Document Y24Xm8bJyx5x24XDq93ZBgxE

Arch. Environ. Contarn. Toxicol. 17, 65-71 (1988) Archi"" of 0 1988 Spnnger-Vcrlag New York fnc. Influence of Termite Treatment in the\Home on the Chlordane Concentration in Human Milk \ S. Taguchi and T. Yakushiji Osaka Prefectural Institute of Public Health. 1-3-69, Nakamichi, Higashinari-ku, Osaka 537, Japan Abstract. An increase of chlordane contamination in humans after termite treatment of their homes was investigated by analyzing human milk samples of breast-feeding women. Chlordane in human milk was analyzed by gas chromatography-mass spectrometry (GC-MS) and the concentrations were compared between women living in termite-treated and non-treated houses. Pglychlorinated biphenyls (PCB) were also analyzed on the same samples and the result was discussed in relation to the concen- tration of chlordane. Mean concentrations of six chlordane compo- nents of the exposed group were three times higher on the average than those of the control group. When the auration of dwelling was less than one year, however, chlordane levels of the treated group were similar to those of the control. A proportional correlation was observed between the concentration of total chlordane and the duration (years) after the termite treatment of the home. The pattern of ch!ordane components observed ir. the milk of the exposed group resembled that of technical chlordane, a commercial product. Chlordane compounds were so stable and persistent in the treated homes that continuous accumulation into the dwellers persisted for at least five years after treatment. The levels of total chlordane in the milk samples of the control group were very low compared to PCB levels. However, when duration of dwelling after termite treatment expanded for two years or longer, chlordane in the milk increased to a level similar to PCB. A close correlation between concentrations of chlordane and PCB of the non-treated group was observed. This fact suggested that a main source of chlordane exposure for the general population was marine products. However, in the case of the treated group, chlordane exposure was inferred to be derived from the inhalation of a contaminated atmosphere in the homes or absorption of chlordane via the skin or uptake of foods contaminated in the homes. Chlordane is a viscous, light yellow to amber-colored liquid. Technical chlordane is a mixture of at least 26 different components (WHO 1984a). The main components are trans-chlordane (24 ? 2%), cis-chlordane (19 t 3%), y-chlordene (9%), heptachlor (7%), trans-nonachlor (7%) (Sovocool et al. 1977). Chlordane has been used for approximately 40 years as a broad spectrum contact insecticide, mainly on non-agricultural crops and animals. In Japan, use of the chemical has been permitted only for the control of termites and wood-boring beetles. Chlordane is so persistent in the environment and stable chemically that it maintains its effects for approximately fifteen years after treatment. The acute toxicity of chlordane in rats is estimated to be similar to that of DDT. Signs of poisoning in various animal species are neurotoxic manifestations such as disorientation, tremors, and convulsions (WHO 1984a). WHO (1984b) rated the toxicity of chlordane as "moderately hazardous". Most of its metabolites are slightly to moderately toxic, with the exception of highly toxic oxychlordane (acute oral LDriofor rat: 19.1 mg/kg body weight). The acute lethal dose of technical chlordane for man is estimated to be 25-50 mg/kg body weight (WHO 1984a). No adverse effects have been reported in occupationally exposed workers. In Japan, the consumption of chlordane has increased rapidly since 1980. Subsequently, its envi- 66 ronmental contamination became a serious social task. Recently, the Environmental Agency of Japan conducted an investigation on the contamination of chlordane in the environment, and reported that the concentration of the chemical in marine organisms was increasing. Based on animal experiments carried out for the evaluation of chronic toxic effects of chlordane, the Japanese government decided to prohibit the use of the chemical in 1986. Chlordane vaporizes gradually in the domestic atmosphere of treated homes (major use of chlordane is termite control) for over ten years. It will accumulate in the dweller's body by the inhalation of vapor, by the intake of foods which adsorbed the vapor, or by skin contact of the chemical. Data are available on the levels 0; chlordane in human milk of breast-feeding women whose homes have not been treated with the chemical for termite control (Mes and Davies 1978; Strassman and Kuntz 1977; Barnett et al. 1979; Savage et al. 1976, 1981). Until now, however, there are few data on dwellers in homes treated with chlordane, except the report of Kawano and Tatsukawa (1982), which reported a relatively high concentration of oxychlordane in one blood sample. With these reports, it is still difficult to relate the chlordane treatment of the home to the increase of the levels of chlordane contamination in man. Our study was conducted to clarify the problem by using human milk samples collected in the Osaka Prefecture from 1984 to 1985. Chlordane levels in the milk of breast-feeding women were analyzed by GC-MS and the results of subjects living in the treated house were compared with that of a non-treated group. Polychlorinated biphenyls were also determined and the results were discussed in relation to the concentration of ch]ordane. Experimental Reagents Column chromatography for the partial purification of chlordane in samples was made with FlorisiP (60-100 mesh, Katayama Chemcial). Florisil" (60- 100 mesh, Wako pure Chemical Industries) was used for the purification of PCBs. Reagents used for the analysis were all special analytical grade for PCBs and pesticides (Wako Pure Chemical Industries). Six chlordane components in a technical chlordane of "Tech. Ref. Std. chlordane 100%" (Velsicol Chemical Corp., Lot. No. b-8113) were determined for the pattern analysis of chlordane residues. ,.- + Samples Human milk samples from 197 breast-feeding women were collected in Osaka prefecture from 1984 to 1985. Milk fat of the samples was extracted by means of the official procedure for S. Taguchi and T. Yakushiji PCB analysis of Japan (Ministry of Health and Welfare Japan 1972) and kept at 5C until used. Exposed group consisted of sixteen samples from women whose homes were treated with chlordane. As a control, seven were selected randomly among 181 human milk samples. Extraction and Clean-up Chlordane Analysis: After milk fat was removed by FlorisiP dry column chromatography (Wakimoto and Tatsukawa 1972; Yakushiji et ul. 1979),organochlorine compounds were extracted with n-hexane. The extracts were concentrated to 3 to 5 ml and pouted into a glass column (30 cm x 15 mm I.D.) containing IS g of Florisilm activated for 15 hr at 130C. Chlordane compounds were eluted with 120 ml of n-hexanelether (9:1, v/v). The eluates were concentrated to 0.5 ml with a Kuderna Danish concentrator. In a preliminary study, three chlorinated hydrocarbons. p,p`DDD, tetrachlorobiphenyls and 6-hexachlorocyclohexane (6HCH) were examined for use as an internal standard for chlordane analysis. Retention times of these `compounds on caprllary gas chromatography were different from all of the chlordane components. However, large quantities of unknown contaminants whose retention time was the same as p,p'-DDD and tetrachlorobiphenyls were detected in the sample extracts. Therefore, 6-HCH was chosen as an internal standard and added to the sample extracts. The concentration of 6-HCH was standardized at 200 pg/L in both sample extracts and stapdard solutions. PCB Analysis: PCBs were extracted according to the official procedure of Japan (Ministry of Health and Welfare Japan 19721, partly improved (Yakushijiet al. 1978). After the milk fat sample was saponified with alcoholic potassium hydroxide, PCBs were extracted with n-hexane and cleaned up by Florisilm column chromatograph y. , Instruments Chiordune Analysis: A doub!e fecming mass spectrometer (Jeol JMS DX-300) coupled to a data system JMA 3100/3500, with a Hewlett-Packard model HP5710A capillary gas chromatograph equipped splitless injector was used for selected ion monitoring of chlordane. Chromatographic conditions were as follows; a methylsilicone fused silica capillary column (Durabond" DB-I, film thickness 0.25 pm, 30 m x 0.32 m m I.D.)programmed from 150C to 310C at 16Clmin; injection temperature 250C; carrier gas (helium) at a flow rate of 20 mllmin. Mass spectrometer operating conditions were as follows; temperatures ("C): separator 297, chamber 200: ionization voltage 70 eV; accelerating voltage 6.5 kV. The monitored ions (m/z)were 217 3-HCH (internal standard); 303 and 305 y-chlordene; 353 and 355 heptachlor epoxide (HCE); 373 trans-chlordane and cis-chlordane; 387 and 389 oxychlordane; 407 frons-nonachlor and cis-nonachlor. PCBs Analysis: A Varian 2100 gas liquid chromatograph equipped with a nickel-63 electron-capture detector, fitted with 2% OV-1 (Gaschrom Q 100-120 mesh, 1.8 m x 2.0 mm I.D.) was used for analyses of PCBs. Temperatures of detector, injector and column was adjusted to 210C, 210C and 250C respec- . .i Chlordane in Human Milk 67 Table 1. Means of the concentration (pg/kg) of chlordane and PCB in milk fat of lactating women Age Na (years) Panty Y.A.T.b (years) y- trans- chlor- chlordene dane cis- oxychlor- chlordane dane transnonachlor cis- total non- chlorachlor dane` HCEf PCB Non-exposed control 7 26.3 1.7 - 3.1 1.5 2.1 19.3 25.6 6.4 58 21.1 261 ExposeddVe 15 28.7 1.8 1.8 11.0 7.9 5.9 33.6 84.9 19.1* 162 54.9* 374 group 1 3 29.7 2.0 0 < Y.A.T. < 1 0.2 0.4 2.0 10.4 19.2 4.4 37 20.9 344 group 2 6 29.8 2.0 1 S Y.A.T. < 2 10.0* 5.9*** 6.0* 26.2 67.9*** 17.6** 134*** 43.5** 375 group 3c 6 27.0 1.5 2 5 Y.A.T. 17.3 13.6 7.7 52.5* 134.9* 28.0** 254* 83.2** 389 a Number of samples Years after termite treatment of house Total concentration of chiordane compomds excep: for heptachlor epoxide {RCE) Group 1, 2, and 3 are combined e One sample ("Y.A.T." was eight years) was rejected (see text) HCE = heptachlor epoxide * P < 0.05. **P < 0.01. ***P < 0.001; significance level against control tively. The carrier gas was nitrogen at a flow rate of 30 ml/min. PCB components whose retention times were greater than p , p ' DDE were quantified by using Kanechlor" mixture (Kanechlor" 300, 400, 500, 600 = l:l:l:l, wlw) as the standard. The Ugawa method (Ugawa et nl. 1973) was used for the quantitation of PCBs. Results and Discussion Chlordane components in human milk were statistically higher in the exposed group than in the control group (Table 1). Levels of cis-nonachlor and HCE increased when the period (in years) after the termite treatment of the house was longer. To indicate the duration of exposure in the house where the termite treatment was carried out with chlordane, the exposure period (in years) is expressed with the term of "Y.A.T." which stands for the year after the treatment. Mean concenrraiion of total chlordane in the milk of exposed women whose exposure was two years or longer (group 3) was 4.4 times higher than that of the control group. Heptachlor epoxide (HCE) is a metabolite of heptachlor and technical chlordane contains approximately 7% heptachlor. Before its use for termite control, however, heptachlor had been used as an insecticide for agricultural purposes. These two origins of HCE contamination confused the source of HCE exposure, particularly for the exposed group. Therefore, the concentration of HCE was not added to the total chlordane in this paper. Three major chlordane residues in human milk, namely oxychlordane, trans-nonachlor and cisnonachlor, have been measured in several countries. These data for the exposure of general population are summarized in Environmental Health Criteria 34 (WHO 1984a). Levels of chlordane in human milk in Canada (Mes and Davies 1978) and the USA (Strassman and Kuntz 1977; Barnett et al. 1979; Savage et al. 1976, 1981; FAO/WHO 1981) are higher than those in Japan. Miyazaki et al. (1985) analyzed chlordane residues in milk samples of Japanese women. Mean concentrations of oxychlordane, trans-nonachlor and cis-nonachlor were 16-20, 24-45 and 4-7 kg/kg, respectively. Our measurements for nonexposed control group (Table 1) were within the same range as reported by Miyazaki et al. (1980, 1985, 1986a). Technical chlordane is a multicomponent insecticide, and the results of our analysis are shown in Figure 1. Miyazaki et al. (1986b) reported that the pattern of chlordane residues in river water and sea water were quite similar to that of technical chlordane. Moderate changes of the component patterns in freshwater fish, marine shellfish and seawater fish were observed at zach stage of the hod-chain. Major changes were the increase in the relative percentages of trans-nonachlor, oxychlordane and cisnonachlor, and the decrease of trans- and cis-chlordane. This tendency was even clearer in human milk samples (Miyazaki et al. 1986a). Our results agreed with Miyazaki's in the changes of the component pattern. The patterns in human milk of both control and exposed group were different from that of technical grade chlordane. Relative percentages of major components of the technical chlordane, trans-chlordane, cis-chlordane and 7-chlordene, were very small compared to those of technical chlordane (Figure 1). These components are presumed to be metabolized relatively easily and to disappear rapidly from the human body. This tendency has been reported in animal experiments (Tashiro and Matsumura 1977, 1978; Barnett and Dorough 1974). 68 S. Taguchi and T. Yakushiji Y ;; 50 mm i E D *Y 20 contra1 group Exposed group 0technical chlordane c Fig. 1. Composition of chlordane compounds detected in the control and the exposed group. Relative percentages in total chlordane (58 pglkg for the control group, 162 figlkg for the exposed group) were calculated by using the mean concentration in Table I. The ratio of oxychlordane to total chlordane (relative %) of the exposed group was less than that of the control group. However, the ratios of trans- and cis-nonachlor to total chlordane of the exposed group were greater than that of the control group. Oxychlordane does not exist in technical chlordane, and is a metabolite formed in a living cell. Brimfield and Street (1979) proposed metabolic routes for cis- and trans-chlordane and the formation of oxychlordane in rat microsomes. Our results confirmed the reported trends. In another investigation, oxychlordane was biotransformed in experimental animals from trans-nonachlor via transchlordane (Tashiro and Matsumura 1978). In humans, however, it was deduced that transnonachlor is very slowly metzbolized to oxych!ordane via trans-chlordane (Kawano and Tatsukawa 1982; Miyazaki et al. 1986a). On the basis of the reported results, the pattern of chlordane components of the exposed group is more likely to be that of a technical chlordane when compared to the control group. Namely, trans-nonachlor is highly accumulated probably through domestic exposure to the technical chlordane in the home, and since the rate of biotransformation of trans-nonachlor into oxychlordane is very slow, the percentage of oxychlordane resulted in a smaller value. Similar results were reported on the ratios between DDT residues in human milk. The ratio of DDE to DDT is lower in countries where DDT is still being used for pest control than in countries where its use has been severely restricted (Slorach and Vaz 1983). When the use of DDT is restricted, the level in vegetable foods falls rapidly. Human Y m c1 5 600 .3 E E .- 500 mCa : 400 A rU 300 Y Y0 'CI O 200 C 0 m.r( Y :: 100 6E1 / /* /' /0' Y-100.4X-18.0 2 Y.A.T. ( year ) Fig. 2. Correlation between the concentration of total chlordane and the penod after termite treatment (Y.A.T.). Linear regres- sion equations were obtained for all the data (smaller slope line) or for the data after one sample was rejected (greater slope line). The data for eight years was rejected , exposure to its metabolite DDE continues, however, through the consumption of animal foods ( e . g . fish) which continue to accumulate this metabolite from the environment longer after the prohibition of DDT. DDT is also metabolized gradually to DDE in the human body. The difference in the ratio of chlordane components between the control and the exposed group in this report indicates that the dwellers in the treated homes are exposed to chlordane which has been applied to the house. On the contrary, relative percentages of chlordane components found in the milk of the control group indicate that the major source of contamination is a biological material in foods. Concentrations of total chlordane were compared to those of PCB, which was analyzed as a representative of organochlorine contaminants in human milk. Mean concentration of total chlordane in the control group was less than '/4 of the mean PCB level (Table 1). However, it increased to a similar level to PCB when the period of exposure extended into two years or more. Correlations between the concentration of total chlordane and the period of exposure are shown in Figure 2. When a linear relationship is assumed, a correlation coeacient (r) for all the data was 0.417. The situation became more clear (r = 0.866) when one result for the longest residence time (eight years) was excluded. This indicates that chlordane i Chlordane in Human Milk 69 Table 2-1. Correlation coeficient matrix between each item for the control group Age Parity Y.A.T." y-chlordene trans-chlordane cis-chlordane oxychlordane trans-nonachlor cis-nonachlor total clordanesb HCE PCB Age -0.595 -0.112 -0.195 0.005 -0.106 -0.397 0.822 -0.492 0.380 -0.156 Parity - 0.032 0.010 -0.319 -0.508 -0.412 0.431 -0.833 -0.164 -0.732 Y.A.T.' Y- chlordene transchlordane cischlordane oxychlordane transnonachlor cisnonachlor total chlordaneb - 0.972 0.766 0.373 - 0.845 - 0.084 -0.159 -0.096 -0.236 - 0.743 0.392 -0.805 -0.239 -0.127 -0.045 -0.242 - 0.864 -0.634 -0.200 0.346 0.476 0.198 - -0.205 -0.404 0.720 0.680 0.488 - -0.280 0.502 -0.059 0.459 - -0.575 -0.064 -0.091 - 0.440 0.785 HCE - 0.492 PCB - a Years after the termite treatment of house Total concentration of chlordane compounds except heptachlor expoxide (HCE) Table 2-2. Correlation coefficient matrix between each item for the exposed group Age Parity Y.A.T." y-chlordene trans-chlordane cis-chlordane oxychlordane trans-nonachlor cis-nonachlor total chlordaneb HCE PCB Age - 0.454 -0.184 -0.147 -0.059 -0.143 -0.020 -0.040 0.072 -0.042 -0.189 0.072 Parity L -0.017 0.189 0.346 0.187 0.162 0.280 0.282 1 0.264 0.034 -0.465 Y.A.T." - 0.761 0.686 0.741 0.916 0.849 0.808 0.866 0.590 -0.063 Y- chlordene - 0.880 0.950 0.850 0.889 0.637 0.904 0.552 -0.135 transchlordane - 0.789 0.833 0.925 0.746 0.924 0.628 -0.086 cischlordane 0.812 0.856 0.617 0.867 0.441 -0.236 oxy- chlordane - 0.955 0.884 0.970 0.712 -0.090 transnonachlor - 0.853 0.996 0.634 -0.087 cisnonachlor - 0.870 0.722 -0.143 total chlordaneb - 0.666 -0.105 HCE - -0.248 PCB - * Years after the termite treatment of house Total concentration of chlordane compounds except heptachlor epoxide (HCE) compounds are so persistent in the environment and stable in the human body that the accumulation continues for at least five or six years after the treatment of the house. As for the sample which was rejected, we have no reason to explain the low value. It is to be determined whether the atmospheric concentration of chlordane in the houses suddenly decreases after five years. Further inves- tigations are necessary in this area. Tables 2-1 and 2-2 show matrices of correlation coefficient between each item. A close relationship was noted between the concentrations of total chlordane and PCB in the control group (r = 0.785). Each component of chlordane was less clearly related to PCB. In another investigation conducted on human milk of residents in Osaka from 1972 to 1977, a proportional correlation was observed between the levels of PCB and pesticides, such as HCH, p,p'-DDT, p,p'-DDE (Yakushiji et af. 1979). The relationship suggested that PCB and the organochlorine pesticides were taken into the human body through similar foods. It is generally accepted that the mqjor source of contamination of organochlorine compounds for the general popula- tion is the ingestion of foods. Intakes from drinking water and air are deermed to be negligible (Bergland 1972; Kolbye 1972). Watanabe et af. (1979) con- ducted a surveillancct of dietary PCB intake of ordinary Japanese womdn, and indicated that approximately 70 to 90% of PCB was derived from eating marine fish. Jelinek and Corneliussen (1976) have also reported that fresh water fish are the major source of PCB exposure in the U.S.A. From the data, it was deduced that the main source of chlor- 70 S. Taguchi and T. Yakushiji Table 3. Mean value of the concentration (@g/kg)of chlordane and PCB in milk fat of lactating women of first and second or more childbirth 7- trans- cis- oxy- trans- cis- total Age average Y.A.T.b chlor- chlor- chlor- chlor- non- non- chlor- Na (years) Parity (years) dene dane dane dane achlor achlor danec HCE` PCB Group (I)d 10 27.0 1.0 1.8 7.2 4.3 4.2 27.3 57.3 15.4 116 53.5 441 Group (IIY 13 30.1 2.4 2.5 12.7 9.9 6.7 36.2 100.5 22.4 188 51.8 312* a Number of samples Years after the termite treatment of house Total concentrations of chlordane compounds except heptachlor epoxide (HCE) Mothers of first childbirth e Mothers of second or mare childbirth HCE = heptachlor epoxide *: P < 0.05 dane for the control group was marine products in their diet. On the other hand, the concentration of total chlordane showed no correlation to that of PCB in the exposed group (Table 2-2). This indicates that the source of chlordane contamination was different from that of PCB. A good correlation is observed among chlordane components. The concentration of six chlordane components is also related to exposure time, as shown in Figure 2. From the data, the main source of chlordane in the exposed group was not from the intake of fish. The route of contamination for the exposed group was inferred partly to be from inhalation in the home of a contaminated atmosphere and partly to be absorption via the skin. Another inference may be the adsorp- tion of chlordane to dietary materials during the pe- riod of storage in the kitchen. The tendency that the main source of chlordane was not fish intake in the exposed group is also apparent in Table 3, where the results of the exposed group are classified into two groups by the parity number. The mean concentrations of PCB in milk fat obtained from mothers after the second or more childbirth (312 Fg/kg) was significantly lower than that from mothers after the first childbirth. It has been recognized that the excretion of PCB via breast milk is a major excretory route. In general, a 30 to 50% decrease in the PCB level of mother's milk is observed after five to six months of breastfeeding (Yakushiji et al. 1982). The half-life of the concentration of PCB in human milk was estimated to be approximately eight months (Yakushiji et al. 1978). In the present study on chlordane, however, such a tendency was not observed. Contrary to the case of PCB, total chlordane measured in the samples of group (11) was such higher than the group (I) (Table 3). Milk chlordane levels of the exposed group were strongly influenced many years after the treatment of the house for termite control. Thus, it was clearly demonstrated that the chlor- dane treatment of a house increased the contamination level of human milk. This study should arouse interests in the toxicological and environmental areas of chlordane pollution, especially in the treatment of houses. Further studies are strongly recommended. Acknowledgment. The authors are grateful to Dr. T. Nishimune for reading the manuscript. References Barnett JR, Dorough HW (1974) Metabolism of chlordane in rats. J Agric Food Chem 22:612-619 Barnett RW, D'Ercole AJ, Cain JD, Arhteur RD (1979) Organochlorine pesticide residues in-human milk samples from women living in northwest and northeast Mississippi, 1973-1975. Pestic Monit J 13:47-51 Bergland F (1972) Levels of polychlorinated biphenyls in food in Sweden. Environ Health Perspect 1:67-71 Brimfield AA, Street JC (1979) Mammalian biotransformation of chlordane: In vivo and primary hepatic comparisons. Ann NY Acad Sci 320:247-256 FAOWHO (1981) Joint FAO/WHO food and animal feed contamination monitoring programme. Summary of data received from collaborating centres-1977-1980. Part BContaminants, World Health Organization, Geneva Jelinek C, Corneliussen PE (1976) Levels of PCBs in the U.S. food supply. In: National conference on polychlorinated biphenyls (Nov. 19-21, 1975, Chicago, Illinois), p 147, Washington DC 20460: US Environmental Protection Agency, Office of Toxic Substances, Washington, DC Kawano M, Tatsukawa R (1982) Chlordanes and related compounds in blood of pest control operators (PCOs). Nippon Nbgeikagaku Kaishi 56:923-929 (Japanese) Kolbye AC (1972) Food exposure to polychlorinated biphenyls. Environ Health Perspect 1:85-88 Mes J, Davies DJ (1978) Variation in the polychlorinated biphenyl and organochlor pesticide residues during human breastfeeding and its diurnal pattern. Chemosphere 7:699706 Ministry of Health and Welfare, Japan (1972) Official method for PCB analysis, January 1972 (Japanese) Miyazaki T, Akiyama K, Kaneko S,Horii S, Yamagishi (1980) - Chlordane in Human Milk Chlordane residues in human milk. Bull Environ Contam Toxicol 25518-523 Miyazaki T, Yamagishi T, Matsumoto M (1985) Confirmation of residue levels of organochlorine compounds in human milk by gas chromatographylmass spectrometry and capillary gas chromatography. J Food Hygienic Society of Japan 26:666673 (Japanese) (1986a) Component patterns of chlordane in human milk, cow's milk, dog adipose tissue, cat adipose tissue, and wild birds. J Food Hygienic Society of Japan 27:481-486 (Japanese) (1986b) Component patterns of chlordane in seawater, river water, fish, and shellfish. J Food Hygienic Society of Japan 27:49-58 (Japanese) Price ON, Young RW, Dickinson JK (1972) Pesticides residues and polychlorinated biphenyl levels in diets, urine, and fecal matter of preadolescent girls. Proc SOCExp Biol Med 139: 1280-1283 Savage EP (1976) National study to determine levels of chlorinated hydrocarbon insecticide in human milk: 1975-1976. US Environmental Protection Agency, (EPA/540/9-78/005, Order no. PB284393) Washington, DC Savage EP, Keefe TJ, Tessari JD, Wheeler HW, Applehans FM, Goes EA, Ford SA (1981) National study of chlorinated hydrocarbon insecticide residues in human milk. Am J Epidemiol 113:413 Slorach SA, Vaz R (1983) Assessment of human exposure to selected organochlorine compounds through biological monitoring. Swedish National Food Administration, Uppsala Sovocool GW, Lewis RG, Harless RL, Wilson NK, Zehr RD (1977) Analysis of technical chlordane by gas chromatographyhass spectrometry. Anal Chem 49:734-740 Strassman SC, Kuntz FW (1977) Insecticide residues in human milk from Arkansas and Mississippi, 1973-74. Pestic Monit J 10:130-133 Tashiro S,Matsumura F (1977) Metabolic routes of cis- and trans-chlordane in rats. J Agric Food Chem 22872-880 (1978) Metabolism of trans-nonachlor and related chlordane components in rats and man. Arch Environ Contam Toxicol 7: 113-127 Ugawa M, Nakamura A, Kashimoto T (1973) Studies on a cal- 71 culation method for polychlorinated biphenyl isomers. J Food Hygienic Society of Japan 14:415-423 (Japanese) Wakimoto T, Tatsukawa R (1972) Report of analytical method for PCBs. Japanese J Water and Waste 14:1125-1133 (Japanese) Watanabe I, Yakushiji T, Kuwabara K, Yoshida S, Maeda K, Kashimoto T, Koyama K, Kunita N (1979) Surveillance of the daily PCB intake from diet of Japanese women from 1972 to 1976. Arch Environ Contam Toxicol8:67-75 W H O (1984a) Chlordane. Environmental health criteria 34, World Health Organization, Geneva -(1984b) The use of WHO recommended classification of pesticides by hazard. World Health Organization (Unpublished Report VBC/84,2), Geneva Yakushiji T, Watanabe I, Kuwabara K, Yoshida S (1978) Identi- fication of low chlorinated biphenyls in human milk by gas chromatography-mass spectrometry. J Chromatogr 154: 203-2 I O Yakushiji T, Watanabe I, Kuwabara K, Yoshida S,Koyama K, Kunita N (1979) Levels of polychlorinated biphenyls (PCBs) and organochlorine pesticides in human milk and blood collected in Osaka Prefecture from 1972 to 1977. Int Arch -Occup Environ Health 43: I I5 Yakushiji T (1982) Review on the PCB contamination in human milk. Jap J Perinatal Care 1:68-74 (Japanese) iYakushiji T, Watanabe I, uwabara K, Yoshida S,Koyama K, Hara I, Kunita N (19 ) Long-term studies of the excretion of polychlorinated bi henyls (PCBs) through the mother's milk of an occupatio ally exposed worker. Arch Environ Contam Toxicol 7:493 495 Yakushiji T, Watanabe I, uwabara K, Yoshida S,Hori S, Fu- kushima S, Kashim to T, Koyama K, Kunita N (1979) Levels of organochlo `ne pesticides and polychlorinated biphenyls (PCBs) in mo hers' milk collected in Osaka Prefecture from 1969 to 19 6. Arch Environ Contam Toxicol 8: 59-66 Manuscript received March 7, 1987 and in revisedform June 20, 1987.