Document zzb6ENKx2mzjg5w9e9qbvb3n0

R&S 105708 BlO-HZDICAL RZSEARCH DOCUMENT DESCRIPTION FORK 63 6S 69 0 0 0 0 0 0^ Duplicate in all cards:--> year as-1961- Pile number [Right justify [Numeric only] Author(s), as Last Naas PS (No Punctuation) and coden for journal as JAKA preceeded-by. one blank space 1 ` 20 "21 nV~, As)<zi-*lizjZ' n. 21 / /-Ai'n/<7 t? (1 TFf-r 1 - e.4--^L 40 41 1 1 1 77 78 ~pzn Sub-Index: Code 60 61 62 n 12 13 ts tie of Renort; end with space-h>*phen-hyphen-space- Follow with Index Terns , *n=------------------ *----;--_ -- * Lon; do not abbreviate. 12 * 1 />U '"'N . * I ' / / C: J )o/>i .< n nr /%.<cI 71 /' t:-/ 'x rj ,'J C)-/- Ltr//.r^ ' ' /.-'/>! _r / //) - 61 '62 21 22 23 24 Source (Journal, Vol., Number, Pages, Date) 12 T'O i-<jT gjtt / p r\ ........... ` ' Yo, ,>,& 333 ~ 3 7o -t ^777; / "7 7Cj 61 62.' 31 32 Srief Str--~ry' 12 10 SUMMARY: A_______ _: - 61 62 61 62 63 64 0000002 359 Mutation Research, 40 ()976) 359--370 Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands VINYL CHLORIDE: DOMINANT LETHAL STUDIES IN MALE CD-I MICE 3J Co ' CO *05709 DIANA ANDERSON. MICHAEL C.E. HODGE and IAIN K.H. PURCHASE Imperial Chemical Iivluslrics Lid. Central Toxicology Laboratory, A Icierley J'url;, AV Mocclcsf:cld, Cheshire SK10 -ITd (England) (Received February 17th, 1970) (Revision received May 19th, 197G) (Accepted June Slh, I97G) Summary The mutagenic activity of vinyl chloride (VC) at three exposure levels was as sessed in ferHle mate CD-I mice with the dominant lethal test. Male mice were exposured by inhalation to VC at 3000. 10.000 and 30.000 ppm for 6 h a day for 5 days. By comparison with control males exposed to air, no mutagenic effects on any maturation stage of spermatogenesis in treated males were detected. There was no significant increase in the number of postimplantational early foetal deaths as shown by the number of females with one or more early deaths or number of early deaths/pregnancy or the number of early deaths/total implants/prepiancy. There was no evidence of pre-implantational egg looses as indicated by the total implanis/pregnant female. There was also no reduction in fertility. The lack of effect was not due to the insensitivity of the system used since a dominant lethal effect was clearly demonstated in male mice dosed i.p. with cyclophosphamide (CTX) at 200 mg/kg body weight and ethyl mcthanesulphonatc (KMS) orally at 200 mg/kg'body weight once a day for 5 days. During dosing these animals were housed under similar exposure conditions to those animals exposed to the test substances but with a flow of air through the ex posure chambers. Thus vinyl chloride is not mutagenic in the mouse at the stated exposure lev els as measured by the dominant lethal test. Introduction Vinyl chloride (VC) used in the manufacture of polyvinyl chloride has been found to cause tumours in rats [12] and man [6]. It has also been shown to R&S 105710 i i i 360 produce chromosome brooks in exposed workers [7,10.13] and to cause muta tions in Salmonella typhimirium [5]. We therefore carried out a dominant le thal study to determine if there were any mutagenic effects of this type in mice after vinyl chloride exposure at 3 different levels. At the same time, negative control animals, exposed to air, and positive control animals, also exposed to air and given ethyl methanesulphonate and cyclophosphamide, were assayed. Materials and methods Chemicals VC was obtained from Air Products Limited, Worsley, Walkden, Lancs, and mixed with air which was metered from the compressed air taps in these labora tories. Kthyl methancsulphonutc (EMS) was obtained from Koch-Light Lim ited, Colnhrook, Rucks, England, and cyclophosphamide B" (Hndoxana) was obtained from Ward Blenkinsop Limited, London. England. Both of these sub stances were prepared as aqueous solutions immediately before use. Animals CD-I mice (Charles River, Manston, Kent) were used throughout the experi ment. Undosed females were S--10 weeks old when mated and male mice im mediately after dosing were 10--12 weeks old. Males were caged individually and females in pairs. They received food and water ad libitum. All females were examined daily during the experiment for evidence of clinical or behavioural abnormalities. Dosing of male mice Dose levels for VC were selected on the basis of toxicity studies. Groups of 6 mice were given 5 daily 6 h exposures of VC at 30.000. 20.000. 15.000. 10.000 and 5000 ppm. Doses of 30.000. 10.000 and 3000 ppm were chosen for the main study. 30,000 ppm was chosen as the highest exposure level as this was found to be in the toxic range and it was desirable that the maximum tolerated dose or higher should be used. The required concentrations were generated by mixing known volumes of VC and air using rotameters as indicators. During dosing the mice were housed individually in compartments in cham bers made of stainless steel and glass with an internal capacity of 3 litres and they received food and water in agar cubes. All groups, including positive and negative controls, were housed under simi lar conditions during the 5-day treatment period with a flow of air (mixed with the appropriate percentage of compound, where necessary) through the cham bers. Groups of between 15 and 25 mice (numbers shown in parentheses below) of proven fertility were treated in the following ways immediately before test mating began. Group 1 (20 mice), air. Group 2 (20 mice), 3000 ppm VC for 6 It per day for 5 days. Group 3 120 mice), 10.000 ppm VC for 6 h per day for 5 days. Group -1 (20 mice), 30.000 ppm VC for ( h per day for 5 days. Group 5 (15 mice), 200 mg cyclophosphamide in water per kg body weight once by i.p. injection on day 5. R&S 105711 36 Group 6 (25 mice), 200 mg ethyl methanesulphonate in water per kg bodj weight orally once a day for 5 days. Mating Fertility testing 200 male mice were caged with `100 virgin female mice, 1 male and 2 female mice in each cage. After 5 days die females were transferred to other cages. The female mice were killed fifteen days after first introducing them to the males and examined for pregnancies. 106 males which survived dosing and were suc cessful in fertilising at least 1 female in their cage were selected for continua tion in the experiment. i 5 Experimental mating 2 virgin female mice 8--10 weeks old were put into each of the 106 cages in which the males were individually housed. After 5 days, the females were re moved and rehoused in pairs. A week after the initial introduction the males were caged with another two virgin females and again left for 5 days. This process was repeated until the treated male mice had been mated at weekly intervals for 8 weeks with virgin, females. The males were then killed and not examined further. No attempts were made to establish whether mating had occurred. Instead, it was assumed that most matings leading to fertilisation would occur 2 or 3 days after intro ducing female mice to the cages containing males. Female mice were killed 13 days after the assumed date of fertilisation, i.e. 15 or 16 days after caging females with males. Assessment Uteri of the killed mice were examined for live implantations, early deaths and late deaths. Statistics The data have been statistically analysed as reported previously [1). Results Mating weeks after treatment are represented in Tables by numbers 1--8 and the mating week before treatment is represented by week 0. Assessment of fe males which became pregnant during the fertility test yielded the data for week 0 but only data from those animals that survived treatment have been included ' in week 0--S. Survival data The number of males surviving treatment are shown in Table I. The only sig nificant mortality occurs in the groups exposed to the highest level of VC. Successful mating frequency The numbers and percentage of males successfully mating at each week is Zi-ZSOJ- S9U tAbLet NUMBER AKIJ PERCENTAGE OF MALE falCfc v/HlCff SORViVED TIlE ATMENT S UCCkSf'ULLY MATING AT RACK WEEK Group i Air Group 2 VC 3000 ppm (G h X&) Group 3 VC 10,000 ppm {0hX61 Group 4 VC 30,000 pmm (fi hX5) Group 5 CTX 200 mc/ki? i.p. Week 0C 1 2 3 4 5 6 7 8 No. of mice % 20 100 20 100 20 100 20 100 20 100 19 95.0 20 300 20 100 20 100 No, ci ( mice 18 18 18 18 18 18 18 18 18 % 100 100 100 100 100 100 100 100 100 No. of mice 19 1 19 19 19 19 19 19 19 % 100 100 100 100 100 100 100 100 100 No. of mice 9 7 9 9 9 7 9 0 8 % 100 77.8 100 100 100 77.8 100 100 88.9 No. ot mice 15 14 14 14 15 15 No Is a 14 14 1 nil b dead 13 % 100 93.3 93.3 93.3 100 100 1 DO 93.3 92.9 % Survival ioo% (20/20) *fl*t treatment 90% (1 B/20) 05% (19/20) 45% (9/20) 100% (15/15) iwccpt at wn-k A No fs, females were not caned with males, ^ nit, male, c Data in week 0 only refers lo animals surviving treatment. Group G EMIj 200 infi/kt oral X 5 No. of m ice % 25 3 23 23 25 25 No fs a 24 25 25 100 12.2 92.0 92.0 100 100 100 100 100 toor-i (25/25) I l t I i TABLE II NUMBER OF MATED FEMALES BECOMING PREGNANT Group 1 Air Group 2 VC 3000 ppm (C h *5) Group 3 VC 10.000 ppm <0 l\ X5) Mo. pre&. No. mated No. prvg. No, mated No. pita. No. muted Week 0 35 40 30 30 28 38 1 35 40 33 36 32 No fs a 2 au 40 30 30 33 3ii 3 34 40 30 3G 34 38 4 as *10 34 36 33 38 b 33 40 34 36 30 as 6 36' 40 33 3G 36 38 7 37 S 37 40 40 31 36 37 32 36 31 38 1 f. dead 37 * No Is, females were not cased with males. ** ns!, male. c /' < o.oot, dP< o.ot. Group 4 VC 30.000 ppm (G h X 5) No, pres. No, mated 15 18 11 18 15 18 1G 18 17 18 12 18 18 18 18 18 14 18 Group 5 CTX 200 mc/kc i.p. G roup G KMS 200 m*:/kn or.il X 5 No. preii. No. mated No. Prcu. No. mated 25 30 37 GO 26 30 4 c 50 25 30 31 d GO 2G 30 40 50 30 30 46 50 28 30 47 GO 20 No Is a 47 No Is a 28 48 22 30 41 GO 23 1 ml " dead 4 5 50 28 363 etzsot s*u 364 also shown in Table I. Numbers remained high during the experiment. No statisLicnlly significant differences in the mating frequency were found between VC treatment groups and the control at any week using a Chi-squared test. There was, however, a significant difference between the EMS-treated group and the control in week 1. Pregnancy frequency The numbers of females in each group which became prepiant at each week of mating is shown in Tabic 11. Statistical differences between treated groups and the control group were only found in the EMS treated group at weeks 1 and 2 using a Chi-square test. These results indicate that VC does not decrease fertility at the administered doses. Total implantations The total number of implants per pregnant female in each group is shown in Table III. The mean values were adjusted to take account of the unequal num bers of pregnant females per male and were compared statistically at each week using an analysis of variance and a Dunnett's `t' test. There were statistically significant differences between the positive control groups (5 and 6) and the negative control. The differences were evident in week 1 in the cyclophos phamide-treated group and weeks 1 and 2 in the EMS-treated group. A signifi cant difference (P < 0.05) in week 4 was also found between the group ex posed to the highest dose of VC (Group 4) and the negative control. Early deaths Early deaths are considered to be important in the assessment of a dominant lethal effect [4], and there are various ways in which the results may be evalu ated, each producing some bias. R&S 105714 TABLE III average total implants per pregnant female Group 1 Air Group 2 VC 3000 ppm (6 h X5) Group 3 VC 10,000 ppm (G h X5) Group 4 VC 30,000 ppm (6hX5) Week O 1 2 a 4 5 6 7 a 11.90 12.03 12.es 12.50 12.3S 13.68 11.75 11.80 12.68 12,G 1 13.56 12.G7 12.53 12.14 12.72 11.94 11.56 12,14 11.07 13.36 13.03 12.30 1 3.GG 14.13 12.50 1 1.76 12.56 13.05 12.68 13. G7 11.16 10.00 3 12.79 12.56 11.94 12.13 1 / < 0.05. b /' < 0.01. Group 5 CTX 200 mg/kg i.p. 11.33 8.82 b 10.93 10.92 1 2.60 12.30 12.46 1 2.50 12.46 Group 6 KMS 200 mu/kn oral X5 11.78 4.00 b 8.43 b 12.50 12.68 12.66 12.10 12.36 12,52 w TABLE IV NUMBER OK PREGNANT FEMALES WITH 1 OK MOHR EARLY DEATHS Group 1 Air Group 2 VC 3000 ppm CG h X 5) Group 3 VC 1 0,000 ppm (G h X 5} ED * 0 >1 0 >1 0 >1 Week 0 18 17 21 0 Iti 22 1 15 20 1G 17 10 21 2 10 17 IB 18 13 20 3 18 1C 1G 14 10 10 4 1 C> 22 17 17 20 13 & 17. 18 18 10 14 22 6 17 10 11 22 21 15 7 18 10 13 18 21 1G 8 13 24 14 18 15 1C a ED, early diMlIis. b P < 0.01. Group 4 VC 30,000 ppm <R h X&) 0 >1 10 5 G5 G9 9G 97 39 7 11 8 10 8G Group 5 CTX 200 mg/ke i-P- 0 >1 22 3 ^ 1 25 b 2 23 b 8 18 14 1G 12 1G 13 13 10 12 13 10 Group G EMS 200 mfc/ku otjI X & 0 >1 21 26 13 r> 2G b 15 25 19 27 25 22 24 23 22 19 22 23 stZ90t ssy tu & cn r 3GG The number of pregnancies with one or more early deaths (Table TV) Cyclophosphamide and EMS treatment caused increases in the number of pregnancies with early deaths. The effect was significant (Chi-square) in weeks 1 and 2 for the cyclophosphamide-treated group and in week 2 for the EMStreated group. Xo differences from the control group were seen in the VC-ireated group. Number of early deaths per pregnancy (Table Va) The number of early deaths per pregnancy was only increased in the cyclo phosphamide and EMS-treated positive controls. A statistical analysis of the data is shown in Table Vb. A large number of low or zero values was encoun tered so it was necessary to stabilise the variance prior to analysis. The variance ratio test on the transformed data showed that groups 5 and 6 were different from group 1. Differences were evident in weeks 1 and 2 with cyclophosphamide treatment and week 2 with EMS treatment, whereas VC treatment induced no increase in early deaths. TABLE Va average early deaths per pregnant female Group 1 Air Group 2 VC 3000 Ppm (6 h X 5) Group 3 VC 10,000 ppm (6 h X5) Croup 4 VC 30.000 ppm (G h X5) Week 0 1 2 3 4 5 G 7 8 0.77 0.86 0.83 0.91 0.89 1.06 0.97 0.76 1.03 0.33 0.91 1.00 0.73 0.79 0.71 0.91 0.87 0.88 0.54 1.00 1.15 0.88 0.61 1.03 0.75 0.68 1.00 0.40 0.4 5 0.93 0.93 0.69 1.00 1.39 0.89 0.71 Group 5 CTX 200 m:/kc i.p. 0.16 4.27 4.8i 1.54 1.13 1.11 0.65 0.62 0.70 Group 6 K MS 200 mc/kt oral X 5 0.59 3.50 2.58 1.45 0.85 0.72 0.85 0.63 0.73 TABLE V5 TRANSFORMED EARLY DEATHS PER PREGNANT FEMALE Group 1 Air Group 2 VC 3000 ppm <6 hX5) Group 3 VC 10,000 ppm (6 h X5> Group 4 VC 30.000 ppm (6 h X5) Week 0 1 2 3 4 5 6 7 8 1.92 2.08 1.9 2.02 2.05 2,10 2.00 1.89 2.19 1.49 2.08 2.02 1.89 1.92 1.82 2.09 2.04 2.07 l.<>4 2.21 2.21 2,01 1.61 2.10 1.90 1.76 1.99 1.59 1.61 2.1 2.0S 3.7 2.32 2.30 1.99 1.85 Group 5 CTX 200 mg/kg i.P. 1.66 4.18 * 4.57 1 2.52 2.16 2.16 1.76 1.91 1.90 Group 6 EMS 200 mc/k* oral X & 1.74 3.06 3.28 a 2.38 1.99 1.S1 1.91 1.69 1.85 R&S 105716 R&S 105717 367 TABLE Via early deaths as a percentage ok total implants per group Group 1 Air Group 2 VC 3000 ppm (0 h X5> Group 3 VC 10.000 ppm (6 h X5) Group 4 VC 30,000 ppm (6 h X 51 Croup 5 CTX 200 mg/kjt Week 0 1 o 3 4 5 6 7 3 G.49 6.(35 6.41 7.35 7.23 7.81 8.16 6.36 3.02 2.67 G.67 7,89 5.87 6*55 5,54 7.50 7.34 7.18 4.42 7.49 9.57 6*99 4.48 7*28 6.02 5,66 8,09 3,09 2,69 6.90 7.82 6.55 7.95 1 l.OG 7.44 5.88 1.42 49.12 44.65 14.08 8.99 9.06 5.21 6.98 5.67 Group 6 fclMS 200 mg/kg oral X 5 5.08 77.73 31.50 11.58 6.70 5.73 7.04 5.10 5.92 TABLE VTb TRANSFORMED EARLY DEATHS PER TOTAL IMPLANT PER PREGNANT FEMALE Group 1 Air Week O 1 2 3 4 5 6 7 8 0.55 0.59 0.55 0.58 0.58 0.57 0.59 0.55 0.62 Group 2 VC 3000 ppm (G h X5) Group 3 VC J 0,000 ppm (6 h X 5) 0.41 0.56 0.57 0.54 0.56 0.51 0.60 0.69 0.59 0.46 0,61 0.65 0.57 0.43 0.57 C.5'1 0.51 0.56 Group 4 VC 30,000 ppm (6hX5) 0.4 5 0.44 0.59 0.71 0.55 0.CG 0.6G 0.59 0.53 Group 5 CTX 200 me/Kg i.P. 0.35 * 1.5G b 1.40 b 0,78 0.61 0.63 0.49 0.56 0.55 Group 6 EMS 200 mc/kg oral X 5 0.50 1.64 b 1.25 b 0,68 0.57 0.51 0.56 0.50 0.52 * P < 0.05. b y < o.oi. Percentage of total implants recorded as early deaths per pregnancy (Table Via) Again it was necessary to stabilise the variance prior to analysis (Table VIb). The variance ratio test on the transformed data showed that groups 5 and 6 were different in weeks 1 and 2 from the negative control, whereas VC-treated groups were not. Group 5, however, also showed a difference from the negative control in week 0 (week 0 being the pre-treatment week). Percentage induced pre- and post-implantational dominant lethality These values are shown in Table VII. The typical variation of the data falls between --14 and +12. Only groups 5 and 6 show large values in weeks 1 and 2 and group 5 also shows a slightly increased value in week 3. A similar value is also seen in week 4 for the highest VC exposure group. The remaining values lie within the typical variation range suggesting that VC shows no induced pre- and post-implantational dominant lethality.  30 cn O cn -A 00 368 TABLE VII PERCENTAGE INDUCED PRE- AND POST-IMPLANTATIONAL DOMINANT LETHALITY Group 2 VC 300a ppm (G h X5) Group 3 VC 10.000 ppm (G h N5) Group 4 VC 30.000 ppm (6hXS) Group 5 CTX 200 mc/kj; i-P. Group 6 EMS 200 orial X5 Week 0 -10.3 -3.0 i --4.8 --2,4 2 + 2.9 + 1.2 , 3 --1.8 +0.7 4 + 1.2 -13.6 s +4,8 -3.8 s --2.3 -9.0 7 + 2,3 +3.3 8 +3.3 +0,3 -13.7 --6.5 --G.O + 11.7 + 19.0 +6.6 -3.6 --0.1 + 2.0 --0.4 +62,3 +4*\3 + 19.1 +0.2 + 11.3 --{) _5 + 3.3 -0,9 -0.5 +9 5.9 +51.3 +4.7 -2.1 + 5.4 --4.4 --6.3 -1.2 Late deaths Late deaths are not as important as early deaths in the assessment of the mu tagenic potential of a test substance [4]. Late deaths were randomly distrib uted throughlout the groups and did not appear to be treatment-related. This would suggest that the induction of late deaths is independent of the males and that males and females contribute to the effect which is not increased by treat ment of the males cither with the mutagens or test substance. Conjoined placentae were seen in this strain. They were the result of very close implantation sites and were not monozygotic twins [3], In this experi ment they were classified as double implantations. Since they occurred with equal frequency in all groups, they did not appear to be correlated with treat ment. Discussion The best indication of mutagenic activity of a substance in the dominant le thal test is an increase in the number of post-implantational foetal early deaths [4]. From the early deaths data there was no evidence in this study of a muta genic effect with VC at the administered exposure levels. This did not appear to be a result of lack of sensitivity of the animals used, since there was a marked response to CTX and EMS. Different evaluation methods were used for the early death data because there is some conflict concerning the commonly-used mutagenic index (early deaths or dead implants/total implants). Epstein et ah [9] 'nave reported a lack of statistical correlation between dead implants and total implants using a weekly mean, whereas Green and Springer [11] do find a correlation, although not a strong one. (The positive correlation indicates that dead implants per fe male should be adjusted by the total implants for that female -- that is essen tially.what the index does.) The U.S. Food and Drug Administration has ana lysed all data in dead implants in two ways (both on dead implants/total implanls/femnle and dead implanls/femalc basis) and infers that equally signifi cant data are pruvided by both [11]. Since there are no standard methods of R&S 105719 CoO analysis both methods have been included in this report, but in an attempt to increase the sensitivity, early deaths only were considered and late deaths were not included. A third method was also used (number of females with one or more early deaths). With ail three methods of evaluating the results the lack of effect of VC was consistent. This was in spite of the fact that exceptionally high levels of VC were used in the* present study where only 9 out of 20 ani mals survived at the top exposure level. Pre-implantation egg losses, whilst representing some of the mutagenic ef fect, are not as important as the post-implantational losses [4]. Pre-implanta tion egg losses in the present study have been considered on their own or pooled with post-implantational losses. Pre-implantation egg losses have been indicated by comparing values of total implants in females mated with treated males and those mated with control males, as suggested earlier [5] (as opposed to counting corpora lutea). There was no pre-implantation egg ioss by compari son of treated groups with the control except at the highest VC exposure group in week 4. This loss was at a low level of significance (P< 0.05). When a posi tive control is significant there is a uniform reduction of implants, whereas in the highest VC exposure group in week 4 the low result was due to a large ex tent to the result from one female. Without this female the mean value for total implants per pregnancy would be 11.19 which is not significantly different from the control, and for percentage induced pre- and post-implantational dominant lethality would be S.7 which lies within the range of normal varia tion. Therefore the results in group 4 week 4 are not biologically significant. . There was no reduction in fertility as measured by the mating and pregnancy frequency at any week at these exposure levels of VC, suggesting no antifertiiity effect. Much of the data was subjected to various methods of statistical analysis, in cluding an analysis of variance. When an analysis of variance is performed, it is assumed that the data are normally distributed, variances are equal in all groups and results are independent. The data for early deaths for individual females had a skewed distribution since there were many females with zero early deaths, so a variance'stabilising transformation was used both for the method of determining early deaths/prognancy and early deaths/total impiants/pregnancy. This helped to reduce the skewedness of the distribution of early deaths for individual females. Since the male was the test organism (and the female the indicator organism) the necessary independence of results was provided by tak ing the male as the basic- unit. The statistical analysis then involved estimating missing values for non-pregr.ant females and using the "between male" variabili ty for assessing statistical significance. The only criterion which was not com pletely satisfied was the normality of the distribution. An analysis of variance however, is relatively robust to departure from normality, whereas it is not so to departure from independence. A fertility study was undertaken prior to the experiment because previously unmated males react differently in the first week of mating when compared with the remaining weeks of the study [11] and also to ensure a high fertility throughout the remainder of the experiment. From these initial fertility data the background dominant lethality of all groups was determined, primarily to ascertain that thei'c were no initial differences between groups. 370 The purpose of the high doses of CTX and EMS used in the present study was to obtain a highly significant positive result. However, the dominant lethal assay in our hands is at least as sensitive as ihat reported for other studies. We have shown EMS to give a positive result in a single i.p. dose of 150 mg/kg body weight which is comparable to that reported previously for the same mouse strain f 14] and HN, to give a positive result where it has previously been reported negative [9]. In conclusion, therefore, whilst mutagenic effects of vinyl chloride, as mea sured by chromosomal aberrations in exposed workers, have been reported [7, 10,13], no such effects as determined in the present study occur in the germ cells of male CD-I mice. Acknowledgment The authors wish to thank Mr. S.H. Ellis, Pharmaceuticals Division. Mereside, for his statistical evaluation of the data, Mrs. S. Palmer, Central Toxicology Laboratory for her technical assistance, and the Inhalation unit. Central Toxi cology Laboratory for exposing the animals to vinyl chloride. References 1 11 1 Anderson, I),, D.ll, McCrcitor, I.F.ll, Purchase, Doniinanl lethal studies with paraquat and diquat in, male CD-I mice. Mutation Re*,, *10 (1976) 349--358. 2 Anderson, D., D.B. McGregor* I.K.H. Purchase, M.C.E. Hodge, J.A. Ciuhbcrt, Dominant lethal test re* sult't with known mutanons in two laboratories (unimhlshed). 3 Hjtciuuu, A.J.. Die ho rial ouc-vytg t vv ms in l hi* mouse. Nature, 187 (I960) 339--3-1C. 4 Bateman, A.J, and S.S. Epstein, in A. HoJIaender (cd.) ChcmicaJ Mu tasens, Principles and methods for their detection. Plenum Press, New York -- London. 1971, pp. 541--5G8, 5 B&rlsch, H., C. MaUvcillc and R. Montesano, Human, rat and mouse liver mediated mutagenicity of vinyl chloride in ,*?. typhimunum strains, Int. J. Cancer, 15 (1975) 429--437. 6 Creech. J.L. and M.N. Johnson. Angiosarcoma of the liver in the manufacture of polyvinyl chloride. J. Occup. Med., 16 (1974) 150-151. 7 Ducatman, A.. E. Hirschom and I.J. Svlikoff, Vinyl chloride exposure and human chromosome aber* rations. Mutation Res., 31 (1975) 1G3--1G8. 8 Epstein, S.S., The use of the dominant-lethal test to detect genetic activity of environmental chemi* cals* Environ, llllh. Perspectives, 6 (1973) 23--27, 9 KpsUni, S.S., E. Arnold. J. Andrea, W. Bass and Y, Bishop. Detection of chemical mutagens by the dominant lethal assay in the mouse, Toxicol. Appl, Pharmac.. 23 (1972) 288--325. 10 Funcs-Cravinto, E. Lambert, J. Lindsten. L. F.htvnbcrg* A.T. Njtarajan and S. Osterman-Golkara, Chromosome aberrations in workers exposed to vinyl chloride, Lancet, 1 (1975) 459. 11 Green, A. and J.A. Springer, Potential limitations and statistical considerations for safety evaluation. Environ. Hlth, Perspectives. 6 (1972) 37--46. 12 Majrnni, C., Ci, Lrfenunc. P. Chieco and D. Curreu. Vinyl chloride carcinogenesis -- current results and perspectives, Med. Lavaro, G5 (l 1-t 2) (1974) 421--444. 13 Purchase, I.F.H., C.R. Richardson and D. Anderson, Chromosomal and dominant lethal effects of vinyl chloride. Lancet, 11 (1976) 410--411. 14 Ray, V.A. and M.L, Hyneck, Some primary considerations in the interpretation of the dominant* lethal assay. Environ. HHh. Perspectives, 6, 23 (1973) 27--35. R&S 105720