Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Chapter 9 TRIFLURALIN AND ORYZALIN NO2 ~ /CH2 CH2 CH3 F3C - / \ - N \=< \CH2CH2CH3 NO2 NO2 ,: CH2CH2CH3 // \\ / NH2.O2 S - ( IN CH2CH2CH3 NO2 Trifluralin (a, a, ~-trifluoro-2 ,6-dinitro-N,N-dipropyl-~ toluidine; also known as Treflan) and oryzalin (3, 5-dinitro-N4, N4-dipropylsulfanilamide) are both members of a class of compounds known as dinitroanilines. Trifluralin is an orange crystalline solid that melts at 49°C. It has a low solubility in water (0 .3 ppm) and a low vapor pressure (2 x 10 4 mm Hg at 30°C). Oryzalin is a yellow-orange crystalline solid that melts at 141°C. It has a much lower vapor pressure (less than 10 7 mm Hg at 30°C) than trifluralin, but is slightly more soluble in water (2.5 ppm at 25°C). Trifluralin is produced via a ser ies of reactions beginning with the reaction of hydrogen fluoride with E~chlorotoluene to produce 4-trifluoromethylchlorobenzene. The latter compound is then nitrated followed by reaction with di-N-propylamine, which replaces tbe chlorine to form trifluralin. me production of oryzalin begins with the nitration of E:chlorobenzenesulfonic acid, followed by the addition of chlorine to form 2,6-dinitro-E~chlorobenzenesulfonyl chlor ice, followed by the addition of ammonia and di-N-propylamine to yield oryzal in . 228
PRODUCTION Eli I`illy and Co. {Elanco Products Division) is the sole U. S . producer of tr if luralin (Stanford Research Institute International, 1979 U.S. International Trade Commission, 1978~. The plant at Lafayette, End., has an estimated 23,000-metric ton capacity for production of trifluralin. Lilly owns the patent rights for oryzalin, but has contracted production to various other firms since receiving approval of oryzalin use in 1975. GAF Corporation made oryzalin for Lilly at Rensselaer, N.Y., from January 1975 to June 1976 (Chemical Week, 1979~. Proctor Chemical at Salisbury, N.C., produced oryzalin from January 1975 to January 1976. The U. S. International Trade Commission listed the Sodyeco Division of Martin Mar ietta Corporation, Sodyeco, N.C., as a producer of oryzalin in 1977 and 1978. Other processors and formulators of oryzalin for Lilly have included Bold Chemical of Tifton, Gal, Central Chemical of Hagerstown, Ark., and Helena Chemical of Helena, Ark. (Chemical Week, 1979~. Although no estimates of production capacity are available for these plants, the production of trifluralin in 1975 and 1978 is estimated at 12, 000 and 16, 000 metr ic tons, active ingredient, respectively. Oryzalin production in 1975 is estimated at 45 mete ic tons, active ingredient; 1978 production is estimated at 450 metric tons of active ingredient. 229
U_ Trifluralin and oryzalin are both used almost exclusively as herbicides. Consumption estimates for each in 1978 are shown in Table 9-1. The Environmental Protection Agency {EPA} teas approved both trifluralin and oryzalin for use as herbicides; however, both compounds have been challenged on several occasions. Trifluralin was challenged on the grounds that it contained nitrosamine contaminants and that its use therefore posed an unacceptable carcinogenic risk. Risk-benefit studies initiated by EPA concluded that benefits of trifluralin outweigh any risks if nitrosamine contamination is kept below 1 ppm (Chemical Marketing Reporter, 19791. Subsequently, Lilly has produced trifluralin with a nitrosamine content of less than 1 ppm, and has continued to receive approval for its use as an herbicide. Oryzalin has been suspected of causing heart-related birth defects among children fathered by workers involved in its manufacture and of causing skin rashes in workers. However, EPA investigations did not disclose any adverse effects, and no regulatory action is planned (Chemical Marketing Reporter, 1980~. Nonetheless, Lilly is required to maintain ongoing teratological studies to support the continued registration of oryzalin (Pesticide and Tbxic Chemical News, 1980~. 230
Table 9-1 Consumption of Trifluralin and Oryzalin Percent of Product Conta in ing compound Tr if luralin 103 Metric Ton. of Active Ingredient Cotton 21. 2 . 7 Deciduous fruits/nuts 1 0 .05 Peanuts 1 0.1 Soybeans 7 4 9 · 7 Sugar beets 1 0.05 Vegetables 2 0.3 Other f ield crops 1 0 .14 Industr ial/co~ercia 1 0.1 Total 100 13.1 OrYzalin Cotton 10 0.05 Deciduous fruits/nuts ~ 10 0.05 Soybeans 60 0.27 Industrial/commercial 20 0. 09 Total 100 0 .45 Stanford Research institute, 1979 231
EXPOSURE The principal routes of exposure to trifluralin and oryzalin are the following: workers are exposed dur ing n~anufactur ing process ; agricultural workers are exposed while applying~the substances to crops; and the general public is exposed through releases of the compounds into air and water during manufacturing processes, through drift, volatilization, and runoff as a result of application, and via contaminated food crops. Because more trifluralin is used than oryzalin, and because much more data about trifluralin exist, this chapter focuses on exposure to trifluralin. Trifluralin is produced at a single site in Lafayette, Ind. (population, 45,000~; thus, exposure of the public to trifluralin during its manufacture is of only local importance. Although no data are available on trifluralin plant discharges, small quantities are known to be discharged into the Wabash River as waste from the manufacturing processes. m e treatment procedure involves activated sludge, followed by settling, and then filtration through activated carbon (Specie and Hamelink, 19791. Parts per trillion (ppt) concentrations of trifluralin have been measured in the water downstream from the plant. However, the tissues of fish captured in these waters contain concentrations of trifluralin several thousand times greater than those in the water (Specie and Hamelink, 1979~. 232
The most likely source from which the general public is exposed to trifluralin is the herbicide's application to soils to control grass and broadleaf weeds. This is practiced primarily in the North Central and South Central States, especially in Illinois, Iowa, and Mississippi (Helling, 1976~. ~ Trifluralin is not an especially persistent chemical. m ere have been some studies of the persistence of trifluralin in soil under actual field conditions (Golab et al., 1979; Golab and Amundsen, 1975; Probst et al., 1967) These studies show that trifluralin concentrations decreased from 10% to 15t of their original value 1 year after application. Trifluralin is lost through volatilization as well as through degradation. At this rate of loss, trifluralin does not tend to accumulate in soils receiving repeated applications. Although the rate of volatilization of trifluralin depends to a great extent on the method of its application, such dissipation provides a signif icant potential exposure route for persons living downwind of treated f ields . If trifluralin is applied to the soil surface, up to 90% of the compound may be volatilized within 2-3 days of application . If it is incorporated into the soil, volatilization losses can be as little as 3% or 4% in 90 days (Taylor, 1978~. In an experiment in which trifluralin was sprayed onto soil and then tilled into the soil, White et al. t1977} measured ~ maximum concentration of 16.5 Ug/m 3 20 cm above the soil during application. The maximum concentration was 3.4 ~g/m 3 20 cm above the soil on the second day 233
after the compound was tilled into the soil, decreasing to 100 ng/m 3 after the 35th day. They estimated total volatilization losses to be at 22.41 during the growing season (120 days) and vapor losses during spraying to be at 3.58. Exposure of persons living downwind of sprayed fields is somewhat reduced because trifluralin is subject to fairly rapid photochemical decomposition. Measurements of trifluralin degradation in air indicate a balf-life of 20 minutes under midday summer sunlight conditions. Woodrow et al. {1978} observed that the half-life increased to 193 minutes in the fall. Trifluralin is stable in the dark. Although relatively immobile in soil, trifluralin may be transported long distances from its initial source of application via runoff into streams and rivers. Indeed, trifluralin has been detected at levels of up to 0.2 ~g/1 in the Cape Fear River, N.C. Trifluralin concentrations tended to correlate with periods of greatest runoff and soil erosion on agricultural lands adjacent to tributaries of that waterway {Horder, 19771. Although there is probably minimal direct exposure.of humans via this route, the tendency of trifluralin to accumulate in fish must not be overlooked (Specie and Hamelink, 19791. Humans may also be exposed to trifluralin through food crops treated with the herbicide. However, analyses of residues in a wide variety of tolerant crops indicate that trifluralin is not 234
incorporated into the leaves, seeds, or fruits (measurement sensitivity, 5-10 ppb). In treated root crops, such as onions and garlic, trifluralin residues are found only in the outer shell, which is usually discarded before consumption. me exception is carrots ~ : 31% of the total residue of 0.65 ppm was found~in the interior of the carrot proper--69% was concentrated in the peel (Protest et al., 1967) 235 .
ANALYTIC METHODS Methods to analyze trifluralin in a variety of vegetables, plant tissues, soil, water, oily crops, wheat grain and straw, and mint and mint hay are described in the Pesticide Analytical Manual of the Food and Drug Administration (1973~. Trifluralin is extracted from crops by blending it with methanol and subsequent extraction from the solvent into dichloromethane. After evaporation of the dichloromethane, the extract is dissolved in hexane, cleaned up on a column of Florisil, and analyzed by electron-capture gas chromatography (EC-GC). GC columns (1.83 m) packed with 3% XE-60 or 5t SE-30 on Chromosorb W are used at approximately 180°C with the carrier gas flowing at 90 ml/minute. The retention time (tR, for trifluralin is 3 to 4 minutes. The analytic procedure is modified as required for aqueous or oily samples. Recoveries of approximately 80% or more can be expected, and the sensitivity ranges from 5 to 10 ppb. Samples containing residues of benzene hexachloride, ethion, and/or zineb require additional cleaning. The interfering compounds are separated by thin-layer chromatography (TEC). Ihe trifluralin- is then scraped from the plate, elated, and assayed by EC-GC. The procedure used at North Carolina State University (T.J. Sheets, North Carolina State University, personal communication, 1980) to determine trifluralin residues in soils requires a 4-hour Soxhlet extraction of the oven-dried sample with benzene-isopropyl alcohol (1:2~. The extract is evaporated just to dryness, dissolved in hexane, cleaned on a Florisil column and subsequently analyzed 236
by EC-GC. Glans columns {1.83 m long 0.635 cm outside diameter) packed with 4% SE-30 plus 61 QF-1 or with 51 Carbowax 20M on Gas Chrom Q are opera ted at 155°C with a Me flow of 80 to 100 ml/minutes. Under these conditions, the tR for trifluralin is approximately 2.0 minutes. Recently reported analytic methods have focused mainly on GC techniques, coupled with various types of detectors including mass spectrometers. Payne et al. (1974) developed a procedure to analyze trifluralin, diphenamid, and paraguat in admixture in soil, sediment, and water. The procedure permits simultaneous GC assays of trifluralin and diphenamid, without cleanup, by employing a Couison electrolytic conductivity detector. Paraquat is determined calorimetrically. m e 1.83 m long glass GC column (0.635 cm outside diameter) was packed with 10% DC-200 on Gas Chrom Q. Recoveries of trifluralin spiked into water at concentrations of O .05 to 10 ppb were 82% to 91%; those from soil spiked at 0 . OS and 1.0 ppm were 861 and 941. Smith (1974 ~ used acetonitr ire-water (9 :1) and ultrasonication to extract residues of four herbicides, including trifluralin, from three types of soil. ffl e addition of excessive amounts of water and saturated sodium sulfate solution to the extract facilitated subsequent partitioning of the herbicide residues into hexane. me hexane extract was then concentrated and assayed by EC-GC. The GC system consisted of ~ 1.5 m long glans column (6 ~ outside diameter), packed with 10% OV-1 on Q~romosorb G-HP operated at 237
ml/minute, tR's for dichlorobenil, trifluralin, dinitramine, and triallate were approximately 1.0, 2.9 , 4 . 5 , and 5.0 minutes, respectively. Recoveries of all four compounds spiked into the three types of soil at 0.05 and 0.5 ppm levels were 92% or more. Lawrence (1976) examined the separation characteristics of Several GC liquid phases (i.e., 31 OV-1, 4% SE-30-6. SP-2401, 10% DC-200, and 5% DEGS on Chromosorb W HP) for 12 pesticides, including trifluralin, by using the Coulson detector. Sensitivities (50% scale) for trifluralin at a tR of approximately 3 minutes on OV-1 at 175°C were reported as 6 ng and 3 ng for the nitrogen and chlorine modes, respectively. Later, Lawrence et al. (1977) reported a confirmatory procedure for pesticides that contain nitrogen dioxide (NOW. ate pesticide residue in ~ m] of benzene was shaken with 0.5 ml of aqueous chromous chloride to convert the NO2 groups to NH2. me solution was made basic and the product extracted with benzene for analysis by GO with a Coulson detector (nitrogen mode). The products were found to be approximately as sensitive as the parents and had tR values of 0.4 to 0.9 minutes relative to the parent compound on the SE-30/SP-2401 column. The procedure was used to confirm trifluralin residues in extracts of potato spiked at levels of 0.5 to 1.0 ppm. Woodrow et al. (1978 ~ used EC-GC to study the behavior of trifluralin vapors released into the atmosphere as emulsifiable concentrate sprays. The experiment wan conducted with a 1.8 m long glass column (3 mm, internal diameter), packed with 3% OV- 17 238
ng/m3. In a similar study, Soderquist et al. (1915) studied photoproducts of trifluralin in air by using EC-GC, TIC, and GC-mas. spectrometry (MS). ffle photoproducts were subjected to Em: on 0.5 mm plates of silica gel G containing 1% zinc ortbosilicate phosphor and developed with hexane-acetone (3:17. file resulting bands were scraped off, elated with acetone, and quantitated GC-MS by using a 1.5 m long glass column (3 mm, internal diameter) containing 25 OV-1 on Chromosorb G. The column temperature was programmed to rise from 150°C to 270°C at 10°/minute with a helium carrier flow of 16 ml/minute. As each compound elated, its mass spectrum we" recorded and identified by comparison with an authentic specimen. Samples of analysis by EC-GC were analyzed under conditions similar to those described by Woodrow (1978~. Downer et al. (1976) compared analytical results for residues of benef in and trifluralin in soil by using a GC-MS procedure and an established EC-GC assay. Comparable sensitivities (<10 ppb) and results were obtained for both the mass fragmentography [multi-ion detector, MID} and the EC-GC procedures . file MID procedure W8S ma id to have a shorter analysis time and to be lest susceptible to contamination than were EC-GC assays. Heck _ al. (1977} reported a high-pressure liquid chromatography (HPLC}-MS procedure developed to obtain pharmacokinetic data from rats dosed with trifluralin. Isotope dilution analysis, with nonfragmenting mass spectrometry offered several advantages over 239
other techniques for determining trace quantities of organic compounds in biologic materials. Residues of the compound were separated from rat tissue and excrete by sequential high-pressure gel permeation and reverse-phase liquid chromatography. The cleaned samples were then quantitated by field ionization MS. 240
HEALTH EFFECTS The mutagenicity and chronic toxicity of both trifluralin and a contaminant, N-nitrosodipropylamine (NDPA), have been reviewed extensively by EPA's Special Pesticide. Review Division, Office of Pesticide Programs (OPP), and is reported in ~Trifluralin (Treflan} Position Document. (Environmental Protection Agency, 19791. A discussion of specific needs of EPA with several representatives from the OPP led to the decision that this committee should focus only on a review of the possible genetic toxicity of trifluralin. Consequently, the committee limited its study to a review of the EPA's documents, which is summarized below with minor editorial changes and updated information from several studies. Tr i f lure 1 in Mutagen ic ity Data Technical trifluralin, containing known and undetermined levels of NDPA, has been tested in a number of system without producing mutagenic e f feats. presented in Tables 9-2 and 9-3. The results of these studies are 241
a) 0 I: 0 0 at: v - ~s 2 I: H Cat 1 0 EM o U) to En · - C) Ed A: Hi: P. - · - :3 · - En V C En :, U) a: · - U] to ·r1 V U) U) U] ~0 EM 01 rat ~ · _t 4 _ - 1 ' ' ' ' ' - 1 ~ ' ' ' ' vl a1 681 1 .1 ~1 =1 a1 m1 1 ~1 0 =1 =1 =1 =1 ~1 ~ ~' ~ =1 =1 =1 ~1 c 471 ~1 ~'1 ~1 c)1 c: .)1 ~1 ~1 O ~ O ~ u~ ~ ~: ~: c: ~ Y ~n Y ~: ~ ~ c: ~ IRIl ~ 'C ~ ~ ~ '8 ~ ~ · - · - ~ · - · - ~ `: ~ ~ ~ ~ · - u~ tn u~ u~ u~ x: u~ u~ u~ u Vl U1 Ot U ~1 ~' Dl .DI ~1 = ~ ~ ~ ~ ~ ~ 0 Z Z Z Z Z Z Z Z Z Z Z Z Z V I Q U] 3 v ~: ~! ^H · - · - s u) . - _' ~ C: u, Q OD ~ _ U) t.)1 ~.J o o ~ ~ ~ ~ 0 ~ ~ ~ · - u~ eG U ~ ~ ~ ~ ·~ 0 s v U) ~3 0 ~0 · - s :x o o C) s Ll . - ·^ 8 · - · - 0 U) o ~ C s o · - · - ~ ~C D O ~ O E~ o 4,' · - O U) Q O O ~: C) 242 0 C} H ~ 0 · - 1 · - C) ~ 60 ~ 0 U) ·,' ·,' 0 U~ O m - V ~ ~ ~ ~ ~ O · - 0 + s ~ ~ ~ ~O ~ ~ ~ ~ O .^ ·,1 O ~ ~ ~ ~ a, _ ~ _t _1 t S c: 1 s m- - ~ ~ ~ ~ ~ t) 3 ~ ~ ~ ~ ~ 3 U) :c ~n 1 1 |: D `: O ~ O ~ :^ t) 0 ~ C' ' ~ ~ ~: ~ dV ~ 2 D V ~ ~ a _ ~ ~ ~ V 14 .,. ~ ~ c: ~ · - ~5 PJ ~ ~ · - ~ ~: U) c: C: ~ ~ ~: . ~ O O £ u, % ~ · - ~ ~ s: a ~ ~ u, x: ~
- ~: lo - E~ o 0 0 E~ - v a: P4 a lo P. - cn a: a lo: . - U] In . - c) VO to En ~ ~ ~ an. o, ~ a: ~~ m · - a: · - 1 O ~1O -1~1~1~1 al =1 =1 JJ1 =1 ~1 ~1= a: ~ :' oo ~ O U O c: `: ~ CD U s s s V x: ~e ~ ~e _I ~e _. . ~1 AS1 I· - r eY z I ~1 1 - eY o Y I ~1 ~1 eY o e~ ~: - 1~1~1~1 - 1 - 1 - 1 a1 U) e~ J~ ~ g) U) U~ U) tO O ~ O O O O O O ~ Z P. P4 Pl P~ P. p. t tn tn C: o E: e~ 60 e~ m · - e~ U O O C'` ~Q ~e '`t E~ O U~ ) t - ) el~ ~ o · - S e~ U 00 m ~n 1 c) L U] · - S 243 0 ~n s e~ · - o u) C . - · - ~: oB o ~s v o 8 s C)
TAB~ 9-3 Mutagenicitv and Related Tests with Formulated Treflan {Part I ) and Unspecif fed Forms of Tr if luralin (Part II ) Part I: Treflan Foraulations Test Systems Species/Strain Result References Gene Mutations Bacter ial Salmonella typhimur ium (8 straine) Neg! Anderson et al., 1972 Insect Drosophila melanogaster Negt,C Murnik, 1918 Chromosomal Mutations Insect Drosonbila ~nelanogaster pofib,c Murnik, 1978 Other studies Exposed Hu~nans posd Yoder et al., 1973 Neurospora Po~ Gr if f iths, 1979 Sordar ia Bond and McMillan, 1979 Mouse bone marrow Pos Nehez et al., 1979 Part IT: unspecified Trifluralin Gene Mutations Bacterial Salmonella typhimurium! Neg Shirasu et al., 1976 TA 1535 Neg Shirasu et al. 1976 TA 1536 Neg Shirasu et al., 1976 TA 1537 Neg Sbirasu et al., 1976 TA 1538 Neg Shirasu et al., 1976 Ch romosoma 1 Mutations Escher ich ia col ia B/r WP2 her~ Neg Shirasu et al., 1976 WP2 her~ Neg Shirasu et al., 1976 Plants Haemanthus katheriniae e Jack~son and Stetler, 1973 Tradeacantia paludosa f Sawamura and Jackson, 1968 Vicia faba f Sawamure and Jackson, 1968 Triticum aestivum Po~ Bartels and Hilton, 1973 Zes ~ys Postl Bartels and Hilton, 1973 Salamanders Triturus helveticus JI Sentein, 1977 Pleurodeles weltl ~ Sentein, 19?7 Pr in~ary DNA Damage UNA Repair, Bacteria Bacillus subtilis B17 (Rec+} /M45 (Rec~ ) Ne" 244 Shirasu et al., 1976
- - a Metabolic activation not used. b Inconclusive results. c Product sample used contained 177 ppm NDPA. d Chromatid lesions in lymphocytes of workers exposed to herbic ides. Decreased number of microtubule~, accumulation of large vesicles at the cell plate region. f Disruption of mitotic process temporarily impeded ch romosome movement. Inhibition of mitosis as a consequence of spindle disruption 245
Bacterial Tests. Simmon et al. (1977) tested 20 pesticides, including technical trifluralin (90%) containing 87 ppm NDPA as a contaminant in reversion-type mutagenic assays. Four Salmonella typhimurium strains and the WP-2 strain of _ cold were used, both with and without mammalian metabolic activation systems. Activation was obtained by using liver preparations from rats pretreated with polychlorinated biphenyl (PCB) Arochlor 1254. Trifluralin produced negative results in this study. Simmon _ al. (1977) also tested the same trif~uralin sample for unscheduled DNA synthesis in human fibroblasts (WI-38 cells), mitotic recombination in the yeast Saccharomyces cerevisiae strain D3, and preferential toxicity in repair-deficient strains of E. cold and Bacillus subtilis, as compared to strains that could repair DNA damage. Each of these assays was performed both with and without mammalian metabolic activation over a wide range of trifluralin concentrations. Findings for trifluralin were negative in all assays. The experimental and data-reporting procedures used in this study appear to have been adequate. Anderson _ al. (1972) evaluated 109 herbicides, including both technical and formulated (44.5%) trifluralin e Specifically, the investigaters looked for induction of point mutations in a battery of standard bacterial and viral plate assays involving base a ir substitution and frameshift reversions as well as forward mutation. mey compared their results to positive results with known mutagens specif ic for each of four assays. Single doses of all test 246
chemicals, including formulated trifluralin (20 or 25 ug/plate), did not induce changes significantly different from the spontaneous rates of mutation in eight histidine-requiring mutants of Salmonella typhimurium, two rII mutants of bacteriophage T4, or E. cold strain KB. This study did not conclusively demonstrate lack _ _ ~ Of mutagenicity because the investigators provided no exogenous metabolic activation to mimic possible conversion of the chemical to potentially active intermediates and they administered only one dose. Shirasu _ al. (1976) studied the mutagenicity of an unspecified form of trifluralin in four histidine-requiring strains of Salmonella ts~phimur ium in a standard Ames assay as well as in l differential toxicity assays with Bacillus subtilis strains H17 (Rec+) and M45 (Rec ~ and in reversion assays with two strains of E. cold that require tryptophan. In each of these assays, the investigators treated bacterial cultures with a single saturated paper disc containing a 0.02 ml solution of a standard sample made up at a concentration of 1 me trifluralin per milliliter of dimethylsulfoxide. Irifluralin showed negative results for mutagenicity in this study. Hbwever,the investigators did not provide mammalian metabolic activation, and they tested only one concentration. mus, these results are inconclusive. Insect Studies. Preliminary results from a study of both _ genetic and chromosomal effects in Drosophila melanogaster (Murnik, 1978 and I. Mauer, EPA, personal communication, 1978) showed 247
No evidence ... that trifluralin induces point mutations in Drosophila,. but some of the results of these two separate studies are contradictory. In one portion of the first study, larvae were fed a diet containing 0.011 formulated trifluralin {44.5% AT) throughout their stages of development, and the number of "'x-linked deaths was recorded in the F2 (second) generation. In replicate tests, no significant differences were found in mortality rates between the treated (0.10% for the first test) and combined control (untreated spontaneous, 0.12%) groups. No positive controls were reported, and the formulated trifluralin tested contained approximately 177 ppm NDPA. In the second portion, adult males fed 0.021 of the formulated trifluralin for 2 days also incurred no increase in sex-linked deaths. Although preliminary results from the two studies showed no evidence that trifluralin induces point mutations in Drosophila, Murnik (1978) reported an increase in chromosomal nondisjunction. Her first study involved the chronic feeding of 0.01% formulated trifluralin throughout the larval stages. Feeding at this stage resulted in a significant increase (0.12%) of XXY males compared to those of a control population (0.048~. However, feeding 0.02t formulated trifluralin to adult male Drosophila for 2 days resulted in no incre ase in nondis j unct ion . XXY nond is j unction wa s the only chromosomal aberration reported in this test. There were no increases over controls in chromosomal loss (XO progeny) or breakage. When the chromosomal portion of the first study was repeated, with technical trifluralin having no detectable NDPA, nondisjunction was not observed in test animals at a level 248
significantly different from that of the control population. Emus, the cytogenetic resul~cs from these two studies are inconclusive as well as contradictory (cna isson, 1978 ~ . Studies with Fungi. Trifluralin has also been tested for nondisjunction in Neurospora crassa (Griffiths, 1979) as well as in Sordar is brevicollis (Bond and McMillan, 1979~. Griffiths . concluded that trifluralin produced aneuploidy in Neurospora. Nondose-related increases were found over a dosage range of 1-75 me trifluralin/liter of culture. No information was given on the purity or source of the trifluralin nor of its ability to induce other genetic effects in Neurospora. m e authors considered the results obtained in Sordaria to be inconclusive. Rodent Tests. Male mice were injected with a trifluralin formulation containing 26% trifluralin (OLITRE ~) in the following dosage regimen: A single dose of 200 mg/kg, two 100 mg/kg doses four 50 mg/kg doses, and a single 0.6 mg/kg dose. The fractional doses were administered at 3-day intervals (Nehez et al., 1979~. Significant increases in chromosomal aberrations in the bone marrow were observed at all but the lowest dosage regimen. The LD50 of the trifluralin formulation was 600 mg/kg. The authors concluded that this preparation induced chromosomal aberrations in bone marrow at total doses of as little as one-third the LD50. Plant Studies. A number of plant studies have been conducted to determine whether trifluralin can disrupt the cellular spindle apparatus. The studies were not performed specifically to assess 249
the issue of mutagenicity. In an In vitro and ultrastructural study of cellwall free endosperm cells of the African blood lily (Haemanthus katherinae), Jackson and Stetler (1973) reported that concentrations of tr i floral in, rang ing f rom 0 . 1-100 .0 ppb, inhibited the rate at which cells progressed through all stages of mitosis from prophase to cell plate appearance. They observed these ef feats by time-lapse phase microscopy dur ing a 2-bour period. Since a. 1 ppb had a near-maximum inhibitory effect, the data presented from all concentrations were pooled. Electron microscopic studies showed a decreased number of microtobules and an accumulation of large vesicles in the cell plate region. The ultrastructural and mitotic index studies appear to have been conducted according to established protocols, but the bioassay used to assess these effects is not well documented. Furthermore, Jackson and Stetler did not establish a dose-response relationship, included no positive control in the study, and provided no information on the amount of NDPA that contaminated the study material. Nonetheless, this study does indicate that trifluralin interferes with the formation and function of plant cell microtubules, and the substance, may therefore, disrupt the mitotic spindle, thereby inducing numerical chromosomal aberrations. Sawamura and Jackson (1968) treated stamina! hair cells of the tetraploid Tradescantia paludosa and leaf cells of Vicia faba with 0.2 to 1.6 ppb of trifluralin. The degree of NDPA contamination for this material was unknown. At the highest dose (1.6 ppb), the 250
authors reported the appearance of bicentric bridges. in late stages of mitosis (anaphase and telophase) in both cell types, and cell elongation in stamina! hair cells only. Miss study demonstrates that trifluralin can disrupt various stages of plant cell mitosis; however, the report is of limited value because the system tested is questionable, the data are not quantitative, and the study was not designed to assess mutagenicity. Bartels and Hilton (1973) treated wheat (Triticum aestivum L., C.I. 5303) and corn (Zea mays L., Yellow Dent U.S. 13~ with technical-grade trifluralin at 10 4M. Cell division in the roots of the germinating seedlings appeared to be arrested at metaphase. me arrested cells showed no microtubule formation. Tr if luralin did not inhibit microtubule assembly nor did it bind to the microtubule protein as does colchicine. The investigators concluded that trifluralin acts on the microtubular organizing centers, rather than on the microtubules per se. Salamander Study. Sentein (1977) reported that trifluralin inhibited mitosis by interfering with the spindle apparatus in two urodele salamanders, Triturus helveticus and Pleurodeles waltl. Eggs of these species were incubated in various concentrations {1/8 through full saturation) of an unspecified form of trifluralin for 1 to 10 mitotic cycles prior to the beginning of cleavage, or at the 2-, 4-, 8-, and 16-blastomere stages. Cytologic observations were made during treatment and after various periods of incubation following transfer of the eggs to a trifluralin-free culture medium. At similar concentrations, the effects were more severe in 251
Pleurodeles than in Triturus eggs, but multinucleate blastomeres and disorganized mitotic figures occurred in both species. Sentein also reported disturbances in chromosomal condensation, especially at chromosomal sites associated with the mitotic spindle attachments, and gaps (discontinuities) at prophase. According to the author, the cytologic effects of trifluralin resembled those induced by classic antimitotic agents, but trifluralin was much less potent. The author concluded that these effects demonstrate that tr~fluralin interferes with the formation or function of cellular microtubular elements. This study is difficult to interpret because of the lack of details pertaining to its protocol. For example, the source and composition of the trifluralin were not stated, nor were control data included (except that the solvent was reported to be polyethylene glycol). However, the study does confirm, in an animal test system, the potential antimitotic action of trifluralin previously found in plant cytolog ic studies. The cytologic studies also support the genetic studies in Neurospora and Drosophila that indicate possible nondisjunctional activity of trifluralin (Mauer, 19781. Human Survey. Yoder et al. (1973) observed chromosomal alterations in lymphocyte cultures prepared from samples obtained from people who apply pesticides. Blood was drawn once during the midwinter lull in spraying operations and again during the peak summer spraying per iod. Forty-two white male workers, who had from 252
1 to 25 years (mean exposure, 8.5 years) of prior occupational exposure to a variety of pesticides, were matched as closely as possible in age and physical characteristics to a control group of 16 businessmen, students, and teachers with no history of involvement with pesticides. The exposed group was divided into two subgroups. One consisted of 16 people who bad been exposed to a variety of 17 insecticides; the other consisted of 26 employees of weed control agencies who had been exposed to 14 herbicides (most frequently to 2,4-D, amitrole, and atrazine, but also to formulated trifluralin). The incidence of chromatic lesions per person in the worker groups increased significantly over that in the control group, but only in blood samples taken in the summer. Although Yoder et al. Observed no heteroploidy (which may be indicative of nondisjunction) in any of the exposed or control cells, the authors did note a small number of chromatic exchange figures among tbe exposed groups. mis study must be regarded as inconclusive in implicating trifluralin as a chromosome breaker, because it was only one of many pesticides used by these workers. NDPA Mutacenicitv Data The principal contaminant of technical trifluralin preparations is NDPA, which is a demonstrated oncogen in rodents (Mbntesano and Bartsch, 1976~. NDPA has been studied in In vitro mutagenicity assays with bacteria and yeasts, as well as in mammalian cell culture, coupled with appropriate mammalian metabolic activation 253
systems (see Table 9-2~. In bacterial assays, NDPA has caused reverse mutations by base-pair substitution at concentrations of up to 1.0 mmol, but only in the presence of complete liver enzyme preparations from rodents. When the cofactors for the microsomal mixed-function oxidase were omitted, the mutagenic effect was absent. Positive results for gene mutation as well as for chromosomal aberrations were also obtained in Chinese hamster lung cell cultures treated with 20 mmol NDPA and a rat liver enzyme preparation (Kuroki et al., 1977; Matsuoka et al., 19797. Krueger (1973) also found direct evidence that NDPA alters genetic material In viva. He reported the presence of alkylated guanine residues in DNA of rats given NDPA. Trifluralin Derivatives The mutagenic potential of degradation and/or metabolic products of trifluralin has also been assessed. Evidence indicates that trifluralin may degrade into a series of products, including substituted benzimidazoles in a mammalian-derived In vitro microsome system (Nelson et al., 19771. Such conversion has been reported to occur under ultraviolet photodecomposition conditions, especially in the vapor phase above treated soil, as well as in the soil. This finding is of concern because some benzimidazoles have been shown to be mutagenic (Seller, 1972~. A report by Nelson et al. (1977) presents some interim results from bacterial mutagenicity assays performed with nine trifluralin 254
metabolites, including some benzimidazoles. In these analysis, the investigators used plate incorporation at concentrations of up to 200 ~g/plate for a standard battery of five test strains of S. typhimurium, both with and without metabolic activation. Summarizing the results, Nelson reported that he had found No potent mutagens among these trifluralin derivatives tested thus far, n as compared to the expected response of positive controls appropriate to each of the test strains. Mutagenic Risk Assessment - Neither technical nor formulated trifluralin (containing NDPA at levels of up to 177 ppm) has shown any mutagenic activity in the studies cited here. me principal contaminant, NDPA at concentrations greater than 20 times those contained in current formulations of trifluralin (< 1.0 ppm), has induced mutations in various test systems (Mauer, 1978~. HDPA is therefore considered to be a mutagen. At least two situations of potential mutagenic risk exist: m e direct effects on DNA and genes related to the NDPA contaminant, and the potential effects trifluralin induces on the spindle apparatus. DNA and Gene Effects. When tests were performed with metabolic activation, technical trifluralin (containing approximately 87 ppm NDPA) did not produce significant gene mutations or primary DNA damage. Formulated trifluralin (as Treflan or unspecified) also 255
produced negative results in some of the same tests; however, the results of these latter tests are inconclusive because they were performed without exogenous metabolic activation. Other preliminary studies indicate that Treflan, containing 177 ppm NDPA, as well as trifluralin with no detectable NDPA, produce negative results in the Drosophila sex-linked recessive lethal test. On the other hand, HDPA by itself has been shown to be mutagenic in several in vitro microbial test systems by causing base-pair substitution and primary DNA damage (Chaisson and Burkhalter, 19781. NDPA concentrations in the trifluralin preparations tested may have been too low to produce gene mutations or direct DNA interaction, especially in the presence of trifluralin {Chaisson and Burkhalter, 1978~. Trifluralin-NDPA mutagenicity data are not adequate to determine, much less to quantify, any risk for gene or DNA interactions posed by trifluralin. Any potential DNA and gene effects are associated with the NDPA contaminant of trifluralin formulations. To pose a potential, heritable genetic risk to humans, a chemical must be mutagenic and must be capable of reaching mammalian germ cells in a metabolically active form. There is no evidence showing whether mutagenically active forms of trifluralin or NDPA do or do not reach mammalian germinal tissue or whether these compounds are metabolized In situ to active forms if they do reach these tissues. me NDPA data in Table 9-2 indicate the need for metabolic activation of this compound before it can induce mutagenic responses in test organisms. Although NDPA has 256
mutagenic activity in some In vitro test systems, including mammalian cells in culture, no in viva tests have been performed. Data on the structurally related aliphatic nitrosamines, dimethylnitrosamine (DMN), and diethylnitrosamine (DEN) can be used to bridge the information deficits concerning the in vivo mutagenic activity of NDPA. DMN and DEN are mutagenic in both the Ames S. typhimurium and Drosophila sex-linked lethal tests. Three mouse dominant lethal studies on these chemicals have also been performed. A single intraperitoneal dose of DEN (13.5 mg/kg body weight) did not significantly increase the number of mutations in the offspring of treated males {Propping et al., 19721. DMN also produced negative results when male mice were administered DEN at 8 or 9 mg/kg body weight by the same route (Epstein et al., 1972~. DMN was reported to produce a weak dominant lethal effect in a second study with male mice (Propping et al. , 1972) . me second study has positive results with a DMN dosage lower than that yielding a negative response-in the first study; however, the mouse strain and route of administration differed. Propping et al. used only a single treatment group. ffl e lack of varying treatment levels precluded any within-experiment replication of the results or knowledge of dose-response relationships. The authors did state that the 4.4 mg/kg dose was the highest dosage of DMN compatible with survival. Because of the great variation in the responses of animals in the dominant lethal test and the positive finding at a level of signif icance just meeting the accepted critical level, the study seems inconclus ive. At face value , it suggests that DMN can reach 257
the mammalian gonad. A negative interpretation, however, is consistent with the finding that neither DMN nor DEN stimulated unscheduled DNA synthesis in the mouse testis following intraperitoneal administration of the test compounds and tritiated thymidine (Gary Sega, personal communication, 1979~. Also, DEN produced negative results in a specific locus test with mice (Russell, 1977~. NDPA itself has not been tested for germinal or In vivo mammalian mutation. ffl e available evidence does not support that Treflan-containing NDPA causes a significant risk to DNA and genes for the following reasons: o NDPA appears to have point mutagenic activity in some in vitro systems, but information is lacking from in viva tests. Some other short-chain alkylnitrosamines have been reported to produced positive results in the Drosophila sex-linked recessive lethal test. O There is no direct evidence that NDPA does or does not reach the mammalian gonad in a genetically active form. As for other nitrosamines, it has been reported that neither DMN nor DEN stimulates unscheduled DNA synthesis in the mouse testis. Only one of three dominant lethal studies with these chemicals in mice suggests a positive effect, and that study reported a very weak positive finding with DMN. However, DEN produced negative results in a specific locus test with mice. O Testing of trifluralin products containing 87 ppm NDPA teas shown negative results for both mutagenic and DNA-damaging activity. 258
o A preliminary study with Treflan containing 177 ppm NDPA produced negative results in the Drosophila sex-linked recessive lethal test. o Exposure of humans to NDPA through trifluralin use is expected to be very low. At this time, it is not possible to quantify the mutagenic hazard that might be associated with the use of trifluralin contaminated with NDPA because information on the presence of the active compound in the mammalian gonad and the results of germinal testing are lacking. Occupational exposures to NDPA ~ 5.05 ~g/yr) and exposures of the general population through consumption of treated food (approximately 1.92 x 10 9 mg/kg body weight/day assuming the presence of a residue and a 5 ppm level of NDPA contamination in Treflan) are very low. Further, even plants and runoff water from fields treated with trifluralin (containing NDPA) did not contain any NDPA. The manufacturer has already lowered the contamination to 1 ppm or less. -thus, any risk is reduced further by a factor of approximately five. Because r isks of adverse ef fects are intimately related to exposure and because the expected exposure of humans to NDPA is low, it is also expected that any risk from point mutagenic effects would be minimal. To obtain a better evaluation of point mutagenic risks, other tests should be conducted on NDPA, including studies assessing it'; ability to reach the mammalian gonad in a metabolically active form. 259
Spindle Effects. The limited studies that have been conducted appear to show that high concentrations of tr if rural in (with or without stated levels of HDPA) can disrupt formation or function of the spindle apparatus in dividing cells, and thus have the potential to cause abnormal segregation of chromosomes (nondisjunction). Tests with formulated tr if rural in (conta in ing approximately 177 ppm NDPA) in Drosophila showed nondisjunction. Replication of these tests with tech n ical tr if rural in having no detectable NDPA produced negative results. However, positive results, showing effects on the spindle, were reported when formulated trifluralin (NDPA content unknown) was tested on Neurospora. The pos it ive chromosomal ef feats repor ted in plants and sa lamanders ind icate the t tr i f rural in (or tr if lure 1 in plus NDPA) may affect spindle fibers by interfering with microtubule formation or function. However no comparable studies in '.~mnalian test systems, either in vitro or in viva, have been reported. Because the mechanism of cell division does not differ significantly between plants and animals, similar spindle effects might be expected to occur in mammals exposed to trifluralin. With this in mind, the committee surveyed mammalian and fish studies for evidence of mitotic disturbances, abnormalities in treated embryos, or any other chromosomal, spindle, or cellular effect of trifluralin on developmental processes. Overt manifestations of such effects include depressed cell formation and maturation, decreased viability of embryos, high resorption rates, 260
or delayed tissue maturation (such as slow rates of oss if ication in neonates). Reports of vertebral hyper trophy in treated fish (Couch _ al., 1978) and variations in skeletal development in mice (Beck, 1977 ~ are not evidence of mitotic spindle effects and do not support the theory of a mutagenic effect of trifluralin in mammalian systems. Evaluation of hematologic values from chronic toxicity studies also did not elicit any such evidence (Mauer, 1978~. Thus, several lines of evidence from both the plant and animal k ingdoms suggest that tr if luralin products, containing known or unknown levels of NDPA, can interfere with the cell division spindle. Mammalian somatic or germinal cells have not been studied, but mammal fan cells would probably respond similarly to cells of other organisms. The existing data regarding effects on the cell division spindle at estimated trifluralin or NDPA exposure levels are inadequate for an assessment of risk. Furthermore, it is not clear whether tr if luralin itself, one of its metabolites, or a contaminant is the active cause of the noted ef feats. Additional studies are needed to clarify these uncertainties. 261
CONCLUSIONS Extensive mutagenicity testing has been performed with trifluralin, with positive results in come instances. mese results are tainted because they were obtained using undefined trifluralin or trifluralin contaminated with the known n~utagen~carcinogen NDPA at levels as high as 177 ppm. Therefore, it is quite possible that the mutagenicity (and carcinogenicity} attributed to trifluralin could be related to the contaminant. NDPA-free trifluralin abould not be considered mutagenic {or carcinogenic} until a carcinogenesis bioassay on NDPA-free trifluralin is performed for comparison with the existing NCI study. The majority of the mutagenicity studies with trifluralin produced Begat ire results . Those that were pos itive for chromosomal damage and aneuploidy may be due to the presence of NDPA. Parallel studies with NDPA have not been reported. NDPA-free trifluralin and pure NDPA need to be tested, in tandem, for the ir abilities to induce cbromosomal damage and aneuploidy. Examination of the mutagenicity data on trifluralin preparations (including the formulated product) con ta ining NDPA reveals an inadequate data base on which to evaluate potential hazards to DNA and spindles in laboratory animals or humans. NDPA may induce mutagenic effects, but the expected low exposures to this chemical suggest that the degree of hazard, even if NDPA are a germinal mutagen, are low. 262
REFERENCES Production, Uses, Exposure Chemical Marketing Reporter . 1979. P. 4 in September 3, 1979 issue. Schnell Publishing Company, New York. Chemical Marketing Reporter. adver se e f feats . 217 {11) :3, 32. 1980. EPA herbicide probe finds no Chemical Week. 1979. Union seeks ban on Eli Lilly herbicide. 125(20):28. Golab, T., and M. E. Amundson. 1975. Degradation of trifluralin, oryzal in and isopropyl in in soil . 3: 258-261 . Environ . Qua 1. Sa f . Suppl . Golab, T., W. A. Althaus , and H. L. Wooten . 1979 . Fate of [14C] tr if luralin in soil . J . Agr ic . Food Chem. 27: 163-179 . Hell ing, C. S. 1976. Dinitroaniline herbicides in soils. J. Environ. Qual. 5: 1-lS . 263 \
Horden, W. 1977. A Report on the Ose of the Macroreticular Resin XAD-2 to Isolate Pesticides from the Cape Fear River. University of Nor th Carolina, Chapel Hill, N.C. Pesticide and Toxic Chemical News. 1980. EPA requires more oryzalin toxicity studies; no FIFRA action taken. 8 (16) :33-34. Probst, G.W., T. Golab, R.J. Herberg, F.H. Holzer, S.J. Parka, C.C. Van der Schans, and J.B. Tepe. 1967. Fate of trifluralin in soils and plants. J. Agric. Food Chem. 15: 592-599. Spacie, A., and J.L. Hamelink. 1979. Dynamics of trifluralin accumulation in river fishes. Environ. Sci. Technol. 13: 817-822. SRI International. 1979. 1979 Directory of Chemical Producers: United States of America. Stanford Research Institute International, Menlo Park, Calif. 1122 pp. Taylor, A.W. 1978. Post-application volatilization of pesticides under f ield conditions. J. Air Pollut. Control. ASSOC. 28: 922-927. U. S . International Trade Commission. 1978. Synthetic Organic Chemicals . Un ited States Production and Sales, 1977. ISI]C Publication 920. U.S. Government Printing Office, Washington, D.C. 417 pp. 264
Wnite, A.W., Jr., L.A. Harper, R.A. Leonard, and J.W. Turnbull. 1977. Trifluralin volatilization losses from a soybean field. Environ. Qual . 6: 105-110 . Woodrow, J.E., D.G. Crosby, T. Mast, X.W. Moilanen, and J.N Seiber. 1978. Rates of transformation of trifluralin and . parathion vapors in air. J. Agric. Food Cbem. 26:1312-1316. 265 J.
Analytic Methods Downer, G.B., M. Mall, and D.N.B. Mallen. 1976. Determination of benefin and trifluralin residues by quantitative gas-liquid chromatography/mass spectrometry. J. Agric. Food Chem. 24:1223-1225. ~ Food and Drug Administration. 1973. Pesticide Analytical Manual. Heck, H.d'A., R.L. Dyer, A.C. Scott, and M. Anbar. 1977. Determination and disposition of trifluralin in the rat: Separation by sequential high-pressure liquid chromatography and quantitation by field ionization mass spectrometry. J. Agric. Food Chem. 25: 901-908. Lawrence, J.F. 1976. Gas chromatographic separation of herbicides of major interest in Canada, with electrolytic conductivity detection in the nitrogen and chlorine modes. J. Chromatogr. 121: 85-87. Lawrence, J.F., D. Lewis, and H.A. McLeod. 1977. Confirmation of some NO2- containing pesticides by chemical reduction and gas chromatography with electrolytic conductivity detection. J. Agr ic. Food Chem. 25 :1359-1361. Payne, W.R. , Jr. , J.D. Pope, Jr. , and J.E. Benner. 1974. An integrated method for trif luralin , diphenamid, and paraguat in soil and runoff from agricultural land. J. Agric. Food Chem. 22:79-82. 266
Smith, A.E. 1974. A aulti-residue extraction procedure for the gas chromatographic deteraination of the herbicides dichlobenil, dinitramine, triallate and trifluralin in ~oila. J. Chromatogr. 97:103-106. Soderquist, C.J., D.G. Crosby, X.W. Hollanen, J.N. Seiber, and, J.E. Wocdrow. 1975. Occurrence of trifluralin and its pt~otoproducts in air. J. Agric. Food Chem. 23: 304-309. Woodrow, J.E., D.G. Crosby, T. Mest, K.W. Mailanen, and J. N. Seiber. 1978. Rates of transformation of trifluralin and parathion vapors in air. J. Agric. Food Cbem. 26:1312-1316. 267
Health Effects Andersen, K.J., E.G. Leightly, and H.T. Takahashi. 1972. Evaluation of herbicides for possible mutagenic properties. J Agric. Food Chem. 20:649-656. Bartels, P.G., and J.L. Hilton. 1973. Comparison of trifluralin, oryzalin, pronamide, propt~am and colchicine treatments on microtubules. Pestic . 8iochem. Physiol. 3: 462-472 . Bartsch, H., C. Mala~reille, and R. Montesa no. 1976. The predictive value of tissue-mediated mutagenicity assays to assess the carcinogenic risk of chemicals. Pp. 467-491 in R. Montesano, H. Bar tech, and L. Tomatis, eds. Screening Tests in Chemical Carcinagenesis. (IARC Scientific Publication No. 12), International Agency for Research on Cancer, Lyon. Beck, S.L. 1977. Postnatal detection of prenatal exposure to herbicides in mice, using normally occurring var. iations in skeletal development. Teratology 15: 15A (Abstract). Bond, D.J., and L. McMillan. 1979. Meiotic aneuploidy: Its origins and induction following chemical treatment in Sordaria brevicollis. Environ. Healtb PersPect. 31:67-74. Brusick, D.J., and V.W. Mayer. 1973. New developments in mutagenicity screening techniques with yeast. Environ. Health Perspect. 6:83-96. 268
Chaisson, C.~. 1978. Summary of evidence on mutagenic potential of Treflan. Toxicology Branch, Office of Pesticide Programs, Environmental Protection Agency, Washington, D.C. Caisson, C.F., and T.D. Burkhalter. 1978. Mutagenicity considerations: Treflan. Toxicology and Plant Studies Branch, Off ice of Pesticide Programs, Environmental Protection Agecy, Washington, D.C. Couch, J.A., J.T. Winstead, D.J. Hansen, and L.R. Goodman. 1978. Vertebral dysplasia in young fish exposed to the herbicide trifluralin. Contribution No. 346, Gulf Breeze Environmental Research Laboratory, Environmental Protection Agency, Gulf Breeze, Fla. Environmental Protection Agency. 1979 . Tr if luralin (Treflan Position Document. Office of Pesticide Programs, Special Pesticide Review Division, Washington, D.C., August 29, 1979. Epste in , S . S ., E. Arnold, J. Andrea , W. Bass, and Y. Bishop. 1972. Detection of chemical mutagens by the dominant lethal assay in the mouse. roxicol. Appl. Pharmacol. 23: 288-325. Griffith~, A.J.F. 1979. Neurospore prototroph selection System for studying aneuploid production. Environ. Bealth Perepect. 31: 75-80 . 269
Jackson, W.T., and D.A. Stetler . 1973. Regulation of mitosis. IV. An in vitro and ultrestructura1 study of effects of trifluralin. Can. J. Bot. 51:1513-1518. Kruger, F.W. 1973. Metabolism of nitrosamines in vivo. II. On the methylation of nucleic acids by aliphatic di-N-alkyl-nitrosamines in viva caused by ~-oxidation: me increased formation of 7-methylquanine after application of 8-hydroxypropyI-propyl-nitrosamine compared to that after application of di-n-propyl-nitrosamine. Z. Rrebeforech. Elin. Onkol. 79:90-97. Kuroki, T., C. Drevon, and R. Montesano. 1977. Microsome- medisted mutagenesis in V79 Chinese hamster cells by various nitrosamines. Cancer Res. 37:1044-1050. Matsuoka, A., H. Hayashi, and H. Ishidate, Jr. 1979. Chromosomal aberration test. on 29 chemicals combined with S9 mix in vitro. Mutat. Res. 66: 277-290 . Mauer, I. 1979. Treflan: Rebuttable Presumption Against Registration Document: Mutagenicity Risk Assessment. Toxicology Branch, Office of Pesiticide Programs, Enviornmental Protection Agency, Washington, D.C. 270
Mauer, I. 1978. Treflan: Summary of Scientif ic Evidence. Toxicology Branch, Off ice of Pesticide Programs, Environmental Protection Agency, Washington, D.C. McCann, J., E. Choi, E. Yamasaki, and B.N. A - es. 1975. Detection of carcinogens as mutagens in the Selmonella/microsome test: Assay of 300 chemicals. 72: S13 5-5139 . Proc. ~tl. Acad. Sci. U.S.A. Montesano, R., and H. Bartscb. 1976. N-n i trove compounds: Res. 32: 179-228 . Murnik, M. R. 1978. Mutagenic and carcinogenic Possible environmental hazards. Mutate Mutagenicity of the herbicide trifluralin in Drosophila melanogaster. Mutat. Res. 53: 23S-236 {Abstract No. 149) . Naka jims, T., A. Tanaka, and R. Tojyo. 1974. me effects of metabolic activation with rat liver preparations on the mutagenicity of several N-nitrosamines on a streptomycin-dependent strain of Escherichia colt. 26: 361-366. Mutat. Res. Nehez, M., A. Paldy, A. Selypes, M. Koro';falvi, I. Lorinczi, and G. Berencsi. 1979. The mutagenic effect of trifluralin~containing herbic ide on mouse bone marrow In viva. Ecotoxicol. Environ . Saf . 3: 454-4S7 . 271
Nelson, J.O., P.C. Kearney, J.R. Plimmer, and R.E. Henzer. 1977. Metabolism of trifluralin, profluralin, and fluchloralin by rat liver microsomes. Pestic. Biochem. Physiol. 7:73-82. Olejos, E.J., and H.H. Cornish. 1976. Mutagenicity of dialkyl- nitrosamines: Metabolites and derivatives. Toxicol. Appl. Pharmacol. 37 :109-110 (Abstract No. 43) . Propping, P., G. Rohrborn, and W. Buselmaier. 1972. Comparative investigations on the chemical induction of point mutations and dominant lethal mutations in mice. Mo1. Gen. Genet. 117:197-209. Russell, L.B. 1977. Validation of the in vivo somatic mutation method in the mouse as 8 pre screen for germinal point mutations. Arch. Toxicol . 38: 75-85 . Sawamura, S., and W.T. Jackson. 1968. Cytological studies ~n vivo of picloram, pyriclor, trifluralin, 2,3,6-TBA, 2,3,5,6-TBA and nitralin. Cytologia 33: 545-554. Seiler, J.P. 1972. Mutagenicity of benzimidazole and benzimidazole derivatives. I. Forward and re~rerse mutations in Salmonella typhimurium caused by benzimidazole and some of its derivatives. Mutat. Res. 15: 273-276. &ntein, P. 1977. Trifluralin, an inhibitor of the achromatic apparatus which damages the chromosomes. Arch. Anat. Microsc. Morphol. Exp. 66:263-277. 272
Shirasu, Y., M. Moriya, K. Kato, A. Furohashi, and T. Kada. 1976. Mutagenicity Screening of pesticides in the microbial system. Mutat. Res. 40:19-30. Simmon, V.F., A.D. Mitchell, and T.A. Jorgenson. 1977. Evaluation of selected pesticides as chemical mutagens. ~ In visor and ~ in vitro. studies . Prepared by Stanford Research Institute, Menlo Park , Calif ., for the Environmental Protection Agency, Health Effects Research Laboratory, Research Triangle Park, N.C. Report No. EPA-600/1-77-028. Available from National Technical Information Service, Spr ingf ield, Va ., as PB-268 647. 251 pp. Yahag i , T ., M. Nagao , Y. Se ino , T. Ma tsush ima , T. Sug imura , and M. Okada . 1917 . Mutagen ic ities of N-n itrosamines on salmonella . Mutat. Res. 48: 121-130. Yoder, J., M. Watson, and W.W. Benson. 1973. Lymphocyte chromosome analysis of agricultural workers during extensive occupational exposure to pesticides. Mutat. Res . 21: 335-340 . 273 .
