as point mutations and interstitial deletions.
Point mutations are limited to intralocus changes that affect one  or a few
base-pairs out of about 10^ to 10$ base-pairs which constitute  a gene
(e.g., base-pair substitutions, base-pair additions or deletions).   An
interstitial deletion is a minor deletion which may involve several  gene loci,
a single gene locus, or a  part of a gene locus (Flamm 1977).
    Before discussing the  results of the mutagenicity tests, it should be

emphasized that a negative result in a mutation test does not preclude the
mutagenic activity of a compound.  The specificity of a test system for the
detection of a particular type of mutagen and the level of sensitivity should
be considered.  Furthermore, the original  form of a chemical  may  not be
mutagenic but rather one of its metabolites may be an active mutagen.   Thus, a
compound inactive in the absence of metabolic activation in an in vitro test
should also be assessed in the presence of mammalian microsomal  enzymes (e.g.,
S9 fraction).  More importantly, the experimental assay may have  been
conducted improperly.  For example, concurrent negative and positive controls
must be included.  In addition, before investigating the mutagenic effect of
any agent, it is essential to determine the quantitative dependence of cell
killing of the chemical dose to be tested.  Not only should an appropriate
concentration range be examined but an adequate number of tests should be
performed to confirm a negative or positive result.
    Bacterial tests in which carbaryl has been reported as negative include
back mutations in Bacillus subtil is, forward mutations in Haemophilus
influenzae, and back and forward mutations in Escherichia coli (Table 1).
Several of these reports (DeGiovanni-Donnelly et al. 1968, Egert and Greim
1976, Fahrig 1974, Fiscor and Lo Piccolo 1972, Shirasu et al. 1976, Uchiyama
et al. 1975) contained no or only a brief description of the protocol.  Most
of these negative reports do not describe the toxicity of the test material
(see Table 1).  The importance of toxicity measurements is that excessive cell
killing by the chemical could result in the elimination of revertants or
mutants (i.e., false negative result) or that an observation of a negative
result and no toxicity may be due to the fact that the test material did not
enter the cell (i.e., a no-test classification).  In addition, because it has

                                                            TABLE I.  POINT MUTATION TESTS USING BACTERIA
Ashwood- Smith
et al. 1972
et al. 1977
et al. 1977
et al. 1978
et al. 1974
Test System
to tryptophan
prototrophy In
E. coll;
spot test
test: plate
and spot test
test: plate
test: plate
Forward mutation
to novoblocln
in Haemophllus

TA 1535
TA 1537
TA 1538
TA 98
TA 100
TA 100
TA 1535
TA 1537
TA 1538
TA 98
TA 100
Wild- type
Induced rats,
Dawley) S9
liver mix
Tested at a con-
centration of
10% (active In-
gredient) In
phosphate buffer
50 nmole
or 11.5 ug/
plate, dissolved
in ethanol
95% purity)
0.2. 2, 20.
400 ug/plate
10 ug to 1500
dissolved In
(96-99* purity)
0.1 mM
for 1 to 9
Reported as
Reported as
at 400ug/
Reported as
negative with
or without

Revertant count data not
Concentration of carbaryl Is
above solubility In water
(0.01%, Carpenter et al . 1976)
Purity of carbaryl not given
Toxicity not described
Revertant count data not
Brief description of protocol
Exact criteria for defining
a negative not given
One concentration examined
No concurrent positive
control data
Abstract without data
Data obtained by personal
Protocol was not available
Data not presented
Did not examine a range of
Purity of chemical not
Exposed to chemical for a
short time
                                                                                                                   (Continued on toilowing  pageT

                                                                   TABLE  1.   (continued)
Egert and
Egert and
Test System
to prototrophy In
E. coll:
TTquId suspension
liquid suspension
Back mutation to
prototrophy In two
TA 1538

System Concentration
House (male NMRI) Not reported
liver mlcrosomes
and NADPII cofac-
Mouse (male 100 uM
NMRI) liver (solvent not
mlcrosomes reported)
and HAOPH.
mice were

Resul t
Reported as
negative with or
without metabolic
In mutation
frequency after
(response was weak)
Reported as

Insufficient Information on
protocol employed and
revertant count data
not presented
Purity of chemical not given
Chemicals tested were
reported not to reduce cell
survival by more than 20%
Positive results were not
confirmed by repeating
Dose-related response not
Purity of chemical not
Report was a review of
published and unpublished
                             auxotrophlc  strains
                             of Serratia
                             marcescens,  and
                             forward  mutation
                             to galactose
                             prototrophy  In
                             E^ coll:  spot test
Protocol and data were not
Purity of carbaryl not given
Toxlclty not described
                                                                                                                        (Continued on following page)


Test System Strain System
Forward mutation to
1. (continued)
Concentration Result
Reported as

1. Report was a review of
published and unpublished
resistance In E.  coll:
Liquid holding test
Protocol and data were not

et al.

Flscor and
Lo Piccolo

et al.

Back mutation
at Indole
locus of
subtil fs

to prototrophy
in E. coll:

plate Incor-



TA 1535
TA 1536
TA 1537
TA 1538



Induced rats
(Male Sprague
Dawley) S9
liver prepa-

0.07% solution

Isotax Insect
with 51
carbaryl , exact
used not reported
50 ug to
1000 ug/plate.
dissolved In

Reported as

Reported as

Reported as
negative with
or without






Purity of carbaryl not given
Toxidty not described
Article was brief on
No concurrent positive controls
Purity of carbaryl not given
Concentration test of permitted
50% growth of control growth
Protocol and data not presented
Toxlcity not described

Data presented only for 1000
ug/plate where a twofold
Increase above background
was observed
Positive control data not
Purity of chemical not given
                                                                                          (Continued on following pageT

                                                                 TABLE 1.  (continued)
et al.
1980 (personal
Test System
plate incorpora-
plate Incorporation
TA 1535
TA 1537
TA 100
TA 98
TA 1535
Aroclor 1254
Induced rats.
S9 liver nix
up to
"purest grade
Analytical grade:
10. 100 ug/plate
1000. 2000.
3000 ug/plate
Experiment 11/28/77-
10. 20. 50, 100.
250, 500. 1000
Reported as 1 .
negative with or
without actlva- 2.
Reported as 1.
weakly positive
Mass screening of 300
Data only presented for 2000
Dose-related increase
et al.

to tryptophan
prototrophy in
E. cpli:
spot test

WP2 uvrA Hone Sevin 85 exact Reported as
WP2 reported concentration used negative
not presented.
report only Indi-
cates that either
1 to 3 mg crystal
or 20-25 ul were
added to each disk
for the various
chemicals tested

Revertant data not presented
Purity of chemical not given
Toxicity not described

                                                                                                                     (Continued on following page)



et al.

et al.

et al.

Test System
plate Incor-

plate Incor-

to tryptophan
pro to trophy In
E. coll:
spot test

to tryptophan
prototrophy In
E. coll:
spot test

TA 1535
TA 1537
TA 1538
TA 100
TA 98

TA 1535
TA 1536
TA 1537
TA 1538

UP 2 uvrA


Aroclor 1254
Induced rats.
(male Sprague
Dawley) S9
liver prepa-




Analytical grade
5. 25, 25. 325.
625 ug/plate
without activation
5. 10. 50.
250. 1250 ug/plate
with activation
0.02 ml of
1 ing/ml solution
dissolve in DMSO

0.02 ml of
1 mg/ml solution

Up to
10 rag/plate
did not report
exact concentra-
tions used or

In strain
TA 1535
In the absence
of metabolic
Reported as

Reported as

Reported as











Data not presented
Insufficient Information
on method
Purity of chemical not given
Toxicity of chemical not
No concurrent positive
Revertant count data not
Insufficient information
on method
Purity of chemical not given
Toxicity of chemical not
No concurrent positive
Brief description of protocol
and revertant count not given
Purity of chemical not given
Toxicity not described

not been clearly established that carbaryl  does not require metabolic
  *     *
activation to produce a mutagenic effect, the incorporation of metabolic
activation by mammalian microsomal  enzymes should be considered in the
experimental  design.  As shown in Table 1,  the majority of these negative
tests did not use activation by an in vitro microsomal  fraction.
    Another point of concern about these negative tests is that the majority
of them were spot tests {see Table 1).  Relatively water insoluble chemicals
are not easily detected as mutagens in a spot test assay (Ames et al.  1979).
In this method, a dose of the chemical to be tested is  placed on a disk of
filter paper.  The chemical then diffuses through the agar medium forming a
concentration gradient.  Because carbaryl has a low solubility in water
(0.01%, Carpenter et al. 1961), it would diffuse poorly through the agar
medium.  An adequate concentration range may not be tested or the chemical may
never leave the area of the disk, thus giving a false negative response.
Also, in a spot test the sensitivity is lowered because less bacteria on  the
plate are exposed to the mutagenic agent at any particular dose in the
resultant concentration gradient.  Table 1  describes the problems with each
test reported as negative.  The negative results reported for the various
bacterial strains do not provide substantial evidence of nonmutagenicity  and
do not reduce the weight of the positive results reported below.
    Carbaryl  mutagenicity was found only in the Salmonella assay using either
plate incorporation testing (Rashid 1978, Cook et al. 1977, McCann 1980,
personal communication) or liquid suspension testing (Egert and Greim 1976).
In plate incorporation, the mutagen is mixed with the bacteria tester strain,
the S9 mix, and the melted top agar and then poured on  the plate.  Generally,
a range of concentrations are tested in a series of plates.  This method
allows a large population of cells to be exposed to a uniform concentration  of

mutagen thus Increasing the probability of detecting weak mutagenic activity.
The liquid suspension method is similar to the plate incorporation assay
except that the chemical to be tested is preincubated with the activation mix
(S9), and the bacteria.  Afterwards, the top agar is added and the mixture is
poured onto plates.
    In a study in which McCann et al. (1975) examined 300 chemicals,
carbaryl  was classified as nonmutagenic in the plate incorporation test using
four Salmonella typhimurium histidine-requiring strains (Table 1).  Metabolic
activation did not augment the mutation frequency.  A conversation with Dr.  J.
McCann (May 1980) revealed that subsequent more through testing of carbaryl
has indicated that it appears to possess weak mutagenicity.  In three
independent experiments using TA 1535 (base-pair substitution sensitive), they
found that carbaryl appeared as weakly mutagenic (twofold or less increase in
background revertant counts) in the absence of metabolic activation.   In one
experiment two replicates of eight doses were examined (0 to 1000 ug/plate)
and carbaryl exhibited a dose-response effect.
    In general, a positive response in the Ames test is broadly defined as a
reproducible, dose-related increase in the number of histidine-independent
colonies (de Serres and Shelby 1979) but the precise criteria for the
designation of a weak response have not been established.  Presently, McCann
and co-workers in collaboration with Malcolm Pike's group at the University  of
Southern California are developing a statistical method for analyzing
dose-response data generated by chemical mutagens.  When McCann and co-workers
used four different models for line-fitting to determine the significance of
the above dose-response data of carbaryl, they found that two of the models
showed a significant (P < 0.01) dose-response (slope was estimated at about  3
revertant colonies/100 ug increase in dose per plate); the other two models

were rejected on the basis of a goodness-of-fit test.  When this determination
  *     *
was carried out, excluding the two data points at 1000 ug/plate dose, on the
grounds that they were artifically low due to toxicity of carbaryl at that
dose, all four models were accepted by the goodness-of-fit test and a
significant (P < 0.002) dose-response (slope was estimated to be about 7
revertant colonies per 100 ug) was shown.  It should be noted that these
statistical procedures are still under development, but McCann stated in a
letter (1980) that any further refinements are not likely to significantly
change the outcome of the basic result of carbaryl.  McCann also noted that
because the mutagenic response was weak, one should consider the possibility
of an impurity with mutagenic activity.
    The mutagenicity of carbaryl was also evaluated by Rashid (1978) employing
five strains of Salmonella typhimurium (Table 1).  Rashid found that carbaryl
produced a weak response in only strain TA 1535 at 125 ug/plate {1.6-fold
increase above the spontaneous mutation frequency) and at 625 ug/plate
(1.9-fold increase above the spontaneous mutation frequency) in the absence of
metabolic activation.  Carbaryl in the presence of rat liver (S9) microsomal
enzymes did not increase the spontaneous reversion frequency.
    Cook and co-workers (1977) tested carbaryl at a concentration range of 0.2
to 400 ug/plate without metabolic activation using the base-pair substitution
sensitive strain TA 100.  A weak mutagenic response (twofold response over
spontaneous mutation frequency) was found only at 400 ug/plate (Cook 1980,
personal  communication).  The data could not be evaluated because the protocol
was not available.
    Egert and Greim (1976) investigated the mutagenic activity of carbaryl
using the Salmonella typhimurium strain TA 1538 (sensitive to frameshift
mutagens).  These authors used a liquid suspension assay and treated bacterial

ce,11s with 100 uM carbaryl (100% survival rate).  Carbaryl was found to be
slightly mutagenic without metabolic activation.  The spontaneous mutation
frequency of this strain was very low (< 1 revertant/lO^ cells) and the
Induced frequency without metabolic activation was 7 revertants/lO^ cells.
After prelncubatlon with phenobarbital-Induced mouse liver nricrosomes and
NADPH (Instead of the rat S9 fraction), however, an Increase 1n mutagenic
activity (80 revertants/lO** cells) was seen.  The authors did not repeat the
assay to confirm this positive response and several concentrations of carbaryl
were not tested to demonstrate a dose-related effect.
    In contrast, some investigators have reported negative responses with
carbaryl using the Salmonella/Ames test (DeLorenzo et al. 1978, Blevins et al.
1977, Marshall et al. 1976).  These negative results could be explained by
solubility problems, or by cytotoxicity.  Also, because the classification of
a result as weakly positive or negative is somewhat arbitrary, the reported
negative results may be explained by the criteria used to classify a result as
negative.  The reported studies of DeLorenzo et al. (1978) and Blevins et al.
(1977) did not state the criteria used to define a negative result and did not
provide the revertant count data to support statements of a negative response.
Marshall et al. (1976) reported a twofold increase over the spontaneous rate
at 1000 ug/plate as a negative result.  However, revertant count data for the
other concentrations tested were not provided and it was not stated if this
twofold increase observed at 1000 ug/plate was reproducible.

Mammalian Cells in Culture
    Ahmed and co-workers (1977a)  examined the ability of 10 uM carbaryl  (66%
survival) to generate drug-resistant mutants in V79 Chinese hamster cells
(Table 2).  After carbaryl treatment and selection for mutants with 1  mM
ouabain, a steroid compound which inhibits membrane Na+/K+ ATPase
activity, these authors observed an induced mutation frequency (15.3
mutants/10^ survivors) significantly higher than the spontaneous mutation
frequency (1.8 mutants/106 survivors).  The apparent ability of carbaryl  to
generate ouabain-resistant mutants would be consistent with its activity as a
base-pair substitution mutagen.   Although carbaryl was reported to be  weakly
mutagenic in this assay, several  inadequacies are apparent in this report
which reduce the weight of the positive result.  The authors stated that the
altered phenotype of five randomly picked clones was stably transmitted
through many successive generations of growth in the absence of ouabain.
Nevertheless, they did not present data to support this statement, i.e., the
growth measurements describing the level of drug resistance acquired by  these
induced clones or measurements demonstrating an altered Na+/K+ ATPase
activity.  In addition, observations were not shown to be reproducible because
data were derived from a single experiment.  Also, the authors did not conduct
experiments to show that the induced mutation frequency to ouabain resistance
increased proportionally with increasing carbaryl concentrations at a  fixed
treatment time.  Concurrent positive controls were not included in the study
design and the purity of carbaryl was not described.
    Hoque (1972) reported carbaryl to be mutagenic in Drosophila melanogaster
(Table 1).  The results of this study are highly questionable.  Hoque  exposed
wild-type females to various concentrations of carbaryl (1 ppm, 5 ppm, and 10

Ahmed et al.

Brzheskll 1972

Hoque 1972

Test System Activation System
Forward mutation None
to ouabaln resis-
tance In V79
Chinese hamster cells

Sex-linked recessive
lethal test in
Visible mutants In
In Drosophila

10 uM

Purity 85%
1% suspension
for 24 hours
1 ppm, 5 ppm.
10 ppm for
5 hours

Weakly mutagenlc

Reported as
weakly mutagenlc

Reported as





Experiments not performed to
demonstrate a dose-related
No concurrent positive
Old not demonstrate an alter
NZ+/K* ATPase activity
or growth measurement In the
presence of ouabatn
Purity of chemical not given
Evidence considered suggestive
Inadequate sample size
No concurrent positive control
Evidence considered suggestive
Article brief on methodology
Purity of chemical not given
Results are considered to be

ppm in the culture medium)  for 5 hours.   Hoque did not describe either the
toxicity effects or the sterilizing effects  of the three concentrations of
carbaryl used in this study.   In addition,  inordinately high  frequencies (in
the FI generation, 69.4*;  in  the F2 generation,  71.8*  at 1  ppm carbaryl)
of a visible variant phenotype (black eye) were observed.   Such high
frequencies are not characteristic of variation  due to genetic alterations,
and thus, in this case, phenotypic variation is most likely the result of some
developmental or physiological error.  In addition, Hoque described an
anatomical defect (head merged with thorax)  caused by  carbaryl  treatment, but
does not state the carbaryl  concentrations where this  anomaly is observed or
the frequency at which it occurs.  Other deficiencies  found in this study are:
1) insufficient sample size,  2) lack of positive control, and 3) lack  of
statistical analysis.
    Brzheskii (1972) used the Drosophila sex-linked recessive lethal  test to
screen for carbaryl mutagenicity (Table 2).   Drosophila melanogaster  males
were exposed to a 1% suspension of Sevin (85% carbaryl) in  dilute sugar syrup
for 24 hours.  This treatment resulted in 50% survival.  A  small increase in
the percentage of complete (heritable mutations) and partial  (not all
mutations transmitted to progeny) recessive  lethals (0.2% +_ 0.07%) compared to
negative control values (0.0% experimental  control, 0.05 +  0.02% in historical
negative controls) was observed.  If only the heritable lethals are
considered, the frequency observed (0.091%  +_ 0.04%) is not  significantly
different from the spontaneous mutation values.   A larger sample size would
have to be examined to conclusively classify carbaryl  as weakly positive or as
    In  summary, the available evidence concerning the ability of carbaryl to
cause gene mutations is considered suggestive.  Additional  tests are  needed

which describe a dose-response curve, for example using mammalian cells j_n_
vitro or Drosophila.  Furthermore, the dependence of carbaryl  on metabolic
activation has not been systematically explored.   Thus, it is  not clear
whether carbaryl  acts predominately as a direct mutagen or is  metabolized  to  a
genetically active chemical.

(Studies summarized in Table 3)
    Mutagenesis is not only characterized by gene mutations but also the  gain,
loss, or rearrangement of portions of chromosomes and the gain or loss  of
intact chromosomes.
    Amer (1965) used root tips from Alii urn cepa (onion) to examine the  ability
of carbaryl to induce chromosome aberrations.  He found that both formulated
and "pure" (degree of purity not defined) carbaryl  prepared at room
temperature or at 60°C produced C-mitotic effects.   C-mitosis  is described as
an inactivation of the mitotic spindle proceeded  by a random distribution  of
chromosomes (Grant 1978).  When onion root tips were exposed for 4 hours  to 25
ppm (0.25 saturated solution) and 50 ppm carbaryl (0.5 saturated solution),
multipolar anaphases, chromosome lagging, "disturbancy of metaphases" (term
not clearly defined), and multinucleated cells were reported.   Continuous
treatment for 24 hours "nearly" arrested mitosis.  This effect, however,  was
temporary because the root tip cells recovered after a 48-hour replacement
period in water.
    Amer and co-workers (1971) found similar results of carbaryl (noted as
"pure" Sevin in the report, however degree of purity not defined) on mitosis
of Vicia faba and Gossypium barbadense seeds.  In this study,  the authors

                                                               TABLE  3.  CYTOGENETIC TESTS
             A. Plants
   Test System
             Amer 1965
Root tips Of All!urn
cepa (onion)
85% sprayable powder
(Union Carbide)
25 ppm and 50 ppro
for 4 hours
C-mltotlc effects;
multipolar anaphases,
chromosome lagging,
multlnucleated cells
             Amer et al,
Root tips of yicia
faba (broad bean)
and Gossyplum
barbadense (cotton)
Carbaryl described
as "pure" Sevln at
25, 50. and 100 ppm
for 4 hours for root
treatment (seed-soak
treatment was also
used; 100 ppm for 6,
12, 24. and 48 hours)
Abnormal mitosis
(reported as distur-
bancy of metaphase,
chromosome lagging.
tetraploid cells),
frequency of In-
duced "abnormalities"
Increased as treat-
ment time or concen-
tration was Increased
1. Frequency of different "abnormalities"
   not reported for control cells
2. Disturbed metaphase and anaphase  eported
   as predominant effect, this effect
   is defined as "chromosomes are spread
   Irregularly over all cells"
             Amer and
             Farah 1968
Heiotlc effects
studied In Vlcia
faba pollen mother
Two-week-old plants
sprayed weekly with
saturated solution
of Sevln for one
month, sprayed every
two weeks for one
month, or sprayed
daily up to 8 days
Abnormal melosis
(reported as "ana-
phase bridges.
stickiness, poly-
ploid. lagging,
1. Chromosome stickiness reported as common
   abnormality, fragmentation was one of the
   least common effects.
2. Treatment did not cause pollen sterility
3. Frequency of each abnormality not
   reported for control cells
                                                                                                                 (Continued on following page)

                                                        Table 3.  (Continued)
A. Plants
   Test System
Melotlc effects
studied In corn
Seeds treated wtth>
Sevln 50 at 0.12%
and 0.25% aqueous
solution for 48 hours,
0.25% solution was
Injected Into anthers
of plants and treat-
ment lasted for 6
Seed treatment:
dose-related Increase
In frequency of "chromtln"
bridges (with and
without fragments)
Anther treatment
chromosome stickiness
(primary effect).
C-mltotlc effects
1. Russian study
Wuu and
Grant 1966
Hltotlc effects
studied In Hordeum
yulgare (barleyl
pollen mother cells
Sevln (purchased
from Union Carbide),
seed treatment at 500.
1000, 1500 ppm for 6.
12. and 24 hours
Chromosome aberrations
(most common effect
reported was fragments,
other effects which
were less frequent
were chromosome lagging
late telophase or Inter-
phase bridges)
1. "mutant" seedlings reported for
   second generation
Huu and
Grant 1967
Melotlc effects
studied In Hordeum
vulgare (barleyj
pollen mother cells
Seeds treated with
Sevln 80 (purchased
from Union Carbide)
at 1000 ppm aqueous
solution for 12 hours,
plants sprayed with
500 ppm
Chromosome aberrations
1. Frequency of each aberration not
                                                                                                    TContlnued on following page)

                                                       Table 3.  (Continued)
B. Mammalian cells  In culture
Reference Test System
Ishldate and Chinese hamster flbro-
Odashlraa 1977 blast cells In vitro

Kazarnovskaya Human embryonic
and Vasllos flbroblasts In vitro
Vasllos et al. Human embryonic
1972 flbroblasts In vitro
(mg/ml )

(reported In

20, 40. and
80 ug/ml for 24
20. 40, and
80 ug/ml for 6.

Polyplo1d(%) Aberrant Cells!*)

48h 24h 48h
2 2 1
1 12 24
1 29 35
0 1+0.9 1+0.6

Chromosomal aberrations

"Pathological forms of
and antlmitotlc






Length of treatment time not
Purity of chemical not reported
Toxlclty of doses tested not
clearly defined
Frequency of each particular
aberration not reported
Data obtained by personal communi-
Russian study
Purity of carbaryl not reported

Russian study
Pathological forms of mitosis not
                                         24. and 48 hours.
                                         technical product
                                         reported as con-
                                         taining 84% active
                                           effect reported
                                                        clearly defined
                                                     3.  Purity  of  carbaryl  not reported
Shplrt 1975
Human embryonic
flbroblasts in vitro
0.001. 0.01, 100,
and 1000 mg/ml
Dose-dependent antlraltotic
effect reported
1. Russian study
2. Carbaryl purity not described
3. 1000 mg/ml resulted In 01
                                                                                                   TContlnued on foTlowing  page)

                                                       Table  3.   (Continued)
C. Rodents
   Test System
Degraeve et al.
Mlcronucleus test In
10-3 M intraperi-
toneally or dally for
one week via Intuba-
                                                                 Reported as nega-
                                                                                       1. Criteria for selecting dose level not
                                                                                          defined or toxiclty of dose used not
                                                                                       2. Purity of chemical not given
Epstein et al.
                 Dominant lethal  In
                 mice (ICR/Ha  Swiss)
                         Orally  for 5 consecu-
                         tive days  at 200 nig/
                         kg/day  and at 10 mg/
                        Reported as nega-
                                                                                       1. Data were not available for an evaluation
Vasllos et al.
                 Mitosis  studied  in
                 epithelium from  small
                 Intestine  crypts and
                 cornea of  rats
                         85% commercial  prepa-
                         ration.  acute:   400
                         mg/kg (one  half LD50)
                         80 mg/kg. 40 mg/kg,
                         and 20 mg/kg
                        Effects on mitotic
                        spindle (mitotic
                        arrest, C-mitosis.
                        and chromosome
                                                                                        1. Russian study
Vasllos et al.
Mitosis studied in
splenic follicles.
corneal epithelium.
and epithelium of
glandulae intesti-
nales of rats
85% commercial prepa-
ration, subacute: 5
and 20 mg/kg (28
orally chronic: O.OS
to 8 mg/kg/day for 6
months orally
Effects on mitotic
spindle (mitotic
arrest. C-mitosis,
chromosome fragmen-
1. Russian study
                                                                                                     "TCorvtlnued on  followlr.-j page)

                                                             Table 3.  (Continued)
      C. Rodents
   Test System
      Well et al.
Domtnant lethals In
rats (Harlan-VHstar)
3-generatlon repro-
duction study, car-
baryl was given In
diet at maximum dally
doses of 200 mg/kg
and dally oral doses
as high as 100 mg/kg
Reported as negative
I. Number of males treated not reported
2. Number of virgin females mated with each
   treated male not reported
3. Number of Implants and fetal deaths per
   female of test or control groups not

found t-hat the percentage of induced chromosome effects  (defined  as  abnormal
metaphases and anaphases) are dependent on  the  carbaryl  concentration  (25 ppm,
50 ppm, 100 ppm)  and treatment time (4, 6,  12,  24,  and 48  hours).
    Amer and Farah (1968) studied the effects of carbaryl  on meiosis of Vicia
faba.  It was found that daily spraying with 0.01%  (saturated  solution) of
carbaryl on plants induced chromosome aberrations (chromosome  stickiness,
lagging, fragmentation, anaphase bridges,  univalents  in  diakinesis,  and
multinucleated cells).   Carbaryl  treatment  did  not  cause pollen sterility.
    Similarly, Wuu and  Grant (1966) induced chromosome effects in  Hordeum
vulgare (barley)  with 500 ppm, 1000 ppm, and 1500 ppm of carbaryl.   Later, Wuu
and Grant (1967)  found  that carbaryl (80%)  also affected meiosis  in  the mother
pollen cells of barley  when seeds were treated  (1000  ppm for 12 hours) or when
barley plants were sprayed with a aqueous solution  of carbaryl (500  ppm).
Meiotic effects were also reported by Brankovan (1972) in  corn after treatment
of the embryonic and generative phases of development with seed treatment of
0.12% and 0.25% solutions of Sevin 50 (50%  carbaryl)  for 48 hours  and  0.25%
injected into anthers for 6 hours.  Tables  of data  were  not provided in the
translation of this report.
    The aforementioned  studies in plants indicate that carbaryl is capable of
breaking chromosomes; however, it predominantly causes mitotic disturbances by
interfering with the spindle mechanism in plants.  This  may result in
chromosome segregation  errors (nondisjunction).
    Although the events measured in plants  are  similar to  those in mammalian
cells, their relevance to humans is sometimes questioned.   It  should be noted
that when one examines the literature an excellent correlation is seen between
the mutagenic effects demonstrated in plants and those in  mammalian  cells in
culture (Flamm 1977).

Mammalian Cells in Culture
    There is suggestive evidence that carbaryl  may act as an antimitotic agent
in cultured mammalian cells.  The evidence comes from studies conducted in
Russia.  The conclusions presented in these reports are difficult to interpret
because these reports only briefly describe the methodology and results, and
the data expression is somewhat unconventional.  Both Vasilos et al. (1972)
and Shpirt (1975) found that carbaryl inhibited the mitotic activity of human
embryonic fibroblasts in vitro.  Shpirt (1975)  reported that this antimitotic
effect was both dependent on the concentration  of carbaryl  and duration of
exposure.  The most pronounced effects were observed at toxic levels of
carbaryl.  Vasilos et al.  (1972) also reported  an increase in "pathological
forms of mitosis" of human embryonic fibroblast in vitro caused by 20 ug/ml,
40 ug/ml, and 80 ug/ml of carbaryl (reported as 84% active material) for 6,
24, and 48 hours, but do not clearly define these forms and only indicate that
the predominant pathological form consisted of  C-mitosis.  Kazarnovskaya and
Vasilos (1977) reported chromosome effects in human embryonic fibroblast
cultures treated with 40 ug/ml and 80 ug/ml of  carbaryl for 24 hours.  The
appearance of these aberrations was dose-dependent.  Chromosome fragments were
observed at 40 ug/ml, a dose which is one-half  of the dose (80 ug/ml) that
results in 50* growth inhibition.  Chromosome and chromatid exchanges were
only reported at 80 ug/ml.  Ring chromosomes and inversions were not reported.
    Ishidate and Odashima (1977 and personal communication 1980) studied
effects of carbaryl on chromosomes of cultured  Chinese hamster fibroblasts.
Three different doses (0.0075, 0.015, and 0.03  mg/ml) were added to cell
cultures and observations were made 24 hours and 48 hours after treatment.  At
the maximum effective dose, 0.03 mg/ml, several types of chromosome effects
(35% aberrant cells:  chromatid gaps and breaks, chromosome breaks,

translocation, ring formation,  fragmentation)  were observed  48  hours  after
treatment that were higher than in nontreated  control  cultures  (1%  aberrant
cells).  The report does not indicate if the high, medium, or low dose  is the
50% growth inhibition dose and  does not define the toxicity  of  the  maximum
effective dose.  At lower doses, 0.015 mg/ml  resulted  in  24% aberrant cells
and 0.0075 mg/ml  did not appreciably affect chromosome structure  (1%  aberrant
cells) 48 hours after treatment.  The toxicity of the  lower  doses was not
given.  Although the authors stated that "gaps" were the  predominant
chromosomal effect, the occurrence for each particular type  of  aberration, or
the frequency of aberrations within a cell  were not given.   It  should be noted
that gaps and breaks may be the result of cytotoxicity.   A small  number of
cells (100) were examined in this study, thus  lowering the sensitivity  of the
assay.  The purity of carbaryl  was not given in this report.
In Vivo Cytogenetic Studies
    In a Russian report, Vasilos et al. (1975a) administered orally a single
dose (400 mg/kg, 0.5 of LDSQ) of commercial carbaryl (85%) to rats.  The
mitotic activity of cells in both the cornea and the epithelium of  the  small
intestine were not affected at this dose.  However, "pathological mitosis"
(e.g., C-mitosis, fragmentation, chromosome lagging, bridges, micronuclei)
were observed in the intestinal epithelium 12  to 72 hours after treatment.
C-mitosis was the most pronounced effect.  Significant effects  were not found
in the cornea cells.  At lower doses, 20 mg/kg (single injection) did not
affect chromosome morphology, but 40 and 80 mg/kg (single injection)  caused  a
1.6- and twofold increase in "pathological" forms of mitosis in intestinal
epithelial cells.
    Vasilos et al. (1975b) continued in vivo cytogenetic  studies  on rats.
They found altered mitosis in intestinal epithelial cells (e.g.,  C-mitosis,

mi-cronuclei, "degenerative forms," chromatid and chromosomal  bridges,
chromosome lagging, and fragmentation) after chronic low doses of 85%  carbaryl
(1 to 8 mg/kg) were given orally to rats daily for 6 months.
    In contrast, Degraeve et al. (1976) did not report adverse chromosome
effects in bone marrow cells or testicular cells of mice 24 hours and  48 hours
after 10~3 M carbaryl (reported to titrate at about 90% active principle)
was administered intraperitoneally or daily for one week by intubation.   An
increase (approximately twofold) in the incidence of diakinesis and metaphases
with monovalent X and Y chromosomes in testicular cells was reported.
However, the results of this study are difficult to evaluate  because the
report does not provide adequate information concerning the design and conduct
of testing.  Thus, no information was provided on the number  of animals
treated, and data are not presented concerning a determination of the  toxicity
of the test substance to mice after exposure to arrive at a maximum tolerated
dose for conducting the in vivo cytogenetic assay.
Dominant Lethal Assay in Rodents
    The dominant lethal assay detects chromosome damage in germ cells.  Using
male Swiss mice, Epstein et al. (1972) administered 1000 mg/kg and 50  mg/kg of
carbaryl (reported as subtoxic) by gavage in daily portions over 5 consecutive
days and reported that this dosage schedule did not produce significant  early
fetal deaths or preimplementation losses.  Data were not presented to  support
this statement.
    Weil and co-workers (1973) looked for dominant lethality  in rats using a
three generation study and found no significant lethal effects.  This  article
is unclear on  several points.  The number of males treated, the number of
virgin untreated females mated with each treated male, and the number  of
implants and fetal deaths per female of test or control groups are not given

i.n this report.
    It should be noted that,  in general,  the dominant lethal  assay  in  rodents
is recognized as an insensitive test for  the detection of weak  mutagens
because of the small  number of animals used in such a study and the high
background of fetal wastage observed in control  animals.   In addition,
chromosomal  effects are usually detected  at higher chemical  doses than are
gene mutations.
    In summary,  the available evidence that carbaryl  causes chromosome
aberrations is merely suggestive.   However, studies conducted in plants and
studies conducted in Russian using mammalian cells in vitro and in  vivo
strongly suggest the ability of carbaryl  to act as a spindle inhibitor.   It
should be noted that spindle effects were primarily reported in studies using
commercial grades of carbaryl, thus the possibility of a  genetically active
contaminant(s) should not be ignored.

(Studies summarized in Table 4)
    The reports discussed in this  section describe tests  which  detect  the
ability of a chemical substance to cause  primary DNA damage (for example, as
manifested by DNA repair) but do not provide a measurement of mutation per  se
(Flamm 1977).
    Unscheduled DNA synthesis is indicative of primary DNA damage and
subsequent DNA repair.  Ahmed et al. (1977b) have shown that exposure  of
virally transformed human cells (VA-4) in culture to carbaryl initiates
unscheduled DNA synthesis at exposures as low as 1 uM as  determined by
photolysis of bromodeoxyuridine (BrdUrd), which is incorporated into DNA
during DNA repair synthesis.  These authors found that the size of  the repair

                                    TABLE 4.  OTHER GENETIC EFFECTS OF CARBARYL
                                    (Evidence Indicative of primary DNA damage)
Ahmed et al .

Regan et al .

Siebert and

Test System
DNA synthesis
as detected by
graphy and
uridine phot-
DNA strand
breaks as
by sedimen-
tation pro-
Gene con-
version at
the loci ade-2
and try- 5
in Saccharomyces
Strain Activation System
SV-40 Liver extracts
transform prepared from
human rats/S9 mix
line VA-4

Human skin
in vitro


Concentration Result
BrdUrd Assay- Positive
1 to 100 uM
for 24 hours
1 to 1000 uM for
8 hours

100 uM for Negative
1 hour

1000 ppm (4.97mM) Negative
dissolved in DMSO
for 16 hours,
purity 99.9%.
1. Metabolic activated
did not increase
genetic activity
2. Purity of chemical
not given
3. Toxicity of doses
used not described

1. Purity of carbaryl
not given
2. Toxicity of doses
used not described

1. Survival at dose
tested was 78%


region 'could be classified as UV-type repair.   In  addition,  when  the  carbaryl
concentration (1 uM, 10 uM, 100 uM for 24 hours) was  increased, there was  a
concomitant increase in BrdUrd incorporation.   The standard  deviation or error
was not given for the BrdUrd photolysis data.   Ahmed  and  associates also used
autoradiography after carbaryl  treatment (1  uM, 10 uM,  100 uM,  1000 uM, for 8
hours) of human VA-4 cells as a method to detect unscheduled DNA  synthesis.
Positive results were obtained from this assay thus confirming  the  results
derived from the BrdUrd photolysis assay. These authors  did not  present the
grain numbers for all of the carbaryl concentrations  examined except  at 100
uM.  Therefore, it is not known if a dose-related  effect  was found with these
concentrations.  Metabolic activation (liver extracts prepared  from rats)  of
carbaryl did not enhance its ability to induce unscheduled DNA  synthesis.  In
this study, neither the cytotoxicity of carbaryl for  the  concentrations
employed nor the purity of carbaryl was given.
    When Regan et al. (1976) treated a culture of  human skin cells with 100 uM
of carbaryl for 1 hour, no evidence of primary DNA damage was detected, but
the cell lines used, the technique employed, and the  endpoint being measured
(DNA strand breaks) were not the same as that employed by Ahmed et al.
(1977b).  Regan determined the sedimentation profiles in  alkaline sucrose
gradients of cellular DNA treated with carbaryl as a  method  for the detection
of primary DNA damage.
    Another system which is indicative of primary  damage  to  DNA includes the
ability of a test agent to induce mitotic gene conversion (intragenic
recombination) in yeast cells.  Siebert and  Eisenbrand (1974) used  a  diploid
strain of Saccharomyces cerevisiae heteroallelic at the gene loci ade-2 and
try-5 to test for the ability of carbaryl to induce mitotic  gene  conversion in
these loci.  In this organism, genetic activity or genetic  damage was not

produced by a 16-hour carbaryl  (1000 ppm,  99.9% purity)  treatment.   Yeast
cells cultured in this solution of carbaryl  showed only  a 22% lethality.   The
low amount of toxicity caused by treatment with 1000 ppm of carbaryl  may
indicate that treatment conditions were not sufficient to induce mitotic  gene
conversion to a detectable level.
    In summary, the results of Ahmed et al.  (1977b)  indicate the ability  of
carbaryl to interact (directly or indirectly)  with DNA and the negative
results reported by Regan et al. (1976) and Siebert and  Eisenbrand  (1974)  are
considered not to contradict this conclusion.

    In order for any mutagen to cause genetic  alterations that may  be
inherited by future generations, it must reach the gonads in an active  form.
As articulated in the Agency's "Proposed Guidelines for  Mutagenicity Risk
Assessments" (Environmental Protection Agency  1980), evidence that  a chemical
reaches the gonads is provided by data demonstrating the alkylation of  DNA or
other cellular molecules, unscheduled DNA synthesis, sister chromatid
exchange, or chromosome aberrations in germinal cells, and non-specific
accumulation of radioactive label  in the gonads following administration  of
the labeled chemical.  When there is a lack of such data, other relevant
evidence includes adverse gonadal  effects following acute, subchronic,  or
chronic toxicity testing; and adverse reproductive effects such as  decreased
fertilization index, reduced sperm count, or abnormal sperm morphology.
    The sperm-abnormality assay is an indicator that a chemical agent may  be
damaging the germ cells (Wyrobek and Bruce 1978).  Wyrobek and associates
(1980) analyzed semen samples from 50 carbaryl production workers and 34  new

hires who served as controls.  Males exposed to carbaryl  had sperm counts or
sperm with" fluorescent bodies (thought to be caused by meiotic
nondisjunction) similar to control values.  The authors also reported a higher
incidence of oligospermic men (less than 20 million sperm/ml) in the exposed
group (14.6%) than in the control group (5.9%).  However, this difference was
not statistically significant.  There was a significant elevation (P < 0.005)
of sperm abnormalities (52.0 ^ 2.6%) with abnormal  head morphology in
currently exposed workers compared to the controls (41.9 ^2.1).  A one-tail
statistical  analysis (P < 0.05)  revealed that previously exposed workers (an
average of 6.3 years since last carbaryl exposure)  exhibited a significant
elevation of sperm abnormalities from control values (Wyrobek 1980, personal
communication).  Because of the sample size, however, it cannot be
conclusively established if carbaryl effects are permanent.  When current
workers were classified as low (supervisors, foremen, maintenance personnel)
and high (baggers, operators) exposure groups, both groups were shown to have
significant differences in sperm abnormalities from the control  group.   But,
there were no appreciable differences between the high and low exposure
groups.  In addition, there was a negative correlation between the number of
years the current workers had been exposed to carbaryl  and the percentage of
abnormal sperm observed.  Wyrobek and co-workers speculated on several
mechanisms to explain this odd relationship: 1) men working longer may be
exposed less because of seniority, 2) biological or pharmacological
adaptation to exposure, and 3) selection for non-affected males.  Although
this study demonstrates a correlation between working in a carbaryl-exposed
area and an alteration of human spermatozoa, it only provides suggestive
evidence that these effects are due to carbaryl and/or its metabolites.  It
must be established that these defects of sperm morphology are not the result

of other factors, such as exposure to chemicals other than carbaryl.1
    Whorton et al. (in press) in an earlier study, examined semen samples
provided by 47 former and current v/orkers at the same plant studied by Wyrobek
et al. (1980).  Similar to the findings of Wyrobek et al.  (1980), the  cohort
with known exposure to carbaryl  were found to have a higher incidence  (14.9%)
of low sperm counts (oligospermia) than a control  group (5.5%),  but this
difference was not considered to be statistically  significant (P = 0.686).
Blood levels of follicle-stimulating hormone, luteinizing  hormone, and
testosterone were not different from control values.  It should be noted  that
the control group was not composed of onsite controls but  was composed of
historical controls (workers from a composite control population with  no known
exposure to infertility-producing agents) collected by the Environmental
Health Associates during previous chemical industry studies.   These data  are
considered to be suggestive of a correlation between oligospermia and  carbaryl
    In a study performed in Russia, Krylova and Denisova (1973)  examined the
process of spermatogenesis in the Mongolian Tree Creeper,  a small rodent.
These animals inhabit an area that was sprayed with 85% Sevin wettable powder
produced in America (flow rate,  2 to 2.5 kg/ha over 60 ha).  One year  after
treatment, 9 out of 22 of the exposed animals were found to have pathological
changes in the spermatozoa from the epididymis (spirally twisted tails, breaks
in the neck and tail, isolated heads and tails).  There was a statistically
significant (P < 0.001) reduction in the number of spermatozoa,  spermatids,
and spermatocytes from the control group (tree creepers residing in an area
that was not sprayed).  Chemical analysis showed that 20 of the 22 exposed
animals contained extremely high quantities of carbaryl in their genitalia
(reported  to be 0.5 to 1.5 mg/kg of body weight).   The reproductive capacity

      l$ee  page 43 for an addendum to this study.

of. the animals living in the carbaryl-exposed area appear reduced with respect
to the animals that resided in an area that was not sprayed.   In this article,
the results were not fully interpretable because the description of the
protocol was brief.  Also, interpretation of results is tenuous because
several factors, e.g., genetic variability of animals, duration and amount of
exposure to the chemical, and the age and health of the animals, cannot be
controlled in such a study.
    Kitagawa et al. (1977) reported a reduction in the number of spermatogonia
and spermatozoa in the testes after rats (Wister) were orally administered 3
mg of carbaryl per rat per week for one year.  Only photographs of
histological slides were provided in the report and not quantitative data to
confirm results.  Although the authors reported that the total  dose used was
equivalent to an acute dose representing the LDSQ, no information concerning
toxicity or the method used to determine the chronic dose level was provided.
The purity of carbaryl was not reported.
    Degraeve et al. (1976) reported an increase in the incidence of abnormal
spermatozoa (no acrosome, abnormalities of flagellae) in the  ductus deferens
in carbaryl-treated male mice (10~^ M administered per dose daily for one
week or single intraperitoneal injection).  These authors, however, did not
demonstrate whether exposed mice showed dose-dependent increases in the
induction of abnormal sperm.  A description of the toxicity of the dosage
level used was not provided in the report.  Carbaryl was described as
commercial Sevin titrating at about 99* active principle.
    Thomas et al. (1974) administered a single dose of radioactive [14C]
carbaryl (24 uCi/kg, 0.9 mg/kg) to Swiss-Webster albino mice.  They found very
small amounts of labeled carbaryl and/or its metabolites in the prostate,
seminal vesicle, testes, seminal plasma, and epididymal fat.   The radioactive

CQunts Vere so low in this study, it is doubtful  that these counts are
significantly different from background counts.
    In a Russian study, Shtenberg and Rybakova (1968) found a decrease in
sperm motility in albino rats at 14 and 70 mg/kg/day after 6, 9,  and 12 months
of carbaryl (reported as 100% active material) treatment.   This effect on male
fertility was dependent on the carbaryl concentration and  duration of
exposure.  The most pronounced effect (P < 0.001)  was observed at 70 mg/kg/day
after 12 months.  Significant effects (P < 0.001)  were also observed at the
lower doses after 12 months of treatment.  The authors reported histological
changes, e.g., oedema of interstitial tissue,  reduction in the number of
spermatocytes and spermatids, and destruction  of  germinal  epithelium in rats
exposed to 70 mg/kg/day for one year.  Slight  interstitial  oedema and some
depression of spermatogenesis were reported in rats exposed to 7  mg/kg/day for
1 year.
    Collins et al. (1971), using a three-generation reproduction study,
administered orally to female rats (Osborne-Mendel) and gerbils 10,000 ppm of
carbaryl (technical grade, 99% purity).  Impaired fertility (P <  0.05) was
reported.  In addition, no litters were produced  from the  second mating of the
second generation of rats and the third generation of gerbils.  It should be
noted that growth depression was observed at 10,000 ppm.
    In contrast, some investigators have reported no significant gonadal
effects attributable to carbaryl.  Dikshith et al.  (1976)  administered
technical  (99% purity) carbaryl (200 mg/kg for 3  days a week) orally to male
albino rats for a period of 90 days.  No histological changes were reported in
the testes and epididymis.  Also, these authors found no effects on the
fertility of male rats.  This study did not involve a quantitative assessment
of the effects on sperm (e.g., sperm morphology,  sperm counts).  Weil  et al.

(1972) observed no significant effects of carbaryl  (10 mg/kg/day)  on female
fertility in a three-generation rat study.  However,  a dose of 100 mg/kg/day
by intubation, which resulted in mortality, reduced fertility.
    In summary, there is no direct evidence, e.g.,  the presence of a
radioactive label  of the chemical in the gonads, that carbaryl and/or its
metabolite(s) reaches mammalian gonads.  However, there are many reports
indicating adverse gonadal and reproductive effects after carbaryl exposure.
Therefore, the weight of evidence suggests that the active form of carbaryl
may reach the germinal tissue of mammals.

    In in vitro experiments, carbaryl has been shown to react with sodium
nitrite under acidic conditions (pH 1) to form nitrosocarbaryl (Eisenbrand et
al. 1974).  Because nitrite is present in human saliva and food products, the
formation of nitrosocarbaryl in stomach physiology is conceivable due to the
widespread use of carbaryl.  Rickard (1979) demonstrated the in vivo formation
of nitrosocarbaryl in the stomach of rats and guinea pigs.  When guinea pigs
were given either simultaneous intubation of carbaryl (1 umol) and sodium
nitrite (1160 umol) or when these components were mixed with feed,
approximately a 1.5% yield of nitrosocarbaryl was detected.  The formation of
this nitroso derivative was dependent on the amount of nitrite and the pH, and
was not particularly influenced by the amount of carbaryl.  Increasing the
amount of carbaryl from 0.025 to 2.5 umol did not increase the yield of the
nitroso-compound.  In rats, where the stomach pH (3.5 to 5.5) is higher than
in guinea pigs (pH 1.5), a very low yield of nitrosocarbaryl was found (0.02%)
at the same concentrations of nitrite and carbaryl.

    Nitrosocarbaryl  has been shown to be strongly mutagenic in bacteria.
Blevins and associates (1977) found that the base-pair substitution sensitive
Salmonella strains TA 100 and TA 1535 were reverted by this agent without
metabolic activation.  The reversion frequency in TA 100 was increased by
approximately 1.6 fold at 1.15 ug/plate and sixfold at 11.5 ug/plate and  in TA
1535 by about threefold at 1.15 ug/plate and 76-fold at 11.5 ug/plate.
Nitrosocarbaryl was not as active on the frameshift sensitive strains TA  98,
TA 1537, and TA 1538.  Marshall et al. (1976) found that nitrosocarbaryl
increased the number of histidine-independent colonies of TA 1535 by
approximately sixfold at 0.5 ug/plate and by 367-fold at 50 ug/plate without
metabolic activation.  These authors also found nitrosocarbaryl  to be slightly
active (above sixfold increase over background values) on the frameshift
sensitive strains TA 1537 and TA 1538 at 50 ug/plate.  Both Blevins et al.
(1977) and Marshall  et al. (1976) found that the mutagenic activity of
nitrosocarbaryl was dose-related.
    Elespuru and co-workers (1974) measured the induction to novobiocin
resistance in Haemophilus influenzae.  These authors found that
nitrosocarbaryl was approximately an order of magnitude more potent than  the
well-known mutagen N-methyl-N'-nitro-N-nitrosoguanidine (MNNG).   In
Escherichia coli, nitrosocarbaryl was also more potent in the induction to
arginine prototrophy than MNNG (Elespuru et al. 1974).  Uchiyama et al. (1975)
found mutagenic activity as tested by the ability to cause reversion at the
tryptophan locus in Escherichia coli.
    Generally, metabolic activation was not required for the mutagenic
response of nitrosocarbaryl.  For example, when Marshall et al.  (1976)
incorporated the S9 fraction in the Salmonella assay, a slight decrease in
mutagenic activity was observed.  Greim et al. (1977), however,  found an

ipcrease in mutagem'city after metabolic activation  by  mouse-liver microsomes.
    Siebert and Eisenbrand (1974)  reported that nitrosocarbaryl was  active  in
causing mitotic gene conversion in Saccharomyces cerevisiae.   Incubation  for 2
hours on 1 ppm of nitrosocarbaryl  increased the relative  conversion  frequency
threefold for the ade-2 and fivefold for the trp-5 locus,  and  at  30  ppm
increases were 139-fold for the ade-2 locus and 885-fold  for the  trp-5 locus.
In this study, a dose-related effect was shown using five concentrations  of
nitrosocarbaryl.  Regan et al. (1976) demonstrated that nitrosocarbaryl was
able to induce DMA damage in cultured human cells as measured  by  unscheduled
DMA synthesis.  In addition, by using methyl  labeled [14C]  and ring  labeled
C^H] nitrosocarbaryl, Regan et al. (1976)  found that the  C^4C] label was
associated with cellular DNA, whereas the [^H] label  was  not.   Because
nitrosocarbaryl has been observed to cause reversion of base-pair substitution
sensitive strains (TA 100, TA 1535), these results suggest that the
nitrosocarbaryl molecule was split and the resultant methyl  group alkylates
    Ishidate and Odashima (1977) reported several chromosome aberrations  (81%
aberrant cells) in Chinese hamster cells in vitro 24 hours after  exposure to
nitrosocarbaryl (0.015 mg/ml).  This was a significant  increase compared  to
control values (1% aberrant cells).  The toxicity of the  concentration used
was not given.

    Although each individual study concerning the mutagem'city of carbaryl
contains deficiencies, the body of evidence (i.e.,  data regarding both the
intrinsic mutagem'city, spindle effects, and the presence of carbaryl or

metabolites in mammalian gonads)  strongly suggests that carbaryl  may  have  the
potential to cause heritable genetic effects in humans.   However,  it  should be
emphasized that because carbaryl  appears to act as a weak mutagen in  the
reported gene mutation test systems, it is likely to act as  a  weak mutagen in
humans as well.
    In the case of nitrosocarbaryl,  there is evidence that it  causes  point
mutations in bacteria and thus, it is likely to have intrinsic mutagenic
activity in other organisms.  If nitrosocarbaryl  were to be  formed or were
present in humans, it might cause somatic mutations which may  be  involved  in
the etiology of cancer.  Although this is an assessment of genetic risk with
respect to somatic cell mutagenicity, an assessment of whether nitrosocarbaryl
has the potential to cause heritable mutations in humans could not be made
because there were no available data on its ability to reach or interact with
mammalian germinal tissue.


Ahmed, F.E., N.J. Lewis,  and R.W.  Hart.   1977a.   Pesticide  induced  ouabain
    resistant mutants in  Chinese hamster V79  cells.   Chem.  Biol.  Interact.

Ahmed, F.E., R.W. Hart,  and N.J. Lewis.   1977b.   Pesticide  induced  DMA  damage
    and its repair in cultured human cells.   Mutat.  Res.  42:161-174.

Amer, S.  1965.   Cytological effects of  pesticides.   I. Mitotic effects of
    N-methyl-1-naphthyl  carbamate (Sevin).  Cytologia 30:175-181.

Amer, S.M., and O.R. Farah.  1968.  Cytological  effects of  pesticides.   III.
    Meiotic effects of N-methyl-1-naphthyl  carbamate {Sevin).  Cytologia

Amer, S.M., M.A. Hammouda, and O.R. Farah.  1971.   Cytological and  morphological
    effects of the insecticide N-methyl-1-naphthyl-carbamate  (Sevin).   Flora
    (Jena) 160:433-439.

Ames, B.N., J. McCann, and E. Yamasaki.   1979.   The  methods for detecting
    carcinogens and mutagens with Salmonel1 a/microsome mutagenicity test.   pp.
    1-17, in Handbook of Mutagenicity Test Procedures, B.J. Kilbey,  M.  Legator,
    W. Nichols, and C. Ramel, eds.   Elsevier-North-Holland Biomedical  Press,
    Amsterdam, Netherlands.

Ashwood-Smith, M.J., J.  Trevino, and R.  Ring.  1972.  Mutagenicity  of
    Dichlorvos.   Nature 240:418-420.

Blevins, R.D., M. Lee, and J.D. Regan.  1977.  Mutagenicity screening of five
    methyl carbamate insecticides and their nitroso  derivatives using mutants
    of Salmonella typhimurium LT2.  Mutat.  Res.  56:1-6.

Brankovan, 0.  1972.  The meiotic effect of Sevin 50 after  treatment of corn in
    the embryonic and generative phases  of development.   Arhiv za Poljoprivredne
    Nauke 25:(90)128-138.

Brzheskii, V.  1972.  The study of the mutagenic properties of an insecticide
    from the carbamate group (Sevin).  Sov. Genet.  8:798-800.

Carpenter, C., C. Weil, P. Palm, M. Woodside, J. Nair, and  H. Smyth,  Jr. 1961.
    Mammalian toxicity of 1-naphthyl-N-methylcarbamate (Sevin insecticide).  J.
    Agr. Food Chemical 9:30-39.

Collins, T., W.H. Hansen, and H.V. Keeler.  1971.   The effect of  carbaryl
    (Sevin) on reproduction of the rat and the gerbil. Toxicol.  Appl.
    Pharmacol. 19:202-216.

Cook, W.L.  1980.  File note of telephone conversation with W.L.  Cook,  Georgia
    State University, and Vicki Dellarco, Reproductive Effects Assessment Group,
    EPA.  July 3, 1980.

Cook, W.L., S.A. Crow, and A.W.  Bourquin.   1977.   Inhibitory effects  of
 '   pesticides and polychlorinated compounds  on  representative  surface  slick
    bacteria.  Annu.  Mtg.  Am.  Soc. Microbiol.  (Abstr)  77:243.

DeGiovanni-Donnelly,  R., S.M.  Kolbye,  and  P.O. Greeves.   1968.   The effects of
    IPC, CIPC, Sevin, and Zectran on Bacillus  subtil is.   Experientia  24:80-81.

Degraeve, N., M. Moutschen-Dahmen, N.  Houbrechts,  and  A.  Colizzi.  1976.   The
    hazards of an insecticide  carbaryl  used alone  and  in  association  with
    nitrites.  Bulletin dela Societe Royale des  Sciences  de  Liege.  45:46-57.

DeLorenzo, F., N. Staiano, L.  Silengo,  and R.  Cortese.  1978.   Mutagenicity of
    Diallate, Sulfallate,  and  Triallate and relationship  between structure and
    mutagenic effects of carbamates used widely  in agriculture.   Cancer Res.

de Serres, F.J., and M.D.  Shelby.  1979.  The Salmonella  mutagenicity assay.
    Recommendations.   Science  203:563-565.

Dikshith, T.S.S., P.K. Gupta,  J.S. Gaur, K.K.  Datta, and  A.K. Mathur.   1976.
    Ninety day toxicity of carbaryl in  male rats.   Environ.  Res.  12:161-170.

Egert, G., and H. Greim.  1976.   Formation of mutagenic N-nitroso compounds from
    the pesticides prometryne, dodine,  and carbaryl  in the presence of  nitrite
    at pH 1.  Mutat.  Res.  37:179-186.

Eisenbrand, G., D. Schmahl, and R. Preussmann.   1976.  Carcinogenicity  in  rats
    of high oral doses of N-nitrosocarbaryl,  a nitrosated pesticide.  Cancer
    Lett. 1:281-284.

Eisenbrand, G., 0. Ungerer, and R. Preussmann.   1974.  The reaction of  nitrite
    with pesticides.   II.   Formation,  chemical properties and carcinogenic
    activity of the N-nitroso  derivative of N-methyl-1-naphthyl  carbamate
    (carbaryl).  Food. Cosmet. Toxicol. 13:365-368.

Elespuru, R., W. Lijinsky, and J.K. Setlow.  1974.  Nitrosocarbaryl,  as a
    mutagen of environmental significance.  Nature 247:385-387.

Epstein, S., E. Arnold, J. Andrea, VI.  Bass, Y. Bishop.  1972.   Detection of
    chemical mutagens by the dominant lethal  assay in  the mouse.   Toxicol. Appl.
    Pharmacol. 23:288-325.

Environmental Protection Agency.  1980.  Mutagenicity  Risk Assessments:
    Proposed Guidelines.  Federal Register 45(221):79488.  (November  13, 1980)

Fahrig, R.  1974.  Comparative mutagenicity studies with  pesticides.  IARC

Fiscor, G., and G.M.  Nil Lo Piccolo.  1972.  Survey of pesticides for
    mutagenicity by the bacterial plate assay method.   EMS News 6:6-8.

F>amm, W.G.  1977.   Approaches to determining  the  mutagenic  properties  of
 "  chemicals;  risk to future generations.   J.  Environ.  Path.  Tox.  1:301-352.

Grant, W.R.  1978.   Chromosome aberrations  in  plants  as  a monitoring  system.
    Environ. Health Persp.  27:37-43.

Greim, H., D. Bimboes, G.  Egert,  W.  Goggelmann,  and M. Kramer.   1977.
    Mutagenicity and chromosomal  aberrations as  an analytical  tool  for  in vitro
    detection of mammalian enzyme mediated  formation  of  reactive metaboTTtes.
    Arch. Toxicol.  39:159-169.

Hoque, M.Z.  1972.   Carbaryl, a new  chemical mutagen.  Curr.  Sci. 41:855-856.

Ishidate, M.  1980.  Letter from Motoi  Ishidate, Jr.  M.D.  National Institute of
    Hygienic Sciences, Division of Mutagenesis,  Tokyo, Japan,  to Dr.  Kuroda,
    Reproductive Effects Assessment  Group,  EPA.  Dated June  25,  1980.

Ishidate, M., Jr.,  and S.  Odashima.   1977.   Chromosome tests  with 134 compounds
    on Chinese hamster cells  in vitro - a screening for  chemical carcinogens.
    Mutar. Res. 48:337-354.

Kazarnovskaya,  M.L., and A.F. Vasilos.   1977.   The effect of Sevin  on the
    chromosomal appartus of cells in  vitro.  Zdravaookhranenie 20(4):14-16.

Kitagawa, K., M Wakakura,  and S.  Ishikawa.   1977.  Light microscopic  study of
    endocrine organs of rats  treated by carbamate  pesticide.   J. Tox. Sci.

Krylova, T.V., and A.V. Denisova.  1973.   Action of Sevin on  the Mongolian Tree
    Creeper.  Biol. Nauki  16:(10)25-30.

Lijinsky, W., and H.W. Taylor.  1976.  Carcinogenesis in Sprague-Dawley rats of
    N-nitroso-N-alkyl carbamate esters.  Cancer Lett. 1:275-279.

Marshall, T.C., H.W. Dorough, and H.E.  Swim.  1976.   Screening of pesticides for
    mutagenic potential using Salmonella typhimurium  mutants.  J. Agric. Food
    Chem. 24(3)560-563.

McCann, J., E. Choi, E. Yamaski,  and B.N. Ames.  1975.   Detection of  carcinogens
    as mutagens in the Salmonella/microsome test:  assay of  300  chemicals.
    Proc. Nat. Acad. Sci.  72:(12)5135-5139.

McCann, J.  1980.  Letter from Dr. Joyce McCann, University  of California,
    Lawrence Berkeley Laboratory, to Yicki  Dellarco,  Reproductive Effects
    Assessment Group, EPA.  Dated June 3, 1980.

Nagy, Z., I. Mile, and F.  Antoni.  1975.   The  mutagenic  effect of pesticides on
    Escherichia coli WP2 try".  Acta Microbiol.  Acad. Sci. Hung. 22:309-314.

Rashid, K.A.  1978.  The relationships between mutagenic and DNA damaging
    activity of pesticides and their potential  carcinogenesis.  Penn. St. Univ.,
    Ph.D. dissertation.

Regan, J.O., R.B. Setlow,  A.A.  Francis,  and W.  Lijinsky.   1976.
 •   MHVosocarbaryl:   its  effect on  human  DNA.   Mutat.  Res.  38:293-302.

Rickard, K.W.  1979.   Chemical  properties  and  potential  toxicological
    significance of N-nitrosocarbamates.   Univ.  Kentucky,  Ph.D.  dissertation.

Seiler, J.P.  1977.   Nitrosation in  vitro  and  in vivo  by  sodium  nitrite, and
    mutagenicity of nitrogenous pesticides. Mutat.  Res.  48:225-236.

Shirasu, Y., M. Moriya,  K. Kato, A.  Furuhashi,  and T.  Kada.   1976.  Mutagenicity
    screening of pesticides in  the microbial .system.   Mutat.  Res. 40:19-30.

Shpirt, M.B.  1975.   Toxicological evaluation  of dichlorodipheny-trichloroethane
    (DDT), hexachlorocyclohexane (HCCH), tetramethyl-thiuramdisulfide  (TMTD),
    Sevin, and Zineb when  acting on  a human cell  structure.   Gig. Tr.  Prof.
    Zabol. 17:(3)32-34.

Shtenberg, A.J., and M.N.  Rybakova.   1968.   Effects  of carbaryl  on  the
    neuroendocrine system  of rats.   Food Cosmet.  Toxicol.  6:461-467.

Siebert, D., and G.  Eisenbrand.  1974.   Induction of mitotic  gene conversion in
    Sarcharomyces cerevisiae by N-nitrosated pesticides.   Mutat. Res.

Thomas, J.A., C.S. Dieringer, and L. Schein.   1974.  Effects  of  carbaryl on
    mouse organs of reproduction. Toxicol. Appl. Pharmacol.  28:142-145.

Uchiyama, M., M. Takeda, T. Suzuki,  and  K.  Yoshikawa.   1975.  Mutagenicity of
    nitroso derivatives  of N-methylcarbamate insecticides  in  microbiological
    method.  Bull Environ. Contam. Toxicol. 14:389-394.

Vasilos, A.F., V.D.  Dmetrienko, and  I.G. Shroyt.  1972.   Colchicine-like action
    of Sevin on human embryonic fibroblasts in  vitro.   Bull  Eskp. Biol. Med.

Vasilos, A.F., V.D.  Dmitrienko, and  I.G. Shroit.  1975a.   Disruption of the
    mitotic system following acute Sevin poisoning.  Izv.  Akad.  Nauk Mold. SSR
    Ser. Biol. i.Khim. Nauk 3:64-67.

Vasilos, A.F., V.D.  Dmitrienko, and  I.G. Shroit.  1975b.   Changes in mitosis in
    chronic poisoning of rats with Sevin.   Izv.  Akad.  Nauk Mold. SSR Ser. Biol.
    i.Khim. Nauk 4:45-47.

Weil, C.S., M.D. Woodside, C.P. Carpenter,  and H.F.  Smyth, Jr.   1972.  Current
    status of tests of carbaryl for  reproductive and teratogenic effect.
    Toxicol. Appl. Pharmacol. 21:390-404.

Weil, C.S., M.D. Woodside, J.B. Bernard, N.I.  Condra,  J.M. King, and C.P.
    Carpenter.  1973.  Comparative effect  of carbaryl  on  rat reproduction and
    guinea pig tetratology when fed  either in  the diet or  by  stomach intubation.
    Toxicol. Appl. Pharmacol. 26:621-638.

     With respect to altered sperm-head morphology reported in
carbaryl-exposed employees (Wyrobek et a',  1980, submitted January 6,  1981,
Environ. Health Perspec.) discussed on page 32 of this document, the
possibility of exposure to other chemicals  was noted.   Dr. H.M.D. Utidjian of
Union Carbide indicated in a letter to EPA  on March 21, 1981 that the  workers
studied by Wyrobek et al. (1980) were not occupationally exposed to other
pesticides.  He stated, however, that there was a potential for concomitant
exposure to the reagent chemicals, alpha-naphthol and  methyl isocyanate,  and
the solvent toluene.  Exposures to methyl isocyanate were kept at very low
levels (below 1 ppm).  Presently, Union Carbide is collecting monitoring  data
in the workplace on these chemicals.  With  respect to  the ability of these
chemicals to induce sperm-head abnormalities, toluene  has been reported as
negative (Topham 1980) and no information was located  concerning the ability
of the alpha-naphthol or methyl isocyanate  to alter sperm morphology.
     It should be noted that the Reproductive Effects  Assessment Group (REAG)
made a recommendation to the Federal Insecticide, Fungicide, and Rodenticide
Act (FIFRA) Scientific Advisory Panel on July 23, 1980, that the
carbaryl-exposed workers be re-examined for altered sperm abnormalities to
confirm the initial findings of Wyrobek et  al. (1980)  and to better define the
dose-response relationship and address the  question of the reversibility  of
the effect.

Topham, J.C.  1980.  Do induced sperm-head abnormalities in mice specifically
     identify mammalian mutagens rather than carcinogens?  Mutat. Res.