Document zdVZ6DY65VgOvKJ3Bv5kRKL16

AKIKUIK 1..4 The definitive effect* of pesticide* on avian liver micro* ftnnmi enzyme* in general have yet tn be described and the literature presents conflicting finding*. Although Ste phen ami emtorkers (11171 > found an increase in sleeping time in chicken*, they were unable to show any consistent change in liver microsomal enzyme activity. Controry to finding a decrease in liver microsomal activity. AhemDonia ond Menzel UPGR) found that liver microsomal oxi dative octivity wits increased in chirks hatched from egg* which had been injected with OUT. Pcakall (1967, 1970) observed an increase in the rate of steroid metabolism by hepatic microsomes of pigeons or doves treated with DDT, also suggesting a stimulation of liver microsomal activity. However, reduction in liver enzyme activity has also been reported. Sell ct al. (1971) found reduced hepatic hydrosyloso octivity after feeding DDT to While Leghorn hens or JapatiCKc quail (Sell ct al., 1972). Gilletl and Arscott (1909) also found hepatic microsomal epoxidase activity markedly depressed in quail fed DDT. These depression* of hepatic enzyme activity arc in accord with the pro longed sleeping times we observed in quail fed DDT. Although lengthened sleeping time is an indication of reduced activity of liver microsomal enzymes which detoxi fy pentobarbital, it is misleading to imply that pesticides which lengthen sleeping lime decrease "liver microsomal enzymes." It would be unusual and unexpected if al! liver microsomal enzymes responded to pe*ticirica in the *ntm' manner. Avoidance of the simplistic use of "liver micro soma) enzymes'* as a single class, all reacting in nn idenimtl fashion, and consideration of specific microsomal enrvme* will provide s greater understanding of the biological eifet-tn of pesticides. LITERATURE CITED Ahou-Donis. M. B.. Menzrl. D. B.. Riochcmintrv7. 1768 (lOGHi. Bitmsn. J.. Cecil, H. C.. Harris. 5. Fries, 0. F., J. Aar hunt Vhtm. 19, 373 11971) Bush. M. T.. Physiol, i'hormncnl 1, 206(190.1). Gillett, J. W,, Arscott, G. H.f Comp, Riachrm. Physiol. 30, 5*9 Hart. L. G., Foots, J. R.. Naitnvn-Schmicdebrrf* Arrh Fro Ikithol. Pharmakni. 219. 480(1965). Pesksll, D. NaluTr HAmdon) 2lfi, 505 (1967). Pcakall. D. U., Science 168,592 (1970). Rosenlwrg,, P., Coon. J. M., /Voc. Sot. Exp. Riot. Med *8. 650 (1968). Sell. J. L., Davison, K. L., Puyear. R. L.. J. Apr. Food CAcm 19. 68(1971). Sell, 4. L.t Davison, K. L., Poonscha, K. B., J Apt Food Chem. 2W3), 563(1972). Stephen, B. J.. Garlich. J. D.. Guthrie. F. E.. Rutt. F.nvimn Con- tom. 7'oxkot. 5,569(1971). Tucker. R. K . Crabtree. D, G.. U. S. Department of the Interior. , Fish and Wildlife Service, Resource Publication No. 84,1970. Received for review July 25,1972. Accepted October 10.1972. Long-Term Studies of Residue Retention and Excretion by Cows Fed a Polychlorinated Biphenyl (Aroclor 1254) G. F. Fries,* G. S. Marrow, Jr., and C. H. Gordon Nine cows were fed 200 mg per day of Aroclor 1254 (PCB) for 60 days. Milk and body fat samples were obtained during and for 60 daya following feeding. Concentrations of PCD in milk fat ap proached equilibrium after 40 days. The aver age concentration in milk from 40 to 60 days was 60.9 * 4.6 pg/g fat. Concentration in body fat was 41.7 11.6 Mg/g at 60 days. When feeding stopped, concentration in milk fat declined 60% within 15 daya. After 15 days the rate of the first-order de cline in concentration was much less. The average rate constant was Q.QlO day"1 and varied among cows from 0.005 to 0.016 day*1. The variation could not be related to such parameters as milk fat production or body weight change. Decline in concentration of PCB in body fat paralleled de cline in concentration of PCB in milk fat. Considerable interest has developed in polychlorinated biphenyl* (PCB), a class of industrial organochlorinc com pound* After many years of use, residues of PCB'a have recently been found distributed widely in environmental and food sample* (Kolbyo 1972; Pcakall and Linccr, 1970). The most extensive use* of PCH's have been as di electric* and plasticizers (Nisbet and Sarofim, 1972). However, the minor use of the PCB Aroclor 1254 in a silo aealnnl pose* u significant source of PCB contamination of milk. Aroclor 1254 is s commercial mixlure of PCB's contain ing an average 54% chlorine, with the major components runging from tetrnchlorobiphenyl to heptnchlorobiphcnyl (Sissons and Welti. 1071). A commercial sealant applied to tho interior* of concrete stave silos contained about 16% Aroclor 1254. Thi* source has caused PCB residues in milk exceeding the U. S. Food and Drug Administration guideline of 5 pg/g in milk fat (Kolbyu, 1972). Agricultural Environmental Quality Institute. Agricul tural Research Center, Agricultural Research Service, BcUsvtllc, Maryland 20705. Residues of PCB in silage occur at high levels adjacent to the treated walls (Skrentny et a/.. 1971). There is some migration of the residue in the silage, but residues seldom occur beyond 3 ft from the wall. Gas chromatographic ex amination of (he residues from silage suggest hide, if any, microbial or chemical change in the composition of Aroc lor 1254 (Fries, 1972; Skrentny et at., 1971). We have compared the behavior of PCB and DDK rcsidue* in milk after environmental sspnsute (Fries *1 nf., 1972). Only the period after removal of the sourres of both contaminants from the diet was studied. The rates of de cline in milk concentrations of the two residues were iden tical. Plalonow ct al. (1971) have studied the distribution of residues in the milk of two cows that received a single dose of PCB. The average excretion in milk was 5% within 4 days. We are not aware of long-term controlled studies on the relationship of dietary intake of PCB to residue ac cumulation and excretion in the cow. This study was carried out to determine milk and body fat residues while cows consumed a fixed level of Aroefor 1254 and to determine the rate of decline of milk arid body fat residue level* after removing PCI) from the diet. HGN$ 067*46 PltlKK, MAHHOW, tiOltlKiN Table I. Days In lactation, Weight, Inlako, and Production of the Cows* Days in Cow liciMitn Weight. kg Dry matter intake, kg/doy Milk. kg/day Fat. Fal. % kg/day t 211 2 207 3 2oe 4 143 6 i 161 159 7 39 46 * 46 5S9 536 667 637 677 628 687 406 607 16.0 16.3 15.0 16.5 15.6 17.9 17.6 16.3 17-1 17.7 4.1 0.72 15.3 4.1 0.62 11.0 4.3 O.S1 18.3 3.9 0.72 14. S 4.2 0.61 20.8 3 8 0.7B 22.7 3.9 0.68 22.9 3.6 0.79 21 9 3.4 0 75 "Days In lactation at iha start ol tha study. Alt other values ora ,veragoft lor iha 120 days ot ma study. '\ MATKHIAI.S AND METHODS Cows. Nine first-lactation Holstein cows were used and their pertinent characteristics are presented in Table I. The cows were selected to provide observations at three stages of lactation and at various levels of production within a stupe of lactation. However, because of the differ* Ing rates of production decline, the differences in produc* tion Iwcarnc less as the experiment progressed. Cows were fed corn silage, alfalfa hay, dehydrated alfal fa pellets, and concentrate. A description of the feeding regime has l>een presented by Miller tt al. (1971). There were some differences in ration composition between cows and within cows during the experiment. These differences did not affect results and only total dry matter intake is presentod (Table I). Dosing and Samplings All cows were fed 200 mg of Arorlor 1254 per day for 60 days. Aroclor 1254 in acetone solution was pipetted on a small portion of the concen* trate and the acotona was allowed to evaporate. The spiked concentrate was hand mixed with the remainder of the concentrate at the time of feeding. Significant feed re fusals were not encountered. Based on the measured dry matter intake and the body weight of the animals, this dose rate was equivalent to 12.4 A 0.9 mg/kg of dry mat ter or0,37 0.2 mg/kg of body weight. Milk samples were collected at 5-day intervals during the feeding period and the following 60 days. More fre quent milk samples were obtained both when feeding was inilioted ond when feeding wok stopped. Body fat biopsy samples from the tailhead area were obtained at 30, 60, 90, and 120 days after the start of the study. Analysis, Fat from the milk and biopsy samples was isolated and cleaned up using U. S. FDA (1968) multipest* icitio residue methodology. The concentration in most samples was relatively high and there were no significant interferences. Thus, it was not necessary to use more elab orate clean-up methods such ns silicic acid chromatogra phy (Armour and Burke, 1970) or dehydrochlorination (Fries et al., 1972). Concentrations of PCB were determined by gas-liquid chromatography using electron capture detection. A Hewlett-Packard Model 7GQ0A instrument equipped with Ni electron capture detector and an electronic integrator was used. The 6 ft x V in. o.d. glass column was packed with 10% DC-200 on 80 to KM) mesh Gas Chrom Q. The carrier gas was a 957# uiK*>n-57 methane mixture with a 120 ml/min Dow rate. A purge gas was not used. Column, inlet, and detector temperatures were 200, 240, and 250*, respectively. Typical chromatograms of milk fat and body fat resi dues as well ns an Aroclor 1254 standard, arc presented in Figure 1. Under these conditions there are 15 peaks in the standard. There were only seven significant peaks in the residue samples. Their retention times were equal to the retention times of peaks. 6 through 15 (except 10) in the standard. The concentration of i'CB in the samples was calculated by comparing the arcos of these seven peaks with the ares of the corresponding peaks of Aroclor 1254 standards. RESULTS AND DISCUSSION Nature of Residues. The peaks with the shorter reten tion times (peaks 1-7) did not occur in either type of resi due samples (Figure 1). The components that did occur in the residue samples (peaks 8-15) were present in approxi mately the same proportion ns In tho Aroclor 1254 stan dard. The relative proportions of these components did not differ significantly among sampling times, among cows, or between types of samples. For this reason data on the individual components are not presented. Sissons and Welti (1971) have determined the composi tion of Aroclor 1254 and the retention indices of its com ponents. From th'*ir work and some comparisons with in dividual PCB's tl .j'ure 1), it was concluded that the major components occurring in the residue samples were penlachlorobiphenyls and hexachlorobiphenyls. The com ponents that were metabolized and did not occur in the residue samples were tctrachlorobiphenyls and some of the pentachlorobiphcnyls. There is little information on the metabolic fate of PCB's. Block and Cornish (1959) reported the formation of 4-chloro-4'-hydroxybiphenyl and some of its conjugation products from 4-chlorobiphenyl in rabbits, it is possible that this reaction is quite general for PCB's if the appro priate positions are open for hydroxylation. The possibili ty of an open position would decrease as the degree of chlorination increased. The relative electron capture responses of PCB'a from tetrachlorobiphcnyls through decachlorobtphenyl general ly do not vary by a factor greater than 2 (Zitko rt al , 1971). Thus, our method of quantitation should compare reasonably well with newer methods in which all compo nents are converted to decachlorobiphenyl. Residues while Feeding. Milk samples obtained before the start of PCB feeding did not contain detectable PCB Figure 1. Typical gas chromatograms ol an Arocior 1254 stan dard. a milk 1st rosidue sample, and a body 1st rescue sample. Retention times ol peaks 1. 2. 4. 6, and 9 wero equal to me re tention times ot the pure compounds 2.S,2'.S`-letrachiorobiphenyl, 2.3,2'.5'-totrachlorobiphenyl, 2.5.3'.4'-ieirachiorobiphenyl. 2.3.4.2'.5'-pentachloroblphenyi, and 2,4,5.2'.4'.S'-heaachtorobiphenyl. respectively. MGNS 067449 AHOCLOH \l%* residues nr Mgnificnnt interference*. Thu* we concluded that the feed* nnd l he cow# were not exposed to sources of PCB other than the material fed. The avrrnue level* of I'Cll in milk fat and body fat duf- Ins (hc 60-day feeding period are presented in Figure 2. The shape of the milk fnt concentration curve is typical of the curves for other chlorinated hydrocarbon compounds (Fries ef al. 10091. After 40 days, the level of milk fat ap proached equilibrium but was still rising slowly. Concentration of PCH in body for r more slowly nnd was always lower than concentration in milk fat. The ratio of concentration in body fat to concentration in milk fat became closer U'.l) at GO days. At equilibrium one would expect a ratio of 1:], if the findings of Stull ct al. (1971) with DDT ran he applied to PCD. The evidence indicates that the cows were not at equi librium at GO days. However, the response from the 40th to the fiOth day is adequate to determine if any character istics of the cows significantly affect levels excreted in\ milk nr accumulated in body fat. Average concentrations of PCB in milk fat from the 40th lo GOth day were similar for all cows (Table II). The average for a)) cows was 60.9 4.5 ug/g and the minor variations were not related to production or any of the parameters listed in Table 1. Because of the smalt varia tion in concentration, the amount excreted per day was directly related to fat production. The average excretion was 42.3 mg/day, accounting for 21% of intake. Concentrations in body fat at 60 days, averaging 41.7 11.6 pg/g, were more variable than concent rations in milk fat (Table 11). With all cows, the concentration in body fat was lower than the concentration in milk fat. As in the case of levels in milk fat, variation among cows cannot be explained by the parameters measured in this experiment. The total amount of body fnt, one of the major parame ters that could affect body fat concentration, cannot be (Manured with ease. In a previous experiment with rats we have shown that the total retention of chlorinated hy drocarbons Is quite constant under a given set of condi tions (Fries et o/., 1971). Concentrations in body fat were inversely proportional to the total amount of body fat. If a similar phenomenon held true for these cows, one would expect that cows with smaller body fat pools would have higher body fat concentrations of PCB. Iteslduc Klimination. There was a rapid initial dacline In the level of PCB in milk fat when PCB feeding stopped (Figure 3). After 15 days the rate of decline was quite slow, Tire shape of this curve is similar to curves that have been established for other chlorinated hydrocarbon compounds (Fries et al., 1969). The decline in concentration in milk fat is adequately BATS Figure 2. Concentration ot PCB In milk tat and body tat of cows fed 200 mg of PCB per day. Each point t an average ot mn cows 4 standard deviation. The curve tor body fat is extrapolat ed before 30 days. described by a two-componcnt first-order system with the equation C - C,-* + Cte *' (1) where C is the concentration at any time, C| is the initial concentration of the first component, C* is the initial con centration of the second component, hi and As are rate constants, and t is time in days. This is a general equa tion describing the concentration of a compound in the ef fluent of two-compartmcnt system. We have discussed Its application to residue elimination from cows elsewhere (Fries etai, 1969). After 16 days the contribution of the first term to the total concentration ie negligible end can be ignored. The values of the constants for the second term were estimat ed by fitting the conventional linear regression of the loga rithmic form of the second term. After subtracting the MOMS 0 6 7 *5 0 Tebfe tl. Fat Production, PCB Eicrotlen In Milk, and PCS Accumulation In Body Fat ot Cows Consuming 200 mg at PC0 per Day0 PCB In mttk tat Milk tat production,1 Concentra EscteKon. Body tat, Cow */)' tion. fig/g mg/day M9/9 1 0.72 59.2 42.6 34.9 2 061 58.9 36.6 39 0 a 0.63 57.9 30.7 39.6 4 0.72 60.1 49.9 25.3 6 0.68 64.2 37.2 54.0 0.77 63. 49.1 63.2 7 0.69 56.6 60.4 37.1 a 0.70 57.6 40.3 92.3 9 0.73 70.6 61.5 60.2 Milk lei production and concentrations In milk tat are averages tor dta 40th to 60W> day ot feeding, Concontraltona In Oody lot art at lha doth dayef feedinp- Figure 9. Concentration of PCB In milk fat and body fsf of cows aftor feeding stopped. Each point Is an average of nine cow* 4 etendard deviation. J. Anr. Food Omm Vnl Si Mn i lot* KUIKS. MAHItONV. COUPON Table III. Decline in Concentration ol PC9 In Milk Pol oiler feeding Slopped Cow Intli.H concentration Milk tot. Body 1*1, p/b po/u Mith fl produc tion. hg/dsy Initial Weight de ch*nqe.* cline,* kg/*y % Bate.*1 day* 1 SO? 2 . 66.3 3 67 9 4 C0.1 6 64.2 63.0 7 66.6 67.6 9 70.6 34.5 39.0 30.6 26.3 54.0 53.2 37.1 32.3 60.2 0.61 0.60 0.44 0.67 0-S6 0 69 0.61 0.72 0.70 0 22 67 0.011 0.42 55 0.005 0.73 66 0.007 0.29 64 0.008 0.58 54 0.006 0.31 52 0.016 0.07 68 0.010 0.07 68 0.006 0.31 49 0.015 *10 to 00 days poll-feeding *100 ((concentration at 0 day* - cencerv nation at to day** (continuation at 0 dnyij). '* in tq i calculated tor 1$to 60 days poiMoedtng. lnClnCj-M (2) vatuct for the second term from the overall concentration, it wo* possible to estimate the constants of the first term. The equation estimated using the averages of all cows was C- 30.6 <*-/ + 32.3 (3) With individual cows, it is not possible to precisely evalu ate the first term of cq 1 bocausc of the limited number of observation*. In practice this is not of great importance. The level of PCI) in milk fat of all cows declined about S in the first lf> days, and there was little variation among cows (Table 111). This is in agreement with our field stud ied (Fries rl ol, 1972). llnlike the initial decline in concentration, the rate of decline for the slower compartment is quite variable among cows (Table 111). Wq have also observed this varia tion in previous studies (Fries et al, 1972). These differ ences ire of practical importance. If two cows start at the same level, the time required to return to the FDA guide line is inversely proportional to the magnitude of the rate constant during the second phase. In this study the range between the extreme cows was a factor of 3. Even wider ranges were observed in the field studies. Despite selecting cows in varying stages of lactation and of varying production levels, it was not possible to estab lish causes of variation among cows. It can be assumed (hot the only significant source of PCB for this compart ment is the body fat stores. This is supported by the de cline in concentration in body fat that paralleled the de cline in concentration in milk fat (Figure 3). The dose positive relationship between concentrations of chlorinat ed hydrocarbons in milk fat and body fat ie presented elsewhere (Fries and Morrow, 1972). If the model is appropriate, there are several factors that could influence the rate of decline in concentration in milk fat. Cows with higher fot production would tend to excrete more PCH, and this could be expected to increase the role of decline. However, there was no significant rela tionship between the rules of decline and the amount of fat produced (Table III). A second factor that could affect the rate is a change in body weight. All the cows were gaining weight in this study (Table 111). The amount of PCB remaining in the body would be diluted by this added l>ody fat. Thus, one would expert the dilution os indicated by weight goin to enhance the rate of decline in milk fat concentration. However, there was no consistent relationship between these two factors (Tabic 111). The third factor, not evaluated in this study, is the lota! amount of body fot. A given rate of milk fat produc tion would clear a larger fraction of body ful if a cow hae a Table IV. Comparison of DOE and PC8 Accumulation and Escrelien by Cows* Parameter Milk tat level. 20 days ltig/g) Milk tat level. 60 days (#ig/g) Body tat level, 60 days (fjg/g) Ci ivg/g) *i (day - *) Ct (*g/g) *, (Mr") DOE 0.21 0.04 0.29 0 05 0.23 0 02 0.12 0.31 0.17 0.013 PCB 0.20 0.02 0.32* 0 03 0.21 * 0 06 0.15 0.32 0.16 0.010 ODE value* are from Fries ef t! (1969) Intake* are nermaMied to 1 mg/day. Ct, h, C*. end S| ere the Conianl | *q 1. small amount of body fat than if she has a large amount of fat. Some indication of the amount of body fat can be obtained from the concentration of PCB in the body fat at 60 days feeding if it is assumed that all cows absorb PCB with the same efficiency. The concentration would be in versely related to the amount of body fat and the most rapid decline would occur with the cows with the highest initial concentration in body fat. However, the overall re lationship between initial concentration in body fat and rate was not consistent (Table HI). The rate of decline is probably controlled by a combina tion of the above and/or other factors rather than a single factor. With the small number of animals involved, it was not possible to do a more elaborate multiple factor analy sis of the data. The length of time required to restore a cow or a herd to a level below the FDA guideline of 5 pg/g of milk fat de pends not only on the rate constant but also on the initial concentration in the milk fat. In the worst case, if the ini tial concentration was 10 **g/g (twice llie guideline) ami rate was 0.005 day-1, the lime required would be 139 days. An additional 139 days would be required for each doubling of the initial concentration. A more reasonable estimate could be made by using the average rale for all cows (0.010 day-1). In this case. 69 days would be re quired to reduce the concentration by half. Similarity to DDE. In earlier work (Fries ct ol, 1972), we suggested that residue behavior of Arocior 1254 was similar to DDE, the most persistent degradation product of DDT. The conclusion was based on the similar rates of elimination after the cows had been removed to clean feed. The results of this study confirm and extend the conclusion. We have conducted experiments with DDE under con ditions similar to this experiment (Fries cf ol, 1969). The parameters of the DDE experiments are In dose agree ment with the parameters of this PCB experiment when all values are normalized to an equal intake (Table IV). The close agreement of the two sets of parameters is re markable considering that PCtt's are a multiple residue and that many systematic errors may have been intro duced by our method of quantitation. The result supports the conclusion that the relative areas of the peaks consid ered did not vary significantly among samples. Systematic errors in estimation of the absolute PCB values would not affect the parameters listed in Table IV. DDE is among the organochlorine compounds most re sistant to metabolic degradation. The agreement of the two sets of parameters suggests that the PCB'e present in the residues arc also among the most resistant compounds to metabolic degradation. LITERATURE CITED Armour. J. A., Burke, J. A., J. An. Offie. Anol. Chtm. S3, 761 (1970). Block. W. D.. Cornish. H. II., J. tliol Chrm 234,3301(1959). Fries, G. F., Adtmn. Chem. 6>r /Vo. 11), 256(1972). MOMS 0 6 7 4 5 1 120 J Anr FnodChem.. Vol. 21. No. 1. 1973 Fries, (. K., Drydcn. I.. I*,. Mnmm-, (J. S., Jr., Hnrtntnn. A. II.. Ahttirnrt. Wi2ml National Mvi'imt! *f (hr Antrrirun (.'hi-micai .Si*rh`Cy,.S`ii 7 12, IfITI. Fries, O. K.. Marrow, <'. S.. Jr.. J. Dairy Sei. M.TOO 0972). Krics, (i. K.. Mormw, (5. S.. Jr., Cordon, C. H., J. Dairy Sri. 52, IMKIUOti!)). Krirn, (. K,, Mnrrow, Ci. S,, Jr., Cordon. C. H., Bull. Environ. Contam. Tuxteiil. 7, 2-V2 (1072). Kolhyr, A. C., Environ, Health l*rr*pect. 1.85 (1972). Miller, It. IT. Honven, N. tV,, (imilh, J. W., J, Dairy Set. 54, 867 (1971). Niftlwt, I. C. T., Simfim, A. F.. Environ. Health Pcrxpecl. I, 21 (1972). Fenkall. 0. B., Linccr, J. L., HiuSei. 20,958(1970). niAZI\ON |\ sJIKIJ* Plutonnw. N. S,, Konmll, II. S,, UulWV, I). II.. Anwar. I) it.. SMMt-lH-iiltrerlicr. I1. VV,, (rtrve, |). (;., J, !hur\ St \. 51. (.Hl.'i (1971). ' SiMMinr. D.. Welti. I).. /. Chromnlour (SO, 15 (1071). Skrenlnv. R. K.. Hcmkrn. K. W,, Borough. H. W.. Hull. Environ. Cantoni. Toxicol 8, 409(10711. Stull. J. W.. Hnnm. W H.. Whiting. K. M.. Hull. Envinm. Con- tom. To.xienl. (*, Aftft 11071). U. S. Food nnd Drug AHtnirmirnriwi. "Pwlieirie Analvtiml Man* ual," Vol. I, Washington. IV (V. 10*18, Zitko, V.. Hutzingrr, 0., Safe. S,, Hull. Environ, Cantnm. Toxi col. , IG0 <1971). Received for review May V 1972. Arrrplcti II, 1972. Toxic Metabolites of Diazinon in Sheep . Norman F. Janet,` Antony F. Machin,* Michael P. Quick, Heather Rogers, Dennis E. Mundy, and Alan J. Cross Diazinon is oxidised in sheep to several cholines* ternsc-inliibiting metabolites. The structures of three of them were determined by direct spectra* scopic measurement. Two are monohydroxy dias* inons, and the third is a dehydration product of one of these; the three structures had already been proposed for diazinon metabolites produced by mice, when the establishment of structure was based on cochromatography with synthetic sam ples. Some quantitative aspects of the distribu tion of the compounds in sheep ere reported. Hecauac diazinon is widely used as an agricultural in* sccticide in conditions where it may be ingested by mam* mals, its metabolism has been studied for many years. It is only recently, however, that specific structures have been pul forward for some of the metabolites formed. In all these structures the pyrimidine moiety has been modified, either before or after cleavage from the diethyl pbosphorothionyl group. Mucke et of. (1970) showed that in rats the isopropyl group of the pyrimidinol derived from diazinon was hydroxylatcd to give two isotneric hydroxypyrimidinones. These compounds ore much less toxic than diazinon be cause they hove lost the phosphorus-containing group. Miyazaki et ol. (1970), in an extensive study of the metab olism of diazinon by mice, suggested structures for several urinary mrtolxrlitc* in which all the phosphate bonds were etiU intact. Identifications were based on a comparison of the chromatographic properties of the metabolites with those of synthetic compounds. The present work is concerned with those metabolites of diazinon in sheep that are still indirect inhibitors of cho linesterase. Three of these were isolated in sufficient quantity to establish their structures by direct spectro scopic measurements. We have reported the occurrence of two of thorn (referred to below as I and fl) in brief prelim inary accounts (Machin et ai, 1971b, 1972), EXPERIMENTAL SECTION Materials. Solvents were analytical reagent grade and were distilled immediately before use. Chloroform was washed free from ethanol and dried before distillation. Diazinon nnd diazoxon were gifts from Fisons, Ltd., Agrochemical* Division. Hydroxydiazinon (0,0*diethyl CM(2-hydroxypri>p*2-yl)*6-melhylpyrimidin-4-yl]l photphoro(hinnnlc) was prepared by irradiation of diazinon (Machin ft of, 1971a). Ministry of Agriculture. Fisheries and Food. Agricultural Development ami Advisory Service, Central Veterinary Laboratory, Biochemistry Department, New Haw, Wey* bridge, Surrey, England. Department of Insecticides and Fungicides, Rothsmsted Experimental Station, Harpcnden, Hens, England. Thin-layer chromatography (tic) was on 8-in. square plates coated with silica gel G or GF/UV254 (Machery, Nagel & Co.); final separations were carried out on (he specially purified grade N-HR/UV254. Column chroma tography was on Florisil, Johns-Manville Cclite 546, and Woelm silica gel and neutral alumina. Gas Chromatography. Residues were determined and preparative separations of the metabolites monitored on a Vartan Aerograph Model 204 chromatograph with a thermionic detector or on a chromatograph (Machin and Morris, 1972) which was also used to purify samples. Col umns were .3 ft or & ft X in. o.d. glass, packed with 1.6-2% XE-60 or 26% SE-30 on Aeropak 30, 100-120 mesh. Isolation of Metabolites. Two sheep were dosed by stomach tube with diazinon () g/kg) which produced moderate Symptoms of poisoning. One was killed after 46 hr. Urine was collected from the other for 3 days and the sheep was allowed to recover. Metabolites I and 11 were isolated from urine collected during this 3-day period. The urine (120 ml) was diluted to 600 ml with water and ex tracted for 48 hr with chloroform (800 ml) in a liquid-liq uid extractor. The chloroform extract was divided into portions (4 x 200 ml), each of which was concentrated to 10 ml and applied to a column of silica gel (Brockman grade 11, 25 g, % in. diameter). The metabolites were elut ed with 1:40 methanol-chloroform (160 ml) and the cluate was concentrated, first in a rotary evaporator and then in a stream of dry nitrogen, to 0.3 ml. After lie with 1:4 acctone-hcxane as mobile solvent, the bands contain ing I and II were eluted separately and chromatographed twice more, first with the same solvent and then with 2:3 ethyl acetate-hexane to remove an unknown contaminant detected by its fluorescence at 3G0 nm. I and II were again eluted with acetone and the solvent was evaporated. Me tabolite I was also seporated from the tissues, as described previously (Machin etal., 1971b). Metabolite III was mainly concentrated in the fat and was extracted by macerating 200-g portions of fat three times with acetone (200 ml). After each maceration the suspension was centrifuged at 3000 rpm for 30 min at ~20* and the supernatant was decanted. The extracts HOXS 0 6 7 4 5 2