MARCH, 1966 THE ANALYST Vol. 91, No. 1080 The Analysis for Residues of Chlorinated Insecticides and Acaricides A Review* BY K. I. BEYNON AND K. E. ELGAR (44Shell” Research Limited, Woodstock Agricultural Research Centre, Sittingbourne, Kent) SUMMARY OF CONTENTS Introduction Sampling and storage Extraction procedures Soils Crops Water Animal tissues and products Column chromatography Paper chromtography Thin-layer chromatography Liquid - liquid partition Precipitation of fats and waxes Chemical methods Other miscellaneous methods Gas - liquid chromatography Total halide analysis Polarograph y Bioassay Colorimetric and ultraviolet analysis Infrared analysis Fluorimetry Clean-up procedures Quantitative analysis Paper chromtography Thin-layer chromatography The positive identification of pesticide residues Recommended methods General procedures Individual procedures (i) Aldrin (ii) BHC (iii) Chlorbenside (Mitox) (iv) Chlordane (v) Chlorfenson (Ovex) (vi) Chlorobenzilate (vii) DDT, TDE and DDE (viii) Dicofol (Kelthane) (ix) Dieldrin (x) Endosulfan (Thiodan) (xi) Endrin (xii) Heptachlor and its epoxide (xiii) Methoxychlor (xiv) Oxythane (Neotran) (xv) Tetradifon (Tedion) (xvi) Toxaphene WE have attempted to prepare a selective and critical review of the work published up to May, 1965, on the analysis for residues of the most widely used chlorinated insecticides and acaricides, and their principal metabolites.We have made no attempt to present an historical account of previous work, but have attempted to produce a balanced picture of procedures that are relevant at the present time and also to give our own assessment of their relative j m port ance .A comprehensive treatise edited by Zweigl was published during 1963 and 1964 on the analysis of pesticides. The second volume of this work gave detailed accounts of procedures for the analysis of most of the compounds considered in this review. However, few of the methods recommended then can be recommended today as being the best procedures available for chlorinated pesticides. Residue analysis procedures, particularly for chlorinated pesticides, were revolutionised in 1961 by the application of gas - liquid chromatography (GLC) with either electron-capture or microcoulometric detection systems. The analysis of small amounts of pesticide is possible with such procedures, and the detection of 0.05ng of some compounds with the electron- capture detector is now routine.In the absence of a valid control sample, GLC analysis cannot provide positive identification of a particular pesticide when only one retention-time value is obtained. Many naturally occurring products respond to GLC detectors and can be mistaken for pesticides, and often such naturaI products are not completely removed by the clean-up procedures that are used. This * Reprints of this paper will be available shortly. For details see Summaries in advertisement pages. 143 However, a few words of caution are necessary.144 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [AfldySt, VOl. 91 behaviour is encountered also with analytical procedures other than GLC.It is a factor, however, that has been ignored by many workers.. The problem of the positive identification of a residue in a sample of unknown history is the most important one facing the residue analyst at the present time and it will be discussed in detail in a later section. We have considered many aspects of the analysis for residues of chlorinated insecticides and acaricides, and the review will be of use to those chemists who wish to analyse samples of unknown history, as well as to those who wish to analyse samples from field trials where adequate controls are available and where the sample history is known. SAMPLING AND STORAGE The principles to be adopted when sampling crops and soils for the analysis for residues of chlorinated pesticides are the same as those when sampling for residues of other compounds.Correct sampling, however, is so important that the principles cannot be emphasised too strongly. Experimental studies of sampling procedures for crops have been reported by Huddleston et L z Z . , ~ Van Middelem et aL3 and by Poos et u Z . ~ Lykken et ~ ~ 1 . ~ 3 ~ have reviewed the literature on the sampling of crops and the second of these reviews,6 in particular, presents a clear account of the correct procedure. Lykken6 recom- mends that the gross samples of crops should be 25 to 100 lb or units, and that this sample should be mixed, quartered and sub-divided to obtain representative replicate 2-lb samples for analysis. The size of the gross sample, however, must be related to the size of the plot and to the size of the crop.Studies of the procedures for soil sampling do not seem to have been reported, but the recommendations concerning the representative nature and size of the sample for crops apply also to soils. Soil samples are generally taken as a core and Lichtenstein et a1.' took 30 cores (2 inch diameter x 9 inches deep) from 500 square feet of dieldrin-treated soil. The depth to which sampling takes place will depend on the depth of penetration of the pesticide. These chlorinated insecticides and their derivatives are strongly adsorbed on all arable soils and do not leach through the soil to any significant extent. I t is usual, therefore, to sample within the cultivated depth unless the total amount or depth variation of the residues is required.The statistical principles of sampling have been summarised by Garber.8 The need for the sample to be representative of the plot from which it is taken6 and the necessity for the participation of the residue chemist in samplingg cannot be emphasised too strongly. These needs are so great that it is surprising that more experimental work has not been reported on the comparison of sampling procedures for residue analysis, especially for soils. STORAGE OF SAMPLES Crop and soil samples must be stored at a temperature at which the residues and the crop do not decompose further whilst awaiting extraction. This may seem an obvious precaution but experimental evidence for such stability is rarely given in the literature. Samples are often stored6 at 1" to 5" C, but this temperature is too high for the extended storage of many crops.Although most of the chlorinated pesticides considered here may be stable for some days when stored at 1" to 5" C, it is recommended that storage for longer than a few days should be at temperatures of -10" C, or below, in closed containers. Whatever the storage temperature, the pesticide residues in the crop and soil must be shown experi- mentally to be stable under the conditions used. Such evidence is best obtained by analysing field-treated samples after storage for different times at different temperatures. It is advisable to record the weight of samples with a high water content prior to deep-freeze storage, for when they regain room temperature disintegration and moisture loss can be rapid.The extracts of crops and soils must also be stored, prior to analysis, in conditions under which further decomposition of the pesticide does not occur, and storage at 1" C, or below, and in the absence of light is recommended. If recovery experiments are carried out at the time of the extraction it is possible to obtain evidence for the stability of the pesticide during the storage of the extract. EXTRACTION PROCEDURES A universal extraction procedure has not yet been developed and the problems peculiar to the extraction of chlorinated pesticides from soils, crops, water, animal tissues and fatty materials, respectively, will be considered.March, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 145 Often, insufficient evidence is given for the adequacy of an extraction procedure, par- ticularly for soils.Most workers carry out recovery experiments by introducing known amounts of pesticide at the extraction stage. However, whilst good recoveries are sufficient evidence that the subsequent stages of the analytical procedure (e.g. , concentration steps, clean-up) do not lead to losses of pesticide, they do not necessarily mean that the extraction procedure is efficient. I t is far more difficult to remove a pesticide from field-treated samples, when the pesticide has penetrated into the sample structure, than to extract a pesticide introduced at the blending stage. The efficiency of an extraction procedure may be estab- lished by extracting field-treated samples with a range of solvents for a range of extraction times.When increasing the severity of the extraction conditions (solvent polarity or extrac- tion time) leads to no further increase in the amount of pesticide extracted, one can be reasonably satisfied that the procedure is adequate. SOILS Many chlorinated pesticides are strongly bound by dry soils, and the adsorptive capacity of the soil will vary considerably with moisture content , organic-matter content, the polarity of the compound and other factors. For example, hexane will extract aldrin from dry soil, but it is not suitable for the extraction of dieldrin residues.10 Extraction with acetone will give a good recovery of most chlorinated pesticides but the co-extracted material can interfere with the analysis. Gouldenll has shown that acetone extraction of soils followed by partition of the pesticide into hexane is a satisfactory procedure when the final analysis is by electron- capture GLC.The use of 10 per cent. acetone in hexane is considered to be adequate for the removal of most chlorinated pesticides from soil without excessive co-extraction of inter- fering substances.1° Benzene - TPA,12 hexane - IPA13 and pentane - acetone13 have also been used. Soxhlet extraction of the air-dried soil with acetone has been used for the removal of DDT from soill4 but air-drying, prior to extraction, cannot be recommended as a general procedure as the pesticide may be volatile. A procedure that can be applied to the extraction of chlorinated insecticides from a range of soil types has been described.15 The soil is mixed with anhydrous sodium sulphate to make it friable and is then extracted with 10 per cent.acetone in hexane in an end-over-end tumbler for 1 hour. In order to improve recoveries for a wide range of chlorinated pesticides the acetone content of the extraction solvent can be increased to 20 per cent. without undue interference, especially if the final analysis is to be by GLC. CROPS Surface rinsing of crops is simple and results in little interference from co-extractives. I t is, however, inadequate for the removal of residues other than those adhering loosely to the crop surface or dissolved in the waxy, surface layer. Since it is rarely used nowadays, we will confine our attention to the extraction of the whole crop. The extraction of pesticides from crops has been reviewed by Bann,lo Heinisch,ls Thornburg17 and Van Middelem.ls Useful comparisons of extraction procedures have been described by Bann,lo Klein and his c o - w o r k e r ~ ~ ~ ~ ~ ~ ~ ~ ~ and by Hardin and Sarten.22 Mills, Onley and Gaither23 have described a useful acetonitrile extraction procedure applicable to many chlorinated pesticides in a range of crops.Prior to the extraction, the crop should be subdivided and mixed thoroughly. Grains such as rice, and seeds such as cotton are broken in a mill prior to the e ~ t r a c t i 0 n . l ~ BannlO showed that aldrin and dieldrin were readily extracted from a range of fresh, un- processed crops such as alfalfa, carrots, corn, dates, figs, beans, turnips and wheat by macera- tion with hexane followed by tumbling.With many crops, especially frozen or canned foods, low recoveries or emulsion problems were encountered. To overcome these problems the use of a mixture of polar and non-polar solvents or the maceration of the crop with solvent in the presence of sodium sulphate has been recommended. The water-miscible solvent is generally removed by water washing prior to analysis. Klein and his co-workers20y21 compared the efficiencies of three procedures for the extraction of DDT, aldrin and methoxychlor from spinach, collards and beans. Blending with benzene - IPA was more efficient than either tumbling with benzene - IPA or Soxhlet extraction with benzene or benzene - IPA. Satis- factory recoveries (95 per cent.) were obtained by blending the crop and IPA (in the ratio of 1 g to 1 ml) for 2 minutes followed by the addition of benzene (2 ml) to the mixture and a further 2 minutes maceration.Complete equilibrium existed between insecticide and sample blend, and two pour-offs were necessary to achieve 95 per cent. recovery.146 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [Analyst, VOl. 91 Hardin and Sarten22 compared the efficiencies of different extraction procedures for the removal of DDT from field-treated collards and their results are summarised in Table I. Blending with IPA followed by the addition of hexane and further blending was the most efficient and rapid procedure, as was found by Klein and his co-workers.19@~21 TABLE I THE EXTRACTION OF DDT FROM COLLARDS DDT extracted (p.p.m.) Extraction procedure (mean values) Tumbling with hexane (30 minutes) .. .. .. .. . . .. 20.9 4 1*1* Blending IPA (2 minutes), addition of hexane + tumbIing (30 minutes) Blending with hexane (2 minutes) + tumbling (30 minutes) . . . . . . Blending IPA (2 minutes) + addition of hexane + further blending (2 minutes) . . . . . . . . .. .. . . .. . . 35.9 f 2-9 Grinding sodium sulphate + tumbling hexane (30 minutes) . . .. .. 24-8 i 1.0 30.4 & 2.6 . . 36.4 5 1.1 *Standard deviation Whilst a procedure suitable for the extraction from a wide range of crops of most of the chlorinated pesticides considered here has not been developed, some general indications are possible. Blending for some minutes with hexane - IPA, benzene - IPA (or with ethanol or acetone instead of IPA) should be suitable procedures.The presence of anhydrous sodium sulphate during the blending should further decrease the emulsification problems. It does not seem to have been established clearly that prior blending with the water- miscible solvent is necessary, and a simple blending procedure with the mixed-solvent system is an efficient procedure for many pesticides.1° 9 1 7 WATER The analysis of pesticides in water has been reviewed recently by Hindin, May and D u n ~ t a n . ~ ~ Chlorinated insecticides are of low solubility in water and they may be extracted with water-immiscible solvents, such as hexane or benzene. However, when it is necessary to extract large volumes of water to achieve the necessary sensitivity, batchwise procedures can be time consuming. Rosen and M i d d l e t ~ n ~ ~ removed the pesticides from 2000 litres of water with a carbon filter, and desorbed them from the carbon by Soxhlet extraction with chloroform.Recoveries of BHC, chlordane, DDT, aldrin, TDE and endrin were in the range 75 to 86 per cent. a t the 2-5 p.p.m. level, but some were lower below the 1 p.p.m. level. Teasley and Cox2s preferred a batchwise liquid - liquid extraction process as they considered that several chlorinated pesticides were unstable on activated carbon. Subse- quently Kahn and Wayman27 described a simple apparatus that can be used for the continuous extract of several hundred litres of water with petroleum spirit, and they obtained a 83 to 100 per cent. recovery at the 0.2 to 340 p.p.b. (parts per thousand million) level with a range of compounds.Previous workers have extracted large volumes of water in order to detect chlorinated pesticides at the p.p.b. level. However, because of the high sensitivity of gas -liquid chromatography, it should now be possible to carry out liquid - liquid extraction by a smaller batchwise process prior to analysis by GLC. ANIMAL TISSUES AND PRODUCTS The chlorinated compounds dealt with in this review and many of their metabolites are fat soluble, and some of them tend to concentrate in the lipoid portion of the plant or animal system. Extracts of fatty materials can have such a large amount of co-extracted material that analysis by the usual residue methods is impossible without rigorous clean-up. Extraction merely involves dissolving the product in a suitable solvent before the necessary clean-up procedures are begun. However, attempts have been made to minimise the fat content of the extract, e.g., by the use of polar-extracting solvents, such as acetonitrile17 or dimethyl s~lphoxide,28~~~ or by the use of an aqueous mixture containing an em~lsifier.~~ However, the solubility of fat in these solvents is low and some of the pesticide may remain in the undissolved fat.Extraction with alcoholic alkali is useful for the alkali-stable cyclodiene insecticide^,^^ 932 and also for DDT and its analogues,= although DDT is unstable to alkali.March, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 147 A general procedure for the preparation and extraction of dairy products has been given by Mills.34 Butter was clarified by warming to about 50" C and filtered.Cheese and milk were blended with sodium (or potassium) oxalate and alcohol, extracted with three volumes of ether and light petroleum, and the combined non-aqueous layers washed with water and evaporated. The fat obtained from all these products was dissolved in light petroleum. The procedure used for milk has been improved by O n l e ~ 3 ~ with an ether - acetonitrile - dioxane mixture as extracting solvent that gave a cleaner extract more quickly. A useful general scheme for extracting dairy products has been given by Langlois et aZ.36 that could also apply to all fatty tissues. A sample containing not more than 1 g of butter fat was ground with 25 to 30 g of Florisil (containing 5 per cent. of water) to give a free-flowing powder.This mixture was placed on top of 25 g of clean pre-washed Florisil in a chromatographic column, and the whole column was eluted with 20 per cent. volume of methylene chloride in light petroleum. Up to 650ml of eluant removed DDT, DDE, lindane and heptachlor and the more polar heptachlor epoxide, dieldrin and endrin with good recovery and precision. Other published methods for animal fats and tissues have usually involved grinding with anhydrous sodium ~ u l p h a t e ~ ~ ,38 939 ?40 or sand41 and dissolution in a suitable solvent. CLEAN-UP PROCEDURES Column chromatography remains the most widely used method of clean-up and its general use is likely to continue for some time. Many other procedures are useful, however, for particular problems. When GLC was first introduced for the analysis of residues of chlorinated pesticides it was hoped that clean-up would no longer be necessary.This is true for many samples of known history when residues of 0.1 p.p.m., or higher, are present. However, when it is necessary to detect residues a t a level less than 0.05 p.p.m., clean-up is generally necessary. For analysis by GLC with electron-capture detection or with the newer version of the micro- coulometer, the clean-up may take place on a smaller scale than was necessary with the older colorimetric procedures. Clean-up is almost always essential with samples of unknown history unless no interference is obtained at the desired level of sensitivity without clean-up. Gas - liquid chromatography has been used successfully for the clean-up of extracts and this application is covered in other sections.COLUMN CHROMATOGRAPHY The convenience and resolving power of column chromatography make it the most commonly used method of clean-up. Despite much work on the theory of chromatography, selection of particular solid or liquid phases is still largely empirical. Any separation may be due to the simultaneous action of adsorption, partition and ion-exchange processes, but as one of these factors usually predominates, they will be considered separately. ADSORPTION CHROMATOGRAPHY- The adsorptive capacity of a material depends on such factors as its structure, method of preparation, the presence of impurities, the treatment or activation it may have undergone, particle size, moisture content and the eluting solvent. I'ariation in the sorptive properties between different batches of many adsorbents tends to be high, and many workers have preferred to work with those with reasonably reproducible properties, such as the synthetic magnesium trisilicate, Florisil, A wide range of materials has been evaluated, however, and Florisil, alumina, silica gel and carbon are commonly used.Evaluations of Florisil by M ~ d d e s ~ ~ and by RiIills and c o - w o r k e r ~ , ~ ~ ~ ~ ~ of silica gel by Moats,43 and of carbon by Coulson and Barnes,44 Cassil et aZ.,45 1 B a e t ~ ~ ~ and have been reported. Coulson4* has advocated the use of aluminium silicate, while cellulose,49 magnesiaz3 931 and clays of the at tapulgite type have also been ~ s e d . ~ 1 The more powerfully the pesticide is sorbed to the adsorbent, the more polar the solvent needed to desorb and elute it from the column.Table IT gives the adsorbents in order of increasing adsorptivity and the eluting solvents in order of increasing polarity. In practice, hexane or benzene is the usual solvent with, if necessary, the addition of a small proportion of ether or acetone to remove the more polar pesticides, The adsorbents may be activated or de-activated as desired by the removal or addition of water. The mechanism of adsorption by charcoal differs from that of the adsorbents in Table 11. Generally, pesticides containing aromatic groups are more strongly adsorbed on charcoal than are the cyclodiene pesticides.148 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF TABLE I1 ADSORPTIVITY OF ADSORBENTS AND POLARITY OF SOLVENTS [Analyst, Vol.91 Adsorbents Cellulose Kieselguhr (Celite) Magnesia Silica gel Magnesium trisilicate Alumina Clays Solvents Hexane C yclo hexane Benzene Methylene chloride Ether Ethyl acetate Acetone Alcohols PARTITION CHROMATOGRAPHY- The partition of chlorinated pesticides into polar solvents from less polar solvents is exploited in partition chromatography, particularly for separations from fats and waxes. Abdallah and Landheerso have used both acetonitrile and dimethylformamide supported on Celite in the clean-up of lindane and DDT from fat by using pentane or hexane as eluant. Hoskins et aLsl successfully used alumina coated with polyethylene for the clean-up of a variety of crops with 40 to 65 per cent. of aqueous acetonitrile as eluant, and achieved an average pesticide recovery of 88 per cent.Thornburgl’ has suggested this method as a “universal” type of clean-up. Coulson et aZ.,52 Zweig et aLs3 and Crosby and LawsM have successfully used gas - liquid chromatography as a method of clean-up. This quite simple technique could have wider application to the stable chlorinated materials. ION-EXCHANGE CHROMATOGRAPHY- Ion exchange has proved useful in clean-ups5 956 and metabolism studiess7 of pesticides, but has been used to only a limited extent with chlorinated compounds. Few members of the group considered here are sufficiently acidic or basic for the technique to be employed, although it has been used for the analysis of DDA in ~rine.~8 PAPER CHROMATOGRAPHY Paper chromatography has been used extensively for the separation of mixtures of pesticides in extracts of plants, animal tissues and dairy products.I t has been used widely for the identification and quantitative analysis of pesticides, but has been little used for the clean-up of extracts prior to analysis by other methods. For this reason a more detailed discussion of paper chromatography will be considered later in the “Quantitative Analysis’’ section. The principles outlined in that section apply also when paper chromatography is used for clean-up. If paper chromatography is to be used for clean-up, a marker spot should be made visible and the assay sample must be eluted from the relevant section of the paper prior to further analysis. Any solvent which produces an RF value of 0.95 or greater for the desired component is satisfactory for the elution.However, the stationary and mobile phases may also be eluted along with the spot and may interfere in the subsequent analysis; this and the slowness of the method and the low capacity of the paper are the reasons why paper chromato- graphy has been used mainly for the identification of pesticides and not for clean-up. Recently Heinisch and Neuberts9 have described the application of wedged-shaped paper strips for the clean-up of plant extracts. TH I N-LAYER CHROMATOGRAPHY Thin-layer chromatography (TLC) may be used for the clean-up of extracts prior to analysis by other methods, for the qualitative analysis of pesticides, and for their direct quantitative analysis. TLC is more rapid and has a higher capacity than paper chromatography, and is more useful for the clean-up of extracts.TLC generally gives sharper resolution of chlorinated pesticides than paper chromatography and, unlike the latter, can be carried out successfully without the impregnation of the adsorbent with a stationary phase. The adsorbents, developing systems and detection systems will be considered in detail in the section on quantitative analysis. When TLC is used for clean-up prior to analysis by other methods, the pesticide is desorbed from the adsorbent at the R, value corresponding to the desired pesticide. TheMarch, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 149 RF value is determined by running a marker spot alongside the extract and only the marker spot is made visible.The correct solvent must be chosen for removal of the separated components from the adsorbent. Little specific information has been given as to a suitable solvent, but of course, any solvent which gives an RF value of 0.95 or greater for the particular component is suitable. Acetone, ether, alcohol or mixtures of these with hexane should remove most of the pesticides considered here. Abbott and Thornsons* have suggested wedge-layers for clean-up by TLC and this system should be useful for chlorinated pesticides. Taylor and Fishwickel have used loose-layer chromatography on alumina with hexane for the separation of aldrin, DDT and its metabolites, BHC, heptachlor epoxide, endrin and dieldrin, and this technique is especially useful for the investigation of solvent systems for the separation of pesticides by column chromatography.LIQUID - LIQUID PARTITION Jones and Riddick62 extracted several pesticides including methoxychlor, chlordane, lindane and DDT with hexane from plants, animal tissues and dairy products. The pesticides were partitioned into acetonitrile, and considerably less interference was encountered in the subsequent colorimetric or polarographic analysis as a result. This liquid - liquid partition procedure has now been used widely, especially for extracts of animal fats and tissues, and its use has been extended to many pesticides and a wide range of solvent pairs are available. Burchfield and Storrse3 showed that lindane, DDT and aldrin partition from hexane into DMF to a greater extent than into acetonitrile, and that the use of the high-boiling DMF was no drawback as the pesticide may be recovered from this solvent readily by dilution with water followed by partition back into hexane.The use of DMSO - hexane and acetonitrile - hexane was compared by Haenni et aLe4 who showed that BHC, aldrin, dieldrin, endrin and heptachlor partitioned to a greater extent into DMSO than into acetonitrile. Extraction of crops and soils with acetone and subsequent dilution of the extract with aqueous sodium sulphate solution and partition into hexane was used successfully by Goodwin et aLe5 This sytem was used also for animal tissues66 but it was found to be unsuitable for animal fats.# The animal fats and dairy products were extracted with hexane, and the pesticides (aldrin, dieldrin, $$’-TDE, $$’-DDT, BHC and heptachlor) partitioned into DMF.The DMF was diluted with water and the pesticide was partitioned back into hexane prior to analysis by GLC. Recoveries from partition processes are generally good, even at the microgram or nano- gram level, as long as the mixing of the solvent phases is effective and the partition is repeated a sufficient number of times.40 The relative volumes of solvents and the number of times that partitioning must be repeated may be determined from the partition coefficients, and TABLE 111 PARTITION COEFFICIENTS6’ OF CHLORINATED PESTICIDES AT 25.5” c Partition coefficient Pesticide Aldrin . . .. . . y-Chlordane . . . . pp’-DDE . . . . oo’-DDT . . , . &b’-DDT . . . . Dieldrin . . .. Endosulfan I . . . . Endosulfan I1 . . Endrin . . . . Heptachlor . . . . Heptachlor epoxide . . Lindane . . .. TDE .. .. .. Telodrin . . . . 7 Hexane - acetonitrile 0-73 0.40 0.56 0.45 0.38 0.33 0.39 0.13 0.35 0.55 0.29 0.12 0.17 0.48 Hexane - 90 per cent. aqueous dimethyl sulphoxide 0.89 0.45 0.73 0.53 0-40 0.45 0.55 0.09 0.52 0.77 0.35 0.09 0.08 0.65 Iso-octane - 85 per cent. aqueous dimethyl formamide 0.86 0.48 0.65 0.42 0.36 0.46 0-52 0.14 0.5 1 0.73 0.39 0.14 0.15 0.63 Iso-octane - dimethyl formamide 0.38 0.14 0.16 0.10 0.08 0.12 0.16 0-06 0.15 0.2 1 0.10 0.05 0.04 0.17 Iso-octane - dimethyl formamide (with 125 mg butter extractive) 0.39 0.16 0.18 0.11 0.09 0.13 0.17 0-07 0.16 0.23 0.1 1 0.06 0.04 0.19150 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [Analyst, VOl.91 several lists of values are a ~ a i l a b l e . * 0 ~ ~ ~ ~ 6 ~ ~ ~ 7 It is generally found that co-extractives do not have any great effect on the partition of the pesticide.67 Beroza and Bowman67 have measured the partition coefficients of a range of pesticides (Table 111) and have shown that liquid - liquid partition can be useful not only for clean-up, but for more positive identification of the pesticides. The extract can be analysed before and after a suitable partition procedure. This will enable the partition coefficient of the component to be determined, and comparison of the results with the values in Table I11 will allow more positive identification. This procedure is rapid and is useful during GLC when the particular conditions will not resolve a given pair of insecticides, and may often be quicker than attempting to alter the GLC conditions in order to attain the desired resolution.More recently these workers68 have extended their work and have used a counter-current distri- bution system to clean-up extracts and also to identify pesticides. PRECIPITATION OF FATS AND WAXES Precipitation procedures have often been used to remove interfering waxes and fats. Fairing and W a r r i n g t ~ n ~ ~ cooled acetone solutions of plant and animal tissues to -15" C to precipitate fats and waxes, which were then removed by filtration, and they reported good recoveries of methoxychlor from apple wax. Williams70 obtained good recoveries of chlordane when the interfering waxes in tomatoes, cabbages and apples were precipitated by cooling a methanolic solution in an ice-bath.Precipitation procedures have been used extensively by McKinley, McCully and their c o - ~ o r k e r s . ~ ~ , ~ ~ 9 7 2 9 i 3 3 7 4 3 7 5 Good recoveries of DDT ( o f and @'), TDE (Rhothane), methoxychlor and dicofol were obtained71 when waxes were precipitated at -70" C from acetone extracts of a range of fruits and vegetables. DDE, DDT and TDE were recovered from a range of animal fats72 when acetone solutions were cooled and a three-stage cooling procedure, one at 5" C and two at -70" C, was necessary to precipitate the fat from large samples. Recently a simple apparatus has been d e s ~ r i b e d ~ ~ ~ ' ~ for the precipitation of fats and waxes which should be useful for the processing of large numbers of samples.With the apparatus benzene - acetone solutions of plant and crop extracts were cooled to -70" C, and good recoveries were obtained by DDT (op' and $@'), lindane, heptachlor, aldrin, heptachlor epoxide, endrin and methoxychlor. Gunther and Blinn76 cooled benzene extracts of avocados to 0" C to crystallise the benzene together with any DDT. The avocado oil was removed from the crystal mush by filtration. McKinley and S a ~ a r y ~ ~ deposited an extract of butter fat on a charcoal column and eluted dieldrin without the butter fat with acetone at -70" C. The precipitation procedure is good for the removal of fats and waxes but not particularly useful for the removal of interference from other sources. Chromatographic procedures will usually remove several of the interfering classes of compounds and have found more widespread application.GLC procedures are sometimes less prone to direct interference from fats and waxes than many of the older colorimetric methods. However, the removal of fats and waxes that do not cause direct interference during G1.C is desirable if extended GLC column life with maintained efficiency is required. Precipitation procedures have also been used in other ways. C H E M I c AL M ETH o D s Chemical methods of removing or modifying co-extracted material can be applied to a limited range of pesticides but are tending to be superseded for normal use by partition and adsorption methods. However, the modification of a pesticide by chemical reaction will probably find wider use as an aid to identification.TREaTMENT WITH ALKALI- Saponification has been widely used as a method of clean-up for the alkali-stable cyclodiene compounds, aldrin,31 dieldrin32 977 and end~-in,~* for DDT (with conversion to DL>E)33Y79 for lindane,s0381 heptachlor epoxides2 and methoxychlor (with conversion to the ethylene compound) .83 A quicker method of clean-up with alkali has been reported for endrin by Albert,84 in which a potassium hydroxide - Celite column was used instead of saponification. One layer of potassium hydroxide - Celite (14 to 17 per cent. water in the mixture) was placed between two layers of magnesium oxide - Celite and the column eluted with light petroleum.March, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 151 TREATMENT WITH ACID- Schechter et aZ.g5 cleaned up chloroform extracts of milk containing DDT and glycerides by hydrolysis with fuming sulphuric acid.Davidow86 improved the method by percolating the extract through a column containing concentrated and fuming sulphuric acid supported on Celite. This technique has also been used for lindane.87 TREATMENT WITH OXIDISING AGENTS- This technique has occasionally been used for pesticides resistant to oxidation such as DDT, dieldrin and lindane as in the method of Gunther et aLa8 for the rapid determination of DT>T in dairy products, and that of O'Donnell et for dieldrin in unsaponifiable materials. Alkaline potassium permanganate has been the most favoured reagent, but chromic acid, acid or alkaline peroxide and acid chromic anhydride may also find a p p l i c a t i ~ n .~ ~ OTHER MISCELLANEOUS METHODS STEAM DISTILLATION- Some of the chlorinated pesticides are steam-volatile, but little attempt has been made to use this property because the volatility is greatly reduced in the presence of crop and particularly tissue extracts. have hydrolysed chlorfenson with alkali and steam distilled the resulting p-chlorophenol, and Gunther and Blinng2 have described a similar procedure for oxythane. Ott and GuntherS3 have attempted to devise a general method for fat clean-up by using forced volatilisation. The method is rapid and recoveries are good except for TDE and methoxychlor. HYDROLYSIS WITH ENZYMES- Cliffordg4 showed that clean-up by using enzymes was effective, but the method is lengthy and the enzyme not readily available so that this observation has not been used to any great extent.Gunther and J eppsongO and Butzler et QUANTITATIVE ANALYSIS Methods that have been used for the quantitative analysis of chlorinated pesticides In will be discussed in turn, and in each the specificity of the procedure will be considered. the last sub-section the specificity of residue methods will be considered in detail. GAS - LIQUID CHROMATOGRAPHY The rapid application of GLC as a technique and in particular the development of selective methods of detection for halogenated compounds have led to its world-wide use in the residue analysis of chlorinated pesticides. Because of the sensitivity and selectivity that it can offer, GLC has become the preferred method for the whole group of compounds con- sidered here.It was once hoped that the relatively small response from most crop constituents would mean little or no clean-up of but because of the interest in much lower residue levels, this hope has not been realised. Nevertheless, GLC has many of the charac- teristics of an ideal residue method. The high sensitivity of the detectors has meant that only small weights of pesticide, and thus smaller weights of co-extracted materials, are injected and this has led to the use of lower loadings of stationary phases and lower column temperatures, which have given more efficiency, resolution and life to GLC columns. NON-SELECTIVE DETECTION SYSTEMS- The first work on the gas chromatography of insecticides in 1958 was carried out by using thermal-conductivity detection.98 Later workers have also used this d e t e c t ~ r , ~ ~ , ~ ~ but the sensitivity is poor and it lacks selectivity.This latter disadvantage also applies to the use of the flame-ionisation detector despite its much greater sensitivity. SELECTIVE-DETECTION SYSTEMS- MiwocouZomet yy- Coulson et a1.9'3 introduced the rnicrocoulometric titrating system as a GLC-detection device in 1960. The method involves combustion of the vapours eluted from the gas chromato- graph and automatic titration of the hydrogen chloride (or sulphur dioxide) produced, and152 BEYNON AND ELGAR: ANALYSIS FOR RESIDVES OF [.Analyst, Vol. 91 it has been reviewed by CassilS9 and by Challacombe and Mch'ulty.lo0 Its usefulness in residue analysis has been assessed by Burke and Johnsonlol and more recently by Burke and Hols- wade.lo2 They conclude that optimum conditions for the analysis of over a hundred chlorine- or sulphur-containing pesticides are a 6-feet x 4.5-mm i.d.aluminium column, packed with 10 per cent. DC 200 silicone fluid coated on an 80 to 90 mesh Celite support, previously acid and base-washed and silanised. This column is conditioned and is operated at 210" C with nitrogen as carrier gas at a flow rate of 120 ml per minute. The maximum sensitivity of the original microcoulometer (Model R-100) is about 0.1 pg chloride, the range to about 1 mg chloride and precision &3 to 5 per cent. The detector has the great advantage of internal standardisation, the silver ions being electrically generated, For a limit of detection of 0.01 p.p.m.of pesticide, the extract of 10 g or more of crop must be injected onto the chromatograph, and thus for reasonable GLC column life extensive clean-up of extracts is required. In addition, the solvent and any materials emerging from the GLC column before the pesticide are usually vented to the atmosphere to prevent in- complete combustion and fouling of the electrodes. To minimise decomposition of pesticide at the high column temperature that is used, CassilS9 recommended the use of a quartz insert in the injection block. The instrument proved immediately popular, particularly in the U.S.A., being put to use not only for analysis of crop residues103~104y105,106 but also for a great deal of environmental monitoring work.24 9 1 ° 7 ,lo* A modified design (Model K-200) has a sensitivity about ten times better than the original.An ultimate sensitivity of about 10 ng of chloride is claimed for this later instrument, a great improvement. The increased sensitivity makes lower column loadings and lower temperatures feasible, which will decrease the tendency for decomposition of some pesticides. E1ectrol.L capture- The quantitative GLC analysis of halogenated compounds with high electron affinity was found to be difficult with some ionisation detectors because of combination of these molecules with electrons. This disadvantage was exploited by Lovelock and Lipskylog for the selective detection of such molecules. In this detection system the eluted vapour passes into an ionisation cell containing two electrodes.The cathode is in contact with a source of electrons which is usually a few hundred millicuries of tritium. The anode collects the electrons accelerated by a d.c. or pulsed potential, and this produces a standing current of 10-9 to Vapours entering the cell capture electrons to a greater or lesser extent and thus diminish the standing current. This change of current is measured. The decrease in current which is normally found is exponentially related to the number of vapour molecules. The working range extends to about 30 per cent. of the standing current, about the first 5 per cent. being rectilinear, the dynamic range being rather less than a thousand. The sensitivity of the detector varies with the electron affinity of the compound, which itself varies not only with the functional group but also with its molecular environment.l1° The measurement of the electron affinities of many compounds has revealed an interesting correlation with biological activity.lll ,ll2 v113 The potentialities of the high sensitivity of the detector for the quantitative analysis of traces of polychlorinated insecticides were quickly seen and applied65 9959979114 and the purely qualitative aspect has received less attention.The device has drawbacks, such as the lack of specificity for halogen, the differing sensitivity among even the halogenated materials and the need for frequent calibration. However, despite these drawbacks, with ordinary care in ~ p e r a t i o n l l ~ j ~ ~ ~ the detector has been a great success. For halogenated pesticides it offers a sensitivity greater than a thousand times that of the microcoulometer, and high selectivity.The characteristics and performance of the detector have been reviewed by L o ~ e l o c k , ~ ~ ~ Landowne and Lipsky,l16 Dimick and Hartmannll' and Clark,l13 and have been shown to be only slightly affected by minor changes in flow rate and detector temperature. For a residue level of 0-01 p.p.m. the extract from only milligram amounts of samples needs to be injected, and although clean-up is necessary at this level, it can be carried out rapidly and conveniently with small volumes of extract, adsorbent and partitioning solvent. Goodwin et aZ.65 described a procedure for the analysis to a concentration of 0-1 p.p.m. of a variety of chlorinated insecticides with a commercial detector, and showed that increased sensitivity could be obtained with one made to a design by Lovelock.llo Both planar or radial geometry of the radioactive source have since been used, with little apparent difference amp.March, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 153 in performance.The detector has been used many times for the analysis of chlorinated pesticides in crops, soils and animal tissues and products, and optimum conditions have been reviewed recently by Burke and Giuff rida.l18 These workers recommend as operating conditions a 6-feet x 4-mm glass column packed with 10 per cent. DC 200 silicone oil (12,500 cS) on 80 to 90 mesh, acid and base-washed and silanised kieselguhr. This column and the detector are maintained at 200" C and 120 ml per minute of nitrogen is passed through.These conditions do not permit the chromatography of endrin without decom- position.G5 Halogen leak detector- This device was first used as a GTX detector by Cremer et a1.120 for the chlorinated solvents, and was extended to residues of the chlorinated pesticides by Goulden et a1.lz1 It consists of two concentric platinum cylinders that act as electrodes. The inner anode is sensitised by treatment with alkali and is indirectly heated to about 800" C and maintained at 250 volts d.c. relative to the cathode, which receives a standing current of positive ions. The presence of halogen in a vapour passing through the cell induces an increase in this current, which can be amplified (if necessary) and recorded in the usual way.The system has, at the moment, the disadvantage of poor stability. Improvements in design will, it is hoped, eliminate or at least reduce this defect. However, it has the great merit of excellent selectivity for halogen, and about 10 ng can be detected in a simple circuit without amplification, and less than 1 ng in a more elaborate circuit. With these advantages it may well have a promising future. Other selective detectors- Beilstein jlame detector-The Reilstein test has been adapted for use as a quantitative GLC d e t e c t ~ r . ~ ~ " l ~ ~ All or part of the effluent from the GLC column is led through a copper thimble held in the hot zone of a Bunsen burner and an intense green flame (Amax. =473 mp) indicates the presence of halogen.The sensitivity by visual observation is in the microgram range for halogen, but this can be increased by the use of instruments. Quantitatively, the time of appearance and disappearance of the coloration can be related to the response from another detector such as a katharometer, coupled prior to the flame or in parallel with it. Solzrtion-cozzdztctivity dete~tor-Sternbergl~~ has made use of the properties of combusted vapours to devise a selective detector. The gases emerging from the GLC column, after combustion, contact a flowing film of distilled water. Any soluble, ionised components produce a change in the electrical conductivity. The claims for sensitivity (6 x 10-9 g for DDT, with signal equivalent to twice the noise level), selectivity (lindane is 4000 times more sensitive than hexane) , and repeatability are g00d.l~~ 91zGy127 Although both of these detectors have involved careful development of ingenious ideas, it would seem that the other selective devices previously mentioned have more immediate promise, and these two will probably not be refined for general use.COLUMN AND INJECTION BLOCK MATERIALS- In its early days GLC was primarily used by workers in the petroleum industry with much emphasis on the high-temperature separation of the inert hydrocarbons, and on this account the problem of decomposition of more thermally labile compounds has sometimes been overlooked. Exposure to hot metal surfaces or carbonised deposits in the injection block can bring about decomposition of many pesticides such as DDT,12* dicofol129 and particularly e n d ~ - i n .~ ~ 91199128 The composition of the column tubing also plays a part in decomposition. Beckmann and BevenuelZ8 compared four materials and found increasing recovery of the insecticide injected with columns of copper, stainless steel, aluminium and quartz, respectively. Dimick and Hartmann117 have shown that Pyrex glass is as effective as quartz. Cassilg9 has overcome decomposition in the injection block by using a quartz liner to the block, and borosilicate glass may be substituted for ~ l u a r t z . l l ~ J ~ ~ I t is becoming normal practice in analysing thermally unstable compounds such as pesticides to dispense with an injection block and to inject directly on to the GLC column in order to avoid decomposition at this point.154 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [A?Zd$St, VOl.91 SOLID SUPPORTS- Although capillary columns have been used for analysis of pesticides131 their evaluation for residue analysis has so far proved di~appointing.~~~ In an attempt to use the high efficiencies of capillary columns to secure more positive identification, the drawback of low column capacity proved insuperable, The extreme dilution of pesticides in residue extracts means that the stream-splitting normally used with capillaries is impracticable. Wider-bore capillaries with their higher capacity may improve this picture, but it is doubtful if a resolution superior to that of packed columns can be obtained in a reasonable time. With packed columns the optimum column bore would appear to be 3 to 4 mm (0.125 t c 0.160 inches), and for best column efficiency the particle-size range of the support material should be narrow (a 10 or 20 mesh fraction). Kieselguhr in its various grades has remained the most popular material, being inert, having a high ratio of surface area to volume anc with reasonable mechanical strength.Recently, a harder and less adsorptive supporl specially developed for GLC (Chromosorb G) has been introduced. Glass micro beadsG5 haw been found to offer no advantage. Low loadings of stationary phases must be used and thc packing of the column can be difficult. The lack of porosity of glass beads can lead to higf back pressures. For the compounds under review the most important property of a support materia is the absence of any tendency to cause decomposition.Goodwin et aL.65 located the mail source of decomposition in the support material and showed that nanogram quantities o endrin could be chromatographed successfully by using small amounts of a polar stationar: phase (Epikote resin 1001) in admixture with the non-polar phase to act as an active-sit suppressor. The proportion of resin used (10 per cent. of the silicone) was insufficient tl modify the resolution of the pesticide mixture given by the silicone. The same worker later found that other epoxy additives could be used132 and that pre-treatment of the suppor by refluxing with epichlorhydrin gave chromatography without decomposition when silicon alone was used as stationary phase. This approach has also been used by Gunther et aL1: TABLE IV RELATIVE RETENTION TIMES OF CHLORINATED PESTICIDES~~~ Pesticide Lindane .... .. I-Eeptachlor . . .. Dicofol . . .. .. Aldrin* . . .. .. op’-TDE olefin . . .. Heptachlor epoxide . . Chlorbenside . . .. pp’-TDE olefin . . .. y-Chlordane . . .. @’-DDE .. .. Endosulfan . . .. Chlorfenson . . . . p-Chlordane . . .. pp’-DDA methyl ester . . pp’-DDE.. .. . . Dieldrin . . .. . . op’-TDE . . .. .. Endrin . . .. .. Chlorobenzilate . . .. Pp’-TDE .. .. op’-DDT.. . . pp:-Methoxychlor olefin * p p -DDT . . .. pp’-Methoxychlor . . Tetradifon . . . . Chlordane . . .. Toxaphene .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. * . .. .. . . .. .. .. .. .. .. .. .. .. .. .. .. .. Relative retention time 0.52 o*s1 0.99, 3.82 1.00 1.12 1-20 1.35 1.34 1.35 1.37 1.44, 1-91 1.46 1-46 1-48 1-69 1-72 1-72 1.90, 2.09 2-07 2.12 2.2 1 2.52 2-70 3.90 4-29 0.52, 0.66, 0.74, 0.82 0.06, 1.14, 1.36, 1.50, 2-3 1.29, 1.45, 1.55, 1.87, 2-17 2.45, 2-85, 3.36, 4.03, 4.52 * Aldrin retention time 5.5 minutes.Column of 10 per cent. DC 200. Carrier-gas flow rate: 120 ml per minute.21 Major peaks are in italics. Temperature of 210’ C.March, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 155 who used tris-(2-biphenylyl)phosphate (Dow K-1110) as a stabiliser, but Beckman and Bevenue128 have reported that this is not always successful. The adsorptivity of the support material has also been reduced by acid and alkali washing, and active hydroxyl groups have been eliminated by conversion to their trimethylsilyl derivatives with hexamethyldisilazane.lo2 t118 STATIONARY PHASES- The stationary phase is selected for the separation it offers of compounds actually or potentially present in the mixture to be analysed.Experience to date has shown that a separation involving the use of silicone compounds based essentially on differences in vapour pressure gives the best resolution of the chlorinated group of pesticides. Work with the dimethyl siloxane polymers DC 200,100~102~118 SE 30,135 SF 96,13e Dow 111179137 and E 301659130 and with the almost non-polar methylphenyl siloxane polymer DC 710138 has been published and tables of retention times have been given by C a ~ s i l , ~ ~ Bevenue,138 Burke and Holswade102 and Burke and Giuffrida.lls The relative retention times of the compounds under review on a non-polar phase are given in Table IV.From this table it can be seen that on the silicone phase there are pairs of compounds difficult to resolve, notably $$'-DDE and dieldrin, $$'-TDE and o$'-DDT, aldrin and dicofol, endosulfan and chlorfenson, heptachlor epoxide and chlorbenside. Other stationary phases with greater polarity have been chosen for the ability to separate these pairs and also to resolve a pesticide from a peak suspected of being due to a co-extracted natural product. Apiezon hydrocarbon greases,130 9139 the silicone polymers modified with fluorine-containing groups38*140 or with nitrile groups (GE XE 60132 and XF 1112132), epoxy resins,65 polyesters141 and the Sonidet P40139 have been used to give this extra resolution with chlorinated insecti- cides, but Ucon and other esters143 91459146 have been used for other pesticides.TEMPERATURE- Excessively high column temperatures lead to decomposition and loss of resolution. The original work with pesticide^^^^^^ was carried out at about 250" C, but the advent of more sensitive detectors has meant, among other advantages, that lower operating tempera- tures can be used and 200" to 210" C has recently been recommended.102~118 Goodwin et al.65 have shown that temperatures as low as 160" C give better chromatography with reasonable retention times with a lower percentage of stationary phase (2.5 per cent.) with 0.25 per cent. of epoxy resin added to prevent decomposition on the uncoated support.To give efficient vaporisation and to prevent condensation the injection block (if used) and the detector must be at a temperature at least as high as that of the column, but it is not necessary for them to be much hotter. Burke147 has evaluated temperature programming for the analysis of chlorinated pesticides, and has shown that a complex mixture of compounds with a range of vapour pressures could be conveniently separated, but the procedure did not achieve separations that were not possible with isothermal operation. Electron-capture detectors containing tritium as the source of radiation cannot be operated at temperatures above 200" C.148 9149 MEASUREMENT IN GLC ANALYSIS- The microcoulometer gives an absolute measure of the weight of compound eluted and burnt, but with other detecting systems the amount of pesticide that each peak represents may be found by calibration with an internal standard, but preferably with an external standard.The peak area or peak height may be measured, although the measurement of the former generally gives more reproducible results150 except for peaks of short retention time. IDENTIFICATION- In the absence of valid untreated control material a single retention time obtained by GLC is not sufficient evidence for the positive identification of a compound. Many naturally occurring products can respond to the GLC detectors used for residue analysis. Although the relative response to electron-capture detectors of these naturally occurring co-extractives will generally be much lower than that of the chlorinated pesticides, they are often present 9143 the polyglycol amide Versamid 90OLa :- n m*-rrh h i m h f i v r-nmr-nmtrqt;nn156 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [A?dySt, Vol.91 There are several methods available of improving the specificity of GLC. Columns with better resolution may help132 and retention times on two or more stationary phases of different type offer additional, though not conclusive, evidence.139 The several-column, single-detector chromatograph produces a characteristic “spectrum” from a single injection and is useful for fairly simple p r ~ b l e m s , l ~ ~ J ~ l but is difficult to interpret for more complex mixtures. Alternatively, information on the identity of a compound may be given by comparison of the responses of the compound to different selective detectors, and this has been successfully demonstrated by Goulden et aZ.121 with the electron-capture and the halogen-leak detectors.Conversion into characteristic derivatives prior to GLC has been exploited by several groups of workers for specific pesticides. Klein et aZ.152 converted DDT and TDE to their ethylene derivatives with alkali, and Beckman and Berkenk~tterl~~ used reduction with sodium in liquid ammonia. Conversion on the nanogram scale can be carried out readily and pre-columns filled with suitable reagents have been used by many workers.154J55~15G~157 9158 Such methods seem capable of offering conclusive evidence of structure. The use of GLC in combination with other procedures is considered later in this section.TOTAL HALIDE ANALYSIS Analysis of pesticide residues by total halide has been thoroughly reviewed in Volume I of the treatise edited by 2weig.l Wet, dry and oxygen-flask combustion and various types of sodium reduction followed by colorimetric or electrometric end-point determination were discussed. Details of the sodium - diphenyl reduction have been given in Volume 2 of the same text, and Grou and Balif159 have used sodamide in paraffin oil for dehalogenation. By present standards the limit of detectability of this method is low, (microgram amounts of halide) ; it is impossible without prior treatment to distinguish between the components of a mixture, and few workers in recent years have concentrated on it. This latter point also applies to neutron activation, another total-halide technique, reviewed by SchmittlG0 and Guinn and Schmitt .lG1 In addition, the sensitivity for chlorine is not particularly high, about 0.1 pg for the 37Cl (n,y) 38Cl reaction, but rather better for bromine, 0.005 pg for the 79Kr (n,y) soBr reaction.This could be a powerful method where many similar samples need to be analysed, but the disadvantages of non-specificity and lack of availability do not encourage its development. PCLAROGRAPHY Polarography has not been used extensively for the analysis of chlorinated pesticides because the method has not the sensitivity of several of the alternative procedures, and also because extensive clean-up of plant and soil extracts is required before analysis by this method. Furthermore, several of the chlorinated pesticides are not reduced during polaro- graphy.The procedure is likely to find more application for the analysis of organo-phosphorus compounds rather than for chlorinated pesticides. The method has been investigated, however, for application to chlorinated pesticides as it is capable of quantitative analysis and also the measurement of the half-wave potentials of the reducible components confers a degree of specificity. Polarography is extremely sensitive for the analysis of reversibly-reducible inorganic Components but the sensitivity is poorer for the analysis of irreversibly-reduced organic compounds. For organic compounds pulse-wave polarography is the most sensitive form of this procedure. Gajan,lG2 R t - a ~ e e ’ ~ ~ and Gajan and Link164 have reviewed the applica- tion of polarography to pesticide analysis.DDT and its metabolites can be determined by p01arographyl~~J~~ and tetramethyl ammonium bromide in aqueous acetone has been used as the base solution167 to determine concentrations of 5 pg per ml. Analysis of DDT by polarography has also been reported by other w ~ r k e r s . ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I t has been reported170 that the alkyl chlorine atoms of DDT are reduced more easily than the aromatic chlorine atoms. Polarography of DDT after nitration has been reported by Davidek and Janicek.171 The polarography of dieldrin has been described by S w a n e p ~ e l , ~ ~ ~ and K ~ s m a t y i l ~ ~ ha5 used polarography for the analysis of residues of DDT, y-BHC and heptachlor in crops and soils after chromatography.Kosmatyi and Shlyapak174 have determined residues of DDT in a range of crops with a limit of detectability of 2.1 x Other references to the polarography of these compounds and to methoxychlor are given by Gajan and Link.lG4 M (about 7 pg per ml).March, 19661 CHLORINATED INSECTICIDES AND ACA4RICIDES 157 BIOASSAY The use of bioassay for the determination of pesticide residues, including residues of chlorinated pesticides, has been reviewed within recent years by Needham,175 Dewey,l76 Sun1" and P h i 1 l i ~ s . l ~ ~ Most of the work that was reported up to 1962 is considered, and many references are given to the application of bioassay in the determination of residues of chlorinated pesticides. Bioassay has been used extensively because it is capable of detecting very small amounts of toxicants.Only the simplest apparatus and a ready supply of test organisms are required. A wide range of test organisms has been used, including vinegar flies (Drosophila melano- gastev) ,175 9 1 7 6 9 1 7 7 9 1 7 8 house flies (Mzisca domestzca) ,175 9 1 7 6 3 1 7 ' 3178 adult and larval mosquitoes ( A edes, A izopheles and Czdex) ,1i5J76 7177 J~~ alfalfa weevil larvae (HjPera postica) ,lS9 brine shrimps (Artemia s a l i m ) ,177 water fleas (Daphnia m a p a ) ,177 niicrocrustacea (Gammarzzs laczistris) ,l80 guppies (Lebistes reticzdatus) ,175 J~~ 3177 ?li8 and goldfish (Carassizis aztratzis) .177 Uavidow and Sabatinol*l and Dewey and Parkerls2 have described a system for the mass rearing of L>a$/ztzia magna for bioassay, and Geroltls3 has described methods for the breeding of Drosoplzila melawgaster for bioassay.Satisfactory methods have also been des- cribed for house flies and mosquito larvae and are given by Needham.175 Of these test organisms the vinegar fly, house fly and mosquito larvae have been used most extensively. All three species are easy to rear and are sensitive to toxicants. Sun177 has made a useful summary of the susceptibility of several test species to chlorinated insecticides and, although DrosoPhila melattogastev are the most versatile, the choice of test organism will often depend on the pesticide to be assayed and on the possible contaminants. The three basic techniques generally used are (i) direct methods in which the material containing the pesticide is fed directly to the test organism, (ii) film methods in which a film of pesticide is deposited on a surface by evaporation of an extract of the sample followed by exposure of the test organism to the film and (G) aqueous methods where the sample extract is transferred to water that contains the test organism, such as fish or larvae.The direct-feeding method is less sensitive than the other methods but it is quicker and is useful when toxicant levels are high. The results in Table 1' indicate that whilst the aqueous methods can be used to detect pesticides in very dilute solutions the absolute sensitivity in micrograms is not as good as that of the dry-film method. In using these procedures it is important that the test conditions should be standardised and the follou~ing factors must be considered.Naturally occurring compounds can be toxic or can have a masking effect and clean-up is often necessary to achieve the highest sensitivity. Mosquito larvae are particularly sensitive to co-extractives. The susceptibility of a test organ- ism will vary with the age and sex of the species, for example, male Drosophila melanogaster are more susceptible to toxicants than females, and the susceptibility generally decreases as the age increases. Henneberry et aZ.18' have described a simple method in which a rising air stream is used to separate the male Drosophila melaiiogaster from the females. The presence or absence of food during the bioassay can also affect the susceptibility.188.189 Apart from these biological factors, physical factors such as the number of test animals, the size of the container, the time of exposure, temperature, humidity and illumination should be st andardised.The mortality obtained with the assay sample is compared with that obtained by using standard amounts of pesticide. Several methods of calculating the results are described by Sun177; a plot of percentage of mortality on a probit scale and amount of pesticide on a logarithmic scale is generally rectilinear and can be used within the range of 20 to 80 per cent. mortality. The susceptibilities of some test organisms by using the three basic procedures are sum- marised in Table V. The values are not from the same source since a direct comparison of all the basic procedures has not been made. The susceptibilities are generally low in the presence of plant extracts due to masking.The masking effect of a different combination of crops and toxicant has been discussed in detail by Phillips.178 When toxicants other than the compound to be assayed are absent, the results of bioassay agree well with the results obtained by chemical or physical methods.176 9178 Bioassay is extremely useful in parallel screening as the comparison of the results of bioassay with those of specific methods may indicate the presence of other toxic components that may be metabolites of the parent compound. Bioassay results and the results of other158 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [AnU&Si!, VOl. 91 non-specific methods may be used in combination to provide added confirmation of the presence of a toxicant.However, the use of bioassay has declined during recent years because of its lack of specificity and also because of the difficulties, in comparison with some other procedures, of obtaining reproducibility. Aldrin .. Dieldrin . . BHC . . . . Chlordane . . Endosulphan Endrin . . Heptachlor . . Toxaphene . . Chlorobenzilate DDT . . .. TDE . . . . Rlethox ychlor y-BHC . . TABLE V THE SUSCEPTIBILITIES OF TEST SPECIES TO TOXICANTS Dry-film method LD,, values* (pg per container) .. .. 0.05 .. .. 0.05 .. .. 1.0 .. . . 0.10 .. .. 0.20 .. .. 0.27 .. . . 0.30 . . . . 0.30 . . . . 10 . . . . 32 .. . . 15 .. . . 1s . . . . 500 Direct feeding? per cent. Aqueous solution: mortality for given LC50 dosage in p.p.m. (P.P*m.1 0.2 p.p.m. 89% 0.0088 0.5 p.p.m. 64% - 0.2 p.p.m. 98% 0.017 - - - - - - 0.5 p.p.m. 54% 0.017 0.1 p.p.m. 55% - 10 p.p.m. 2076 - - - 2.0 p.p.m. 0% 0.035 - - 5-0 p.p.m. 0% - * Ref. 184. t Ref. 185. $ Ref. 186. 25 x 200-mm tubes, 24 hours exposure of twenty, 7 to 31-hour old Drosophzla Twenty grams of macerated potato containing pesticide were exposed to Twenty-five 4th instar larvae of Culex p . quinquefasciatus in 100 nil of water melanogaster to pesticide in the absence of plant extracts. 50 one-day old Drosophzla melanogaster for 24 hours. exposed for 24 hours to the pesticide in the absencc of plant extracts. Several attempts have been made to improve the specificity of bioassay. Methods involving a combination of chemical and physical clean-up have been used.lgl Also, the comparison of the response of several different test-organisms to the toxicant can be used to fingerprint the ~ e s t i c i d e .l ~ ~ 7 ~ ~ ~ Other methods of achieving specificity by clean-up, by comparison of the ratio of photomigration, by comparison of rates of mortality, and by the use of several different test-species, have been summarised by Sun,177 Dewey176 and Needham.175 Although bioassay is useful in parallel screening procedures and for the confirmation of results of other methods, its use cannot be recommended as the sole procedure to be adopted for the measurement of residues of chlorinated pesticides. COLORIMETRIC AND ULTRAVIOLET-SPECTROMETRIC ANALYSIS Colorimetry, with ultraviolet or visible radiation, has been until recent years the main source of specific residue methods. Sensitivities of a few micrograms and reasonable specificity have been attained for the whole group of pesticides under review, and a full discussion and details for thirteen of the group are given in volume 2 of the text edited by Zweig.l The need for lower levels of measurement and increasing emphasis on the analysis of metabolic products are being met by progress in other methods with much higher intrinsic selectivity and sensitivity, and the usefulness of colorimetry is declining.Comparisons have been carried out between ultraviolet and visible spectrometry and other methods, and agreement of the results has been good.993194 INFRARED ANALYSIS Infrared spectroscopy is one of the few specific methods available for residue analysis, but there are considerable difficulties involved in its application.The first difficulty, that is gradually being overcome, is the need for a relatively large amount of the pesticide to obtain a spectrum. The second difficulty is the need for extensive clean-up and this difficulty is accentuated by the concentration required before analysis. However, in spite of these difficulties infrared spectroscopy has often been used for the quantitative and qualitative analysis of residues of chlorinated pesticides. Its application in this way has been reviewedMarch, 19661 159 by Blinn and Gunther,lg5 Brucelg6 and by Frehse.lg7 These reviews give many references to the analysis of residues of aldrin, chlorbenside, chlordane, DDT, dieldrin, endrin, endo- sulphan, lindane, methoxychlor and tetradifon in a range of samples including crops, soils, air and water, and these references will not be repeated here.Blinn and Guntherlg5 have also reviewed the use of the different infrared regions and have compared the use of solutions, mulls and potassium bromide pellets. In much of the earlier residue analysis 60 pg or more of pesticide was required to obtain a recognisable spectrum. Smaller amounts could be analysed quantitatively by using the strongest absorption band for measurement, but such procedures lose some of the advantage of the specificity. Blinn et aZ.1g8 carried out measurements with 0.3 ml of carbon disulphide solution in a 5-mm light-path, sodium chloride cavity-cell and, under these conditions, 214 pg of aldrin in solution produced an absorbance (1250 cm-l) of 0-1 unit.Later, Johns and Braithwaitelg9 described a micro-specular attachment which reduced the amount of sample needed to the range 25 to 50 pg. Kreuger and \‘olkmann200 have des- cribed a micro attachment for an infrared spectrometer that will increase the instrument’s sensitivity 40-fold and will measure the spectrum of 0-1 ,ug of lindane. Chen201 has obtained recognisable infrared spectra with 1 pg of methoxychlor and other pesticides. Recently Crosby and Laws54 obtained recognisable spectra for 5-p1 solutions containing 5 pg of pesticides by using cavity-cells of 1-mm path length. This last work was described for organophosphorus compounds, but will be applicable to chlorinated pesticides, although not with quite the same limit of detectability.Sparagana and 1CIason202 have described a dual-beam condensing unit that uses reflecting- type optics and have obtained infrared spectra by using 0.05 pg of a compound. Although they did not apply the procedure to pesticides the method shows great promise for such applications. Infrared spectroscopy can often be used, after GLC, to confirm the identity of a component, and Giuffrida203 has described a useful method of trapping-out components from GLC columns directly on potassium bromide powder. The sensitivity of infrared spectroscopy has been increased considerably during recent years, and because of its specificity the method will assume increasing importance for residue analysis in the future. FLUORIMETKY Fluorimetry is a sensitive method for the analysis of many compounds, and characteristic activation or emission spectra can often be obtained.M a c D ~ u g a l l ~ ~ ~ and H ~ r n s t e i n * ~ ~ have reviewed the application of fluorimetry to residue analysis but make no mention of its application to chlorinated pesticides. Fluorimetry has not been used for the analysis of the chlorinated pesticides considered here, because they are insufficiently fluorescent. Although they could be converted to fluorescent compounds by suitable chemical methods, such studies are hardly worthwhile in view of the extensive clean-up that is necessary before fluorimetric analysis, and also because several sensitive methods are already available for the analysis. C H LO I< I h’ AT E D I N S E CTI C I D E S ,4 h’ D AC A R I C I D E S PAPEK CHROMATOGKAPHY Paper chromatography has been used e ~ t e n s i \ ~ e l y ~ ~ y206 ,20T *208 3209 for the separation of mixtures of pesticides in extracts of plants, animal tissues and dairy products, and for the qualitative and quantitative analysis of pesticides. Both aspects will be considcred here.The application of paper chromatography for clean-up has been discussed in a previous section. The RF values obtained by paper chromatography are useful for the identification of the components, but a single K , value is not sufficient for positive identification unless it can be used in conjunction with another parameter such as the GLC retention time. Paper chroma to- graphy is generally used not for clean-up of extracts, but for the identification of pesticides, and the clean-up by other methods is generally necessary to achieve the maximum sensitivity of detection.37 y 2 0 6 Early work on the paper chromatography o f pesticides has been reviewed by Block et aZ.211 and the more recent work has been summarised by Zweig212 and by McKinley.213 The paper chromatography of pesticides was described by Mitchell in a series of In one of these222 the chromatography of 114 pesticides including almost all of the compounds considered here was described.Mills37 showed that chlorinated pesticides in a range of fruits, dairy products and animal tissues, could be readily detected by using160 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [AndySt, VOl. 91 paper chromatography and similar work was reported almost simultaneously by McKinley and Mahon.206 The basic techniques established by Mitchell, Mills, McKinley and Mahon or minor modifications of them are those used widely with success today.PAPERS, AND MOBILE AND STATIONARY PHASES- Whatman No. 1 filter-paper has often been used for paper chromatography, and if phenoxyethanol-silver nitrate is to be the spray reagent the paper is generally washed before the chromatography to remove background interference. Washing procedures have been described with water aqueous silver nitrate followed by water,206 and aqueous silver nitrate followed by ammonia - water.209 Most paper chromatography has been carried out with impregnated papers, and in Table VI are listed a few of the combinations of mobile and immobile phases that have been found to be useful for the separation of chlorinated pesticides, together with the corre- sponding RF values.The preferred system will depend on the compounds to be separated, and the use of different systems is recommended to obtain several RF values if more evidence of identification of a compound is desired. DETECTION- Phenoxyethanol- silver nitrate as a spray reagent followed by ultraviolet irradiation has been shown by to be capable of detecting as little as 0.01 pg of some chlorinated pesticides. has also shown that 4 to 5 minutes of ultraviolet irradiation is the optimum without the production of an excessive background, and he has also compared the limits of detectability with and without water-washing prior to chromatography.Although washing has been an accepted and recommended procedure for some time, these recent results show that the process has little effect on the limits of detectability for many chlorinated pesticides. Indophenol blue in the presence of an aliphatic acid226 and methyl yellow followed by ultraviolet irradiation227 have also been used as spray reagents for paper chromatography. Several, but not all, of the spray reagents used for TLC are also applicable to paper chromatography . When the pesticides are not resolved from one another or from co-extractives they can sometimes be identified by the use of specific spray reagents. TABLE VI RF VALUES OF CHLORINATED PESTICIDES BY PAPER CHROMATOGRAPHY R p values RL values (RF relative to lindane) I A 1.A - Component 2,2,4- 2,2,4- Mobile phase 2-Methoxy- Trimethyl Trimethyl pentane222 pentaneZo6 Immobile phase Soya-bean 2-phenoxy- 2-phenoxy- oil ethanol ethanol Aldrin . . . . .. Chlorbenside . . .. Chlordane . . .. Chlorfenson . . . . Chlorobcnzilate . . y-BHC.. . . .. op’-DDT .. .. pp’-DDT .. .. Dicofol .. .. Dieldrin . . .. Endrin .. .. Heptachlor . . .. Heptachlor epoxide . . Methoxychlor . . .. Oxythane . . .. Tetradifon . . .. Toxaphene . . .. 0.37 0.92 2.63 0.65 0.42 1.0 0.34 0.65 1.40 0.08 to 0-57 0.8 to 1.0 1-83 0.74 0.3 1 0.46 0.79 0.39 2.26, 0.52 - 0.80 1.98 0.34 0.67 1-66 0.69 0-30 0.31, 2-11 0-42 0.64 2-64, 1-55 0.41 0.64 1-48 0.3 1 0.94 2.11 0.38 0.77 - 0.68 0.30 0.67 - - 1-23 0.48 0.16 - 0.05 to 0.48 0.36 to 0.92 - and streak Figures in italics are the main components.40 per cent. aqueous pyrid ine*06 mineral oil 0.27 1.0 0-G1 streak 1.00, 1-28 1-69, 0.73, 0.30 0.55 0.55 1.09, 0.28, 0.93, 0.40 0.08, 0.33, 0.73 0.7 1 0-38 1.09 0.42 - - - 70 per cent. aqueous acetone206 mineral oil 0.10 1.0 0.32 0-18 1.00 1-12, 0.85, 0-55 0.18 0.18 0.73, 0.35, 0.09 1.82, 0.33, 0.14 0.45 0-15 1-24 - 0.11, 0.50March, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 161 QUANTITATIVE ANALYSIS- Quantitative analysis of pesticides can be carried out after paper chromatography by measurement of the size or intensity (or both) of the separated spots and comparison with standards. The principles of such measurements for quantitative analysis by using paper chromatography and thin-layer chromatography are similar and are discussed together under the heading Thin-layer Chromatography. THIN-LAYER CHROMATOGRAPHY (TLC) TIX has been used many times for the qualitative and quantitative analysis of chlorinated pesticides and these aspects will be considered here.The application of TLC for the clean-up of extracts has been discussed briefly in a previous section. Conkin22s has reviewed the early work on the application of TLC to residue analysis, and other early work on the TLC of chlorinated pesticides has been summarised by Ganshert et aZ.229 Walker and Ber0za~~0 studied the separation of 68 pesticides by TLC, and Kovacs231 and Morley and Chiba232 have successfully applied the method to the analysis of residues of chlorinated pesticides in crops. Recently Abbott et aL2= have studied the separation of chlorinated pesticides by TLC over a wide range of conditions and some of their results are summarised in Table VIT.TABLE VII RF VALUES OF CHLORINATED PESTICIDES BY TLCW RF x looinsystem Compound Aldrin . . .. pBHC . . .. pp'-DDE . . . . op'-DDT . . . . pp'-DDT . . . . Dieldrin . . . . Endosulfan A . . Endosulfan B . . Endrin . . . . Heptachlor epoxide Methosychlor . . Heptachlor . . . . PP'-TDE . . , . 7 1 88 87 71 72 69 - - 82 - 66 - 2 98 58 98 90 91 58 - 98 - 77 3 73 87 74 78 53 - - 69 - 58 4 58 74 50 52 30 - - 48 38 67 - 5 69 37 62 58 54 48 52 52 62 36 46 - - 6 7 8 9 1 0 1 1 67 70 79 64 62 67 27 - 47 18 19 46 61 68 73 57 56 65 54 62 71 46 48 59 49 60 69 39 40 57 41 46 63 48 54 65 47 63 61 35 31 58 12 42 58 65 26 26 49 61 65 73 53 52 65 39 27 30 45 10 13 - 33 45 59 26 28 52 - - - - - - - - - - Key to systems 12 98 94 98 97 98 88 94 86 88 88 88 92 - 13 14 i5 82 95 70 78 95 65 73 89 50 69 89 42 52 37 12 64 65 17 9 4 2 61 51 13 78 95 58 57 49 17 57 71 25 55 78 - - - - System NO.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Plate Silica gel - alumina (1 to 1) Silica gel - alumina (1 to 1) Silica gel Silica gel Silica gel Silica gel Silica gel Silica gel Silica gel Silica gel Silica gel Kiesclgu hr Alumina Alumina Silica gel ~ - 3 Mobile solvent Cyclohexane - liquid paraffin (20%) Cyclohexane - silicone oil (8%) Cyclohexane - hexane (1 to 1) Cyclohexane - benzene (1 to 1) - liquid paraffin (10%) Cyclohexane - liquid paraffin (20%) - dioxane (10%) Cyclohexanc - liquid paraffin (2076) - dioxane (5%) Cyclohexane - liquid paraffin (10%) - dioxane (3.5%) Cyclohexane - liquid paraffin (5%) - dioxane (2%) Light petroleum (40" to 60') - liquid paraffin (20%) Light petroleum - liquid paraffin (10%) Light petroleum - liquid paraffin (5%) - dioxane (1%) As 11 As 11 Hexane Hexane The measurement of the RiF value may assist in the identification of a chlorinated pesticide although a single RF value is insufficient evidence.Quantitative analysis may be carried out by measuring the spot area and intensity. To achieve the greatest sensitivity the layer is generally washed prior to the separation to remove interfering compounds from the absorbents and the extracts are often submitted to clean-up by other methods before the thin-layer chr~matography.~~~ However, Morley and Chiba232 were able to measure residues of aldrin, DDE, DDT and heptachlor in wheat, down to 0.1 to 0.2 p.p.m.(0.05 to 0.1 pg) without prior clean-up. After rigorous clean-up162 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [Analyst, Vol. 91 of extracts, however, K o ~ a c s ~ ~ l was able to detect residues of many chlorinated pesticides in a range of foodstuffs at the p.p.b. (parts per thousand million) level. ADSORBENT AND DEVELOPING SYSTEM- Silica gel, alumina, kieselguhr and some mixtures of these have been used as adsorbents. Kieselguhr is reportedz33 to be of limited use for the separation of chlorinated pesticides and silica gel has been the most widely used. Chloroform alone or mixed with polar solvent~,2~0 hexane2= or h e ~ t a n e ~ ~ l and cyclohexane alone or mixed with polar solvents have been used successfully for development.The use of a simple system such as hexane or heptane and silica gel or alumina will effect the separation of many chlorinated pesticides.231 3233 I t is not possible to recommend one system for general application since no one combination of adsorbent and developing solvent will separate even the small number of compounds considered here. However, by suitable choice from one of the many systems now available230~231~233 any pair of compounds could be separated. Furthermore, Abbott et aZ.233 have shown that resolution of chlorinated pesticides, not possible at room temperature with a given system, may be effected at a higher or even lower temperature. DETECTION- The most commonly used locating agent is phenoxyethanol - silver nitrate followed by ultraviolet irradiation.K o ~ a c s ~ ~ * could detect 0.01 pg of aldrin, DIIE, heptachlor and its epoxide, DDT, TDE, endrin, dieldrin, methoxychlor and dicofol, 0.05 pg of BHC and 0.1 pg of toxaphene and chordane. The adsorbent layer was washed with water prior to the chromatography; the limits of detectability may be higher without this washing because of the presence of inorganic chlorides in the adsorbent. Morley and Chiba232 and Yamamura et aZ.234 used ammonium hydroxide in place of phenoxyethanol. Abbott et aL2% reported that treatment with ethanolic silver nitrate by ultraviolet irradiation was adequate for the detection of chlorinated pesticides on TLC. Abbott et aZ. obtained interesting results by using combinations of aqueous silver nitrate and a pH indicator (with or without ultraviolet irradiation) and recommended the use of bromophenol blue - silver nitrate,235 but did not report the limits of detectability.Other spray reagents that have been used for the detection of chlorinated pesticides on TLC include 0.1 N potassium ~ermanganate,"~ zinc chloride or iodine plus di~henylamine,~~T silver nitrate - formaldehyde - potassium hydroxide - nitric acid - hydrogen peroxide in succession,238 P-dimethylaminoaniline hydro~hloride,~~~ alcoholic o-toluidine or o-dianisidine,=O sulphuric iodine230 and bromine - fluorescein.230 M'ith several of these reagents ultra- violet irradiation is necessary. The limits of detectability obtained with these reagents are greater than those claimed by Kovacs, but many of them will detect 5 pg or less.Although silver nitrate in some form is the most common reagent the other reagents mya serve to increase the certainty of the identity of a particular spot. QUANTITATIVE ANALYSIS- The principles for the quantitative measurement of the spots of chlorinated pesticides in thin-layer and paper chromatography are the same as that for other compounds as outlined by Purdy and Truter,241 and by T r ~ t e r . ~ ~ ~ Often the spot area alone has been measured and is related to the amount from a calibration graph, preferably obtained by the chromato- graphy of standards on the same plate. The comparison is often made visually how- ever.37,206,232,243 The area of the spot can be related to the amount of compound present211 but the relationship is not rectilinear.Evans210 found that a rectilinear relationship existed between spot area and log (amount) in the range 2.5 to 14 pg for several chlorinated pesticides, and he obtained a precision of 10 per cent. for the measurement of 2.5 pg or more of these pesticides with paper chromatography. Purdy and T r ~ t e r ~ ~ l have reported excellent repro- ducibility ( * 6.6 per cent. in the 2 to 30 pg range) with a rectilinear calibration-graph obtained by plotting (area): against log(weight), a relationship first used by Fisher et aZ.244 Zweig212 obtained rectilinear plots in the 0.5 to 44pg range for DDT by plotting amount against area and density. I t is likely that relationships that consider both area and density will give the most consistent results for a wide range of compounds.Densitometric measurements may be made photometrically by scanning the plate or paper, or by scanning a photograph of them.March, 19661 TLC is rapid and quantitative in future years. A difficultv CHLORINATED INSECTICIDES AND ACARICIDES 163 the equipment is inexpensive and it can be used for clean-up, qualitative analysis. These advantages will undoubtedly be exploited even further THE POSITIVE IDENTIFICATION OF PESTICIDE RESIDUES )ecoming more and more prominent in this field of anal]& is that of establishing t h i identity %f a residue. Mani naturally occurring products *in the extract can respond to the analytical procedure and can be mistaken for pesticides when a valid control is not available.Such components are not always removed by the clean-up procedure that is used. I;or this reason GLC, total halide analysis, polarography, bioassay, colorimetry ultraviolet analysis, fluorimetry, thin-layer or paper chromatography cannot normally be regarded as specific methods, and cannot usually provide positive identification of a given component in the absence of a valid control. Inirared analysis can be reasonably specific but relatively large amounts of pesticide are required as well as rigorous clean-up of the extracts. These problems are being overcome gradually, and infrared analysis has been used successfully for the positive identification of residues of chlorinated pesticides after clean-up by gas - liquid chromatography, column chromatography or by thin-layer chromatography.Mass spectrometry can generally provide a positive identification of a cornponcnt, but little has been reported so far on the application of this technique to residue analysis. The problems are similar to those encountered with infrared analysis, in that rigorous clean-up and relatively large amounts of compound are required. However, Ryliage245 has obtained recognisable mass spectra with 20 ng of esters of long-chain carboxylic acids that were passed directly from a capillary GLC into a mass spectrometer. Whilst such equipment is expensive it merits application for identification of residues. Such a procedure may perhaps be more convenient than infrared spectroscopy as the effluent from the GLC column is already in the vapour phase, and can be fed directly into the mass spectrometer with perhaps only limited interference from co-extractives and without the need for trapping-out components.These two procedures can, at best, offer conclusive evidence of identity. The specificity of other methods can be improved, but such improvements cannot make interpretations from them completely unambiguous. A single retention time obtained by GLC or a single RE’ value obtained by thin-layer or paper chromatography is not sufficient for the positive identification of a component. Whilst the measurement of several retention times or RF values with different operating conditions may show that the component in question is not present, such evidence cannot be used as proof of the presence of the component. Improvements in specificity can be obtained by the combined use of different clean-up procedures.The use of liquid - liquid partition and column chromatography or chemical treatment prior to GLC improves the specificity, and the additional use of thin-layer chromato- graphy prior to GLC increases the specificity even further. Increased certainty of identification can better be obtained by analysis using two or more methods. This has been done successfully in the past and extracts have been analysed by colorimetry and bioassay, by total chlorine and bioassay and by GTX. Such combined procedures cannot always be applied, however, especially for the analysis of low residues as few of the other methods are as sensitive as GLC for the analysis of chlorinated pesticides. If infrared analysis or mass spectrometry cannot be used, reasonable specificity can be achieved if the extract is subjected to partition, column chromatography and thin-layer chromatography in turn, and is analysed by gas - liquid chromatography and by another procedure such as colorimetry or bioassay.Beroza and Bowman6’ 768 have used a combination of liquid partition and GLC for the identification of pesticides by using the GLC retention time and the partition coefficient as parameters to identify a component, and this approach shows promise for the positive identification of pesticides. RECOMMENDED METHODS In the previous sections, work on the chlorinated pesticides has been reviewed in a general way. In this last section, procedures will be recommended for the extraction, clean-up and analysis which at the present time seem to be the best.Procedures that can be applied to whole groups of compounds will be outlined first. Then each chemical will be considered164 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [Analyst, Vol. 91 individually, pointing out any particular difficulties in the analysis, referring to published work on each chemical to illustrate the use of the recommended methods on a variety of substrates, and giving alternative methods. GENERAL PROCEDURES EXTRACTION- soizs- Mix a suitable weight of wet soil with anhydrous sodium sulphate and tumble them end- over-end with a solvent mixture. A ratio of 1 g of soil to about 0.5 g of sodium sulphate and 2 ml of a mixture of 20 per cent. acetone in hexane is recommended, tumbling for 1 hour.Filter the mixture and remove the water-soluble portion by water washing. crops- Macerate a convenient weight with anhydrous sodium sulphate and a mixture of water- immiscible and miscible solvents, benzene or hexane and acetone or isopropanol. A ratio of 1 g of crop to about 0.5 g of sodium sulphate and 2 ml of a 1 to 1 volume solvent mixture, with blending for 1 minute is usually sufficient. Filter and remove the water-soluble portion by water washing. Animal tissues and products- a suitable solvent such as hexane. Mills% already described may be used. CLEAN-u P- For the removal of polar interference use column chromatography with adsorbents such as alumina, Florisil or magnesia. Elute the non-polar compounds such as aldrin, DDE, DDT and heptachlor with hexane, and the more polar ones such as dieldrin, endrin and the oxidation products of chlorbenside with hexane containing a small amount of acetone or ether.For non-polar interference, such as lipid, use liquid - liquid partition with hexane - acetonitrile,62 hexane - dimethylformamidegO v63 or hexane - dimethylsulphoxide64 as solvent pairs. To determine low residue levels on materials with a high fat content, further clean-up by using column chromatography is usually necessary. It should be stressed that while these are the general methods of clean-up, others referred to in a previous section may prove helpful for particular problems. Grind the tissue with anhydrous sodium sulphate or sand and extract by warming with Alternatively, the methods of Langlois et al.36 or of ANALYSIS- Gas - liquid chromatogyaphy- The operating conditions given in the papers of Goodwin et al.,65 Burke and Holswade102 and Burke and Giuffrida1l8 should be consulted.The following points will serve as guides in the choice of conditions for particular systems. A non-polar silicone stationary phase has the best separating power; other thermally-stable polar phases may be needed to give particular separations or additional evidence of identification. The following are recom- mended to minimise the decomposition of pesticides ; on-column injection, suppression of support activity by treatment with Epikote resin or hexamethyldisilazane, glass or other non-reactive materials for the column and the use of column temperatures as low as con- veniently possible, General screening methods with GLC, designed to reveal the presence of chlorinated pesticides, have been reported by Goodwin et aZ.,95 Kaufman and Jackson,246 McCully and McKinley,38 Minyard and Jackson135 and Wells.=' T hin-lay er chromatography- Where GLC equipment is not available TLC offers an inexpensive and suitable alternative with good separating power and fair sensitivity.Like paper chromatography, it is a useful complementary technique for identification purposes. The best separations have been achieved with silica gel and a non-polar eluting solvent such as hexane or heptane. Alumina gives poorer separations, but if silver nitrate is used as detectinp apent it gives better sensitivitv. A 300-w layer of adsorbent is preferred sinceMarch, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 165 The best sensitivities have been obtained with Mitchell’s silver nitrate - phenoxyethanol spray reagent231 (i) AZdrin- INDIVIDUAL PROCEDURES c1 c1 Materials treated directly or indirectly with aldrin must be examined for both aldrin and dieldrin (its oxidation product).Both products are stable to the GLC operating conditions described above. On a silicone stationary phase, aldrin is difficult to resolve from dicofol and o$’-TDE olefin but these are separable on more polar phases. Residues to a level of 0.01 p.p.m. can be determined readily in crops and tissues. The GLC method has been used for the determination of aldrin in soil and root crops by Lichtenstein and his c o - w ~ r k e r s , ~ ~ ~ ~ ~ ~ ~ by Stewart et aZ.250 and by Decker et in grapes by Hascoet and Adam252; in vegetables and dairy products by Watts and Klein1S6; in milk by Henderson2M; and in water by Hindin et aZ.24 Alternative methods to GLC are the phenylazide colorimetric method and bioassay, both of which are discussed in detail by Porter.254 (ii) y-BHC (Linda?ze)- c1 co c1 C1 Cl yBHC may be measured at the residue level by gas - liquid ~ h r o m a t o g r a p h y l ~ ~ ~ ~ ~ ~ 9 and this is the recommended method of analysis. Electron-capture detectors have a large response to this compound and residues of 0.01 p.p.m.can be determined readily and the limit of detectability is even lower. GLC has been applied to the analysis of y-BHC in a wide range of crops,ll 946 9659255 dairy I t is rarely necessary to analyse samples for residues of the other isomers of BHC but the isomers can be conveniently separated by GLC,102J18,257 TLC229 or paper chromato- g r a ~ h y .~ ~ 8 9259 The Schechter - Hornstein colorimetric procedure or modifications of it have been used by many workers for the analysis of a wide range of samples.260t0268 The collaborative studies of the Schechter - Hornstein method are discussed by Klein269 but the procedure is much less sensitive than GLC. Other methods of analysis that have been used include bioassay,l77 p ~ l a r o g r a p h y ~ ~ ~ and infrared spectroscopy.271 9272 and tissues.255 (iii) ChZorbenside (Mitox)- Chlorbenside can be analysed with the GLC conditions described previously and is resolved from the other pesticides considered here.GLC does not seem to have been reported for residue analysis of chlorbenside but the previous work1029118 has shown this to be feasible. GLC of the metabolites of chlorbenside, the sulphoxide and sulphone does not seem to have been reported but should be possible. Gunther, Blinn and Barnes273 have measured residues of chlorbenside in pears by infrared spectroscopy. The method is also suitable for analysing for the sulphoxide oxidation product166 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [Analyst, 1'01. 91 of chlorbenside. The sample is extracted with hexane and is subjected to column chromato- graphy which separates the chlorbenside and its sulphoxide. The separated sulphoxide and chlorbenside are determined by infrared spectroscopy.The alternative colorimetric procedure of Higgons and K i l b e ~ ~ ~ ~ involves oxidation of the chlorbenside to the sulphone, followed bjr nitration and spectroscopic measurement at 575 mp. Higgons and Kilbey did not separate the chlorbenside and its metabolites before analysis, but there is no reason why the separation procedure by Gunther could not be used for this. This separation procedure may also be of use before GLC analysis. (iv) Chlordaue- c1 c1 Chlordane is produced by the chlorination of chlordene and contains 60 to 75 per cent. of the p- and y-chlordanes (the trans and cis isomers, respectively), and 25 to 40 per cent. of analogous materials containing 6, 7 and 9 chlorine atoms. The gas chromatogram of technical material has a characteristic appearance, the peaks due to p- and y-chlordane predominating.N'hile in some respects the gas chromatography of chlordane residues is made easier by the readily identifiable pattern, the feasibility of distinguishing peaks from natural materials and from other pesticides that may be present, such as clilorfenson, chlorbenside and the major isomer of endosulfan, is not an easy matter, and must be overcome by prior clean-up or by the use of more than one stationary phase. Chlordane has rather specialised uses and little work has been published on residues arising from them, but reference may be made to the paper of Gutenmann and Lisk255 where a sensitivity of 0.01 p.p,m. of chlordane in soil was achieved. The alternative method is a colorimetric one, based on its reaction with methanolic potassium hydroxide and diethanolamine, details of which are gixren by The acaricide chlorfenson can be analysed satisfactorily by gas - liquid chromatography under the conditions previously described, but it is not readily resolved from endosulphan and p-chlordane.Such resolution might be accomplished by using other, perhaps more polar, stationary phases, but the GLC of chlorfenson has not been studied in detail for residue analysis. Butzler et aZ.91 have reported a colorimetric procedure that is sensitive to less than 5 pg of chlorfenson in orange pulp. In this procedure the chlorfenson is extracted with benzene, and the extract is hydrolyscd with alcoholic potassium hydroxide to convert the chlorfenson to $-chlorophenol which is separated by steam distillation and nitrosated.The product is then separated by- column chromatography before spectroscopic analysis at 430 mp. The analysis for residues of chlorfenson does not seem to have been reported very frequently in recent years. (vi) ChZorobenzilate- GLC promises to be a sensitive and convenient method for such analysis. Other colorimetric procedures have been described.g* 3276 9 2 7 7 CO,C,H, I OH The acaricide chlorobenzilate may be analysed with good sensitivity by gas - liquid chromatography and is stable under the conditions described. I t is readily resolved from the other acaricides considered here. Its retention time, however, is close to that of 09'-DDT and $$'-TDE when a column temperature of 200" C is used.l18March, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 167 The analysis of residues of chlorobenzilate in grapes and cotton seed with a sensitivity of 0.05 p.p.m.has been carried out successfully by microcoulometric GLC278 with 20 per cent. Dow 11 silicone grease on Chromosorb P. Benzene was used as the extraction solvent and clean-up was by partition into nitromethane followed by column chromatography on Florisil for cotton-seed extracts. Infrared spectroscopy was used to confirm the identity of the residue. The use of electron-capture has not yet been reported for the determination of chloro- benzilate residues but the work of Burke and Guiffridalls shows that this should be possible with high sensitivity. A suitable alternative method of analysis is the Delley colorimetric procedure as modified by H a r r i ~ ~ 7 ~ and described in detail by Margot and Stammbach.280 In this procedure the chlorobenzilate is extracted with benzene, saponified with methanolic alkali to form dichloro- benzilic acid.which is extracted and nitrated and the absorption is measured at 538mu. Clean-up with Nuchar carbon was used for grapes. This procedure is sensitive to 2 pg of chlorobenzilate and haibeen applied to of samples.279 Blinn et a1.28l have described a similar procedure but without step, and have applied this to analysis of residues in citrus. a wide range the nitration (vii) DDT, TDE, DDE- DDT and the The technical cc1, pp’-DDT The analvsis of residues of the maior and minor constituents of technical metabolic priducts is one requiring carehl experimentation and interpretation. material contains upwards of 70 per cent.of the $p’-isomer, the bulk of the remainder being the o$’-isomer, but some f$’-TDE and a small amount of op’-TDE are usually present. The principal metabolic product of DDT is DDA, which can be analysed by GLC after methylation. Other metabolic products are the olefin DDE, but TDE may also be produced and the olefin of TDE. In addition DDT may undergo thermal and catalytic conversion to DDE or TDE. Thus an extract of crop or tissue could conceivably yield DDT and TDE as such, or as dehydrochlorinated products, each as 99’- or o$’-isomers; and to add to the difficulty, derivatives may arise as impurities in the technical material, from metabolism or from decomposition. Exceptional care must therefore be taken when interpreting chromatograms.On a silicone stationary phase three pairs of the DDT analogues are difficult to resolve and require other stationary phases to show adequate separation. These pairs are $$‘-TDE olefin and of’-DDE; fp’-DDE and 09’-TDE; and $9’-TDE and op’-DDT. The latter two pairs may also be separated by conversion to their olefins by treatment with alkali, followed by GLC.141 ~ a t e r , 2 ~ W crops,136 ~lothing,~8~ dairy products,152$2~,256 tissues,3g 9285,286 92879288 and the total diet289 with a sensitivity sometimes of less than 0.01 p.p.m. Colorimetric methods of analysis have been reviewed by M i s k ~ s . ~ ~ ~ GLC has been used for the analysis of residues of DDT and its derivatives in (viii) Dico fol (Keltha%e)- OH c1 --c1 cc1, Dicofol is reported212 to decompose almost quantitatively during GLC and although Burke and Johnsonlol and Burke and Holswadelo2 have reported retention results for dicofol they have not shown that the response that they obtained was from dicofol itself.However, Gunther et aZ.129 have shown that, when the GLC column (silicone on firebrick) was “condi- ditioned” to dicofol the extent of decomposition was small, and that 90 to 95 per cent. of the dicofol was not decomposed during the chromatography, and the remainder was converted to 4,4’-dichlorobenzophenone. Thus, GLC shows considerable promise for the analysis for residues of dicofol, and some results have been given in a study of total-diet samples.28s168 BEYNON AXD ELGAR: ANALYSIS FOR RESIDUES OF [ArtdySt, Vol.91 Colorimetric analysis has been carried out successfully for dicofol in a wide range of crops and animal products. The basic procedure is that used by Rosenthal et aZ.2g1 involving hydrolysis of dicofol to chloroform which is determined colorimetrically by the Fujiwara reaction. Gordon, Haines and have carried out the hydrolysis with sodium hydroxide (for crops) or with tetraethylammonium hydroxide (for fatty samples) and their procedure has been described in detail by Gordon and S c h ~ c k e r t . ~ ~ ~ This procedure will detect 10 pg of dicofol and it has been applied to the analysis of dicofol in milk,292 butter-fat and animal body fat,294 and is applicable to a range of crops and animal products. Similar procedures have been used for a range of samples by other workers,295~296,297 and Gunther and Blinn298 have described an alternative procedure involving hydrolysis of dicofol to 4,4’-dichlorobenzo- phenone which is measured by ultraviolet spectroscopy.(ix) Dieldrin- Dieldrin is stable to the operating conditions described previously for GLC analysis. On non-polar stationary phases the retention times of dieldrin and +$’-DDE are almost coincident, and op’-TDE is also close. It is essential therefore to acquire complementary evidence for the identity of die1dri11.l~~ It is fairly stable to metabolic change, though several workers have reported metabolites.2gg 9300 3301 9302 Recent examples of the analysis of dieldrin residues by GLC are those carried out by Hardee et aZ.,303 by Morley and Chiba304 and Decker et aZ.251 on soil, by Harvey and Harvey on pasture,305 by Williams, Mills and hlcDowel1 on milk,256 by Hunter et aZ.286 and Dale and ~Quinby~~7 on human fat and by on total-diet samples.Residue levels of 0.01 p.p.m. can be attained without excessive clean-up. The alternative specific method for dieldrin is the phenylazide colorimetric procedure discussed in detail by Porter.306 (x) Endoszdfan (Thiodan)- c1 C1 Endosulfan has two stereoisomers, the lower-melting isomer constituting two-thirds of the technical product, the higher-melting isomer being the other major component. These can decompose to endosulfan alcohol, and the so-called endosulfan ether may also be present in technical material. The alcohol may also be produced by metabolism, and the presence of endosulfan sulphate from metabolic oxidation has also been re~orted,~O73~~8 so that once again the application of one technical material can give rise to a complex residue analysis.Zweig et u Z , ~ ~ have used GLC as a means of separation in their work on the infrared determination and identification of endosulfan residues. Details of the GLC microcoulometric procedure are given by Graham et aZ.309 but few results specifically relating to residues of endosulfan have been reported.ggJ05,307 p 3 l o The major component of endosulfan may be difficult to resolve under normal GIX operating conditions from p- and y-chlordane isomers and from chlorfenson, and the other isomer from endrin or chlorobenzilate, but the presence and the ratio of the several peaks from technical endosulfan gives some evidence of identity.March, 19661 CHLORINATED IXSECTICIDES AND ACARICIDES (xi) Endrin- c1 169 c1 Although endrin is sensitive to temperature,llg it can be analysed without decomposition by GLC.65 Endrin has a less toxic product, the so-called “delta keto,” an isomeric ketone that occurs in plants but not in animals.Endrin is resolved from other common pesticides on a silicone stationary phase. The pattern of peaks due to decomposition on chromato- graphy at high temperature could be used as an aid to identification. GLC analyses of endrin residues in alfalfa,255 water,24 milk,256 tissue39 and the total diet2sg have been reported in recent years. Colorimetric procedures for the analysis of residues of endrin have been reviewed by Terriere.311 (xii) Heptnchlor and heptnchlor epoxide- c1 Like aldrin, heptachlor forms an epoxide in biological media, so that both must be deter- mined as residues following heptachlor applications. Both are thermally stable and can be chromatographed on the sub-microgram scale without trouble. Rusk and Fahey312 have pointed out that y-chlordane present in technical heptachlor can form a significant proportion of the total residue, and its retention time on non-polar stationary phases is similar to that of heptachlor epoxide. They have given details of a chromatographic separation of heptachlor, its epoxide and y-chlordane on Florisil. Recent residue work carried out by GLC include that in s0il,2489250 crops,248 9250 milk,256 9313 tissue66 and the total diet.289 Colorimetric methods have been reviewed by Bowery.314 (xiii) Methoxychlor- CCI, H Methoxychlor may be analysed by gas - liquid chromatography under the standard conditions described previously.Because of its long retention time there is little interference from other chlorinated pesticides. The method has not been widely reported for residue analysis but it has been used by BaeW who also described a suitable clean-up procedure, and by Williams.289 A colorimetric method of analysis was developed by Fairing and W a r r i n g t ~ n ~ ~ and is described in detail by Lowen et aZ.315 The method, which is applicable to crops and animal products, involves partitioning of the pesticide into nitromethane followed by dehydro- chlorination, column chromatography and sulphonation, to give a coloured product that is measured spectroscopically.The method is sensitive to 2 pg of methoxychlor. This procedure was used by Cluett et nl.316 in combination with total halide analysis to determine methoxy- chlor and its possible metabolites in milk. Other methods of analysis have been summarised by Lowen et aL315170 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [AfldySf, Vol. 91 (xiv) Oxythane (Neoiran)- C l e OCH,O Oxythane was not included in the work of Burke and Holswadelo2 or of Burke and JohnsonlOl but it is very likely that the analysis for this compound can be carried out by GLC. Little work has been published on the analysis for residues of oxythane or on its thin-layer and paper chromatography . Gunther and Blinn3I7 have described a colorimetric procedure involving extraction with benzene followed by hydrolysis with hydrobromic acid to form 9-chlorophenol.The phenol is separated from interfering compounds by steam distillation, and is reacted with 4-amino- antipyrine, before spectroscopic measurement at 510 mp. (xv) Tetradi f o n (Tedion)- c1 / Ci Gas - liquid chromatography is suitable for the analysis of residues of tetradifon, and microcoulometric detectors have been used by Coulson et aZ.,96 Cassilg9 and by Burke and Mills105 for the analysis of a range of crops, and less than 5 pg of tetradifon could be detected. The use of electron-capture detection would improve the sensitivity considerably. The GLC of tetradifon has been carried out successfully by other workers who have used both micro- coulometric and electron-capture detection.102s118 A colorimetric procedure based on the Fujiwara reaction was developed by Fullmer and CassiPls and is described in detail by Cassil and Yaffe319 who have also considered many of the possible sources of interference in the procedure. Gunther and Blinns9 have described a total chlorine method, and Gunther et have measured tetradifon residues by infrared analysis with chromic oxide - acetic acid for the clean-up. (xvi) Toxaphene- As toxaphene is a multi-component mixture of chlorinated terpenes, it is not possible to recommend a specific method. Residue analysis has generally been carried out by a totd- chlorine procedure or the zinc chloride - diphenylamine colorimetric method of Graupner and D ~ n n , ~ ~ l both of which have been described by D ~ n n .~ ~ ~ Nikolov and done^^^^ have recently described a modification to the colorimetric procedure that increased the sensitivity 10-fold. The GLC method with halogen-selective detection is an advance on these in that sensi- tivity is improved and some semblance of specificity is offered due to the pattern of peaks produced. However, at low residue levels this specificity is unreal as the presence of peaks from natural products or other pesticides would be impossible to distinguish from the Toxaphene components. After rigorous clean-up, determinations of fairly low residue concentrations are possible on short columns which, having poor efficiency, give a single or only a few peaks for the Toxaphene mixture.324 An attempt to analyse total diets for toxaphene residues has been reported.289 We would like to thank Mr.E. S. Goodwin, Mr. R. Goulden, Mr. D. S. Penny, Mr. A. Richardson and Mr. P. C. R. Webb for reading the manuscript and for their helpful advice. REFERENCES 1. Zweig, G., Editor, “Analytical Methods for Pesticides, Plant Growth Regulators, and Food 2. Huddleston, E. W., Thompson, K. H., Gyrisco, G. G., Lisk, D. J., Kerr, T. W., jun., and Olney, 3. Van Middelem, C. H., Wilson, J. W., and Hanson, W. D., Ibid., 1956, 49, 612. 4. Poos, F. W., Dobbins, T. N., and Carter, R. H., U.S. Department of Agriculture, Bulletin E-793, 6. Lykken, L., Mitchell, L. E., and Woogerd, S. M., J . Agric. Fd Chem., 1967, 5, 501. 6. Lykken, L., in Gunther, F. A., Editor, “Residue Reviews,’’ Springer-Verlag, Berlin, Volume 3, 7.Lichtenstein, E. P., Mueller, C. H., Myrdd, G. R., and Schulz, K. R., J . Econ. Ent., 1962, 55, 216. 8. Garber, M. J., in Zweig, G., 09. cit., Volume 1, p. 491. Additives,” Academic Press, New York and London, Volumes 1 to 4, 1963-64. C. E., J . Econ. Ent., 1960, 53, 1078. 1950. 1963, p. 19.9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. March, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 171 Westlake, W. E., Analyt. Chem., 1959, 31, 724. Bann, J. M., Paper presented at the 131st American Chemicd Society Meeting, Miami, April, 1957.Goulden, R., Qualitas PI. Muter. Veg., 1964, 11, 381. Koblitsky, L., and Chisholm, R. D., J . A s s . 08. Agric. Chem., 1949, 32, 781. Lichtenstein, E. P., and Schulz, K. R., J . Econ. Ent., 1960, 53, 192. Taschenberg, E. F., Mack, G. L., and Gambrell, F. L., J . Agric. Fd Chem., 1961, 9, 207. Shell Method Series, SMS-551/55. Heinisch, E., NachrBl. dtsch. PJSchutzdienst, Berlin, 1959, 13, 161. Thornburg, W. W., in Zweig, G., op. cit., Volume 1, p. 87. Van Middelem, C. H., in Zweig, G., 016. cit., Volume 1, p. 25. Klein, A. K., J . Ass. Off. Agric. Chem., 1958, 41, 551. -, Ibid., 1960, 43, 703. Klein, A. K., Laug, E. P., and Sheehan, J. D., jun., Ibid., 1959, 42, 539. Hardin, L. J., and Sarten, C. T., Ibid., 1962, 45, 988.Mills, P. A., Onley, J . H., and Gaither, €3. A., Ibid., 1963, 46, 186. Hindin, E., May, D. S., and Dunstan, G. H., in Gunther, F. A., op. cit., Volume 7, 1964, p. 130. Rosen, A. A., and Middleton, F. M., Analyt. Chem., 1959, 31, 1729. Teasley, J. I., and Cox, W. S., J . Amer. Wat. Wks ,4ss., 1963, 55, 1093. Kahn, L., and Wayman, C. H., Analyt. Chem., 1964, 36, 1340. Eidelman, M., J. Ass. Off. Agric. Chem., 1963, 46, 182. -, Ibid., 1962, 45, 672. Lampert, L. M., J . Dairy Sci., 1964, 47, 1013. O’Donnell, A. E., Neal, M. M., Weiss, F. T., Bann, J . M., DeCino, T. J., and Lau, S. C . , J.Agric. O’Donnell, A. E., Johnson, H. W., jun., and Weiss, F. T., Ibid., 1955, 3, 757. Prickett, C. S., Kunze, F. M., and Laug, E. P., J . Ass. 08. Agric. Chem., 1950, 33, 880.illills, P. A., Ibid., 1961, 44, 171. Onley, J. H., Ibid., 1964, 47, 317. Langlois, B. E., Stemp, A. R., and Liska, B. J., J . Agric. Fd Chem., 1964, 12, 243. RiIills, I?. A., J . A s s . Off. Agric. Chem., 1959, 42, 734. McCully, K. A., and McKinley, W. P., Ibid., 1964, 47, 652. Taylor, A,, Rea, R. E., and Kirby, D. R., Analyst, 1964, 89, 497. de Faubert Maunder, M. J., Egan, H., Godly, E. W., Hammond, E. W., Roburn, J., and Thomson, Krzcminski, L. F., and Landmann, W. A., J . Agric. Fd Chem., 1963, 11, 81. Moddes, R., J. Ass. Ojf. Agric. Chem., 1961, 44, 169. Moats, W. A., Ibid., 1962, 45, 355. Coulson, D. M., and Barnes, E. G., Stanford Hes. Inst., Pesticide Res. Bull., 1963, 3, No. 2. Cassil, C. C., Gortner, W., and Stoner, H., quoted in Zweig, G., op.cit., Volume 1, 1962, p. 97. Baetz, R. A., J . Ass. Off. Agric. Chem., 1964, 47, 322. Moats, W. A., Ibid., 1964, 47, 587. Coulson, D. M., Stanford Res. Inst., Pesticide Res. Bull., 1962, 2, No. 4. Wheatley, G. A., Hardman, J. A., and Strickland, A. H., PI. Path., 1962, 11, 81. hbdallah, M. D., and Landheer, C. A., J . Chromat., 1962, 9, 245. Hoskins, W. M., Erwin, W. R., Miskus, R., Thornburg, W. W., and Werum, L. N., J . Agric. Coulson, D. M., Cavanagh, L. A., and Stuart, J., Ibid., 1959, 7, 250. Zweig, G., Archer, T. E., and Rubenstein, D., Ibid., 1960, 8, 403. Crosby, N. T., and Laws, E. Q., Analyst, 1964, 89, 319. Calderbank, A., Morgan, C. B., and Yuen, S. H., Ibid., 1961, 86, 569. Calderbank, A., and Yuen, S. H., Ibid., 1965, 90, 99. Plapp, F. ?IT., and Casida, J.E., Analyt. Chem., 1958, 30, 1962. Cueto, C., Barnes, A. G., and Mattson, A. M., J . Agric. Fd Chem., 1956, 4, 943. Heinisch, E., and Neubert, P., J . prakt. Chem., 1963, 22, 267. Abbott, D. C., and Thomson, J., Chem. & Ind., 1964, 481. Taylor, A., and Fishwick, B., Lab. Pract., 1964, 13, 525. Jones, L. R., and Riddick, J. A., Analyt. Chern., 1952, 24, 569. Burchfield, H. P., and Storrs, E. E., Contr. Boyce l’homson Inst. PI. Res., 1953, 17, 333. Haenni, E. O., Howard, J. W., and Joe, It-. L., jun., J . Ass. 08. Agric. Chem., 1962, 45, 67. Goodwin, E. S., Goulden, R., and Reynolds, J. G., Analyst, 1961, 86, 697. Taylor, A., Ibid., 1962, 87, 824. Beroza, M., and Bowman, M. C., Analyt. Chem., 1965, 37, 291. Fairing, J. D., and Warrington, €1. B., Adv.Chem. Ser., 1950, 1, 260. Williams, D. W., J . Ass. 08. Agric. Chem., 1957, 40, 254. Anglin, C., and McKinley, W. P., J . Agric. Fd Chenz., 1960, 8, 186. McKinley, W. P., Savary, G., and Webster, C., Ibid., 1962, 10, 226. McKinley, W. P., and Savary, G., Ibid., 1962, 10, 229. McCully, K. A., and McKinley, W. P., J . Ass. Off. Agric. Chem., 1964, 47, 859. McKinley, TV. P., Coffin, D. E., and McCully, K. A., Ibid., 1964, 47, 863. Gunther, I;. A., and Blinn, R. C,., J , Agric. Fd Chem., 1953, 1, 325. Cueto, C. J., jun,, Ibid., 1960, 8, 273. Fd Chem., 1954, 2, 573. J., Ibid., 1964, 89, 168. Fd Chem., 1958, 6, 914. I , J . Ass. Ofl. Agric. Chem., 1965, 48, 358. -~172 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 9 9 . 100.101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 123. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. BEYNON AND ELGAK: ANALYSIS FOR RESIDUES OF j’ilqzalyst, Vol. 91 Bann, J . M., Lau, S. C., Potter, J . C., Johnson, H. W., O’Donnell, A. E., and Weiss, F. T., Ibid., 1958. 6, 196. Carter, R. H., I n d . Engng Chem., Analyt. Ed., 1947, 19, 54. Frawley, J. P., and Davidow, B., J . A s s . Off. Agric. Claem., 1949, 32, i58. Tufts, L. E., Darling, G. W., and Kimball, R. H., Ibid., 1950, 33, 976. Meyer, C. F., Malina, M. A., and Polen, P. B., J . Agric. Fa? Chem., 1960, 8, 183. Helrich, K., Hansens, E. J . , and Granett, P., Ibid., 1958, 6, 281. Albert, R. A., J . Ass. Off. Agric.Chem., 1964, 47, 659. Schechter, M. S., Pogorelskin, M. A., and Haller, H. L., I n d . Engng Chem., Analyt. Ed., 1947, Davidow, B., J . Ass. Off. Agric. Chem., 1950, 33, 130. Bradbury, P. R., and Standen, H., J . Sci. Fd Agric., 1954, 5, 252. Gunther, F. A., Blinn, R. C., and Ott, D. E., J . Ass. Off. Agric. Chenz., 1962, 45, 359. Gunther, F. A., and Blinn, K. C., “Analysis of Insecticides and Acaricides,” Interscience Pub- Gunther, F. A., and Jeppson, L. R., J . Econ. Ent., 1954, 47, 1027. Butzler, G. J., Luce, E. N., and Wing, R. E., J . Agric. Fd Chem., 1957, 5, 42. Gunther, F. A., and Blinn, R. C., op. cit., p. 333. Ott, D. E., and Gunther, F. A., J . Agric. Fd Chem., 1964, 12, 239. Clifford, P. A., J . A s s . Off. Agric. Chem., 1947, 30, 337. Goodwin, E.S., Goulden, R., Richardson, A., and Reynolds, J. G., Chem. & I d . , 1960, 1220. Coulson, D. M., Cavanagh, L. A,, Dc Vries, J . E., and Walther, B., J . Agric. Fd Chem., 1960, Clark, S. J., Paper presented a t the 140th American Chemical Society Meeting, Chicago, September, Adlard, E. R., and Whitham, B. T., in Desty, D. H., Editor, “Gas Chromatography 1958,” Butter- Cassil, C. C., in Gunther, I?. A., op. cit., I‘olume 1, 1962, p. 37. Challacombe, J . A., and McXulty, J . -I., iiz Gunther, F. A., op. cit., Volume 5, 1964, p. 5 i . Burke, j., and Johnson, L., J . Ass. OH. Agric. Chem., 1962, 45, 348. Burke, J., and Holswadc, \IT., Ibid., 1964, 47, 845. Coulson, D. M., and De Vries, J. E., Stanford Res. Inst., Tech. Rep. No. I I , Report No. 6, 1958. Coulson, D.M., De Vries, J . E., and Walther, B. A., Stanford Res. Inst., Tech. Rep. No. I V , Report Burke, J., and Mills, P. A., J . Ass. 08. .4gric. Chem., 1963, 46, 177. Beckman, H., and Bevenue, A., J . Agric. Fd Chem., 1964, 12, 245. Coulson, TI. M., Huene, A. S., and Cavanagh, L. A., Stanford Res. Inst., Tech. Rep. No. I , Report Coulson, D. M., Stanford Res. Inst., Tech. Rep. No. 11, Report No. 5 , 1961. Lovelock, J. E., and Lipsky, S. K., J . Amer. Chem. SOC., 1960, 82, 431. Lovelock, J . E., Analyt. Chem., 1961, 33, 162. -_ , Nature, 1961, 189, 729. Lovelock, J . E., Zlatkis, A., and Becker, R. S., Ibid., 1962, 193, 540. Clark, S. J., in Gunther, F. h., op. cit., Volume 5, 1964, p. 32. Moore, A. D., J . Econ. E n t . , 1962, 55, 271. Lovelock, J . E., Analyt.Chem., 1963, 35, 474. Landowne, R. A., and Lipsky, S. K., Ibid., 1962, 34, 726. Dimick, K. P., and Hartmann, H., in Gunther, F. A., op. cit., I’olume 4, 1963, p. 150. Burke, J., and Giuffrida, L., J . A s s . Off. Agric. Chem., 1964, 47, 326. Phillips, D. D., Pollard, G. E., and Soloway, S. B., J . Agric. Fd Chem.,. 1962, 10, 217. Cremer, E., Kraus, T., and Bechtold, E., Chem.-Ing.-Tech., 1961, 33, 632. Gonlden, K., Goodwin, E. S . , and Davies, I,., Analyst, 1963, 88, 951. Monlrman, J . T,., and l>ubois, L., in Socbels, H. J . , JITall, R. F., and Urenncr, S., Editors, “Gas Gunther, F. A., Rlinn, R. C., and Ott, D. E., Annlyt. Chew., 1962, 34, 302. Sternbcrg, J . C., Paper presented at the 13th Pittsbiirgh Conference on Analytical Chemistry and Johns, T., Paper presented at the 14th Pittsburgh Confvrence on Analytical Chemistry and Applied Johns, T., and Braithwaite, C.H., jun., in Gunther, F. A,, op. rit., Volume 5, 1964, p. 45. Coulson, D. &I., J . Gas Chromat., 1965, 3, 134. Beckman, H., and Bcvcnue, 4., J . Clzro~nat., 1963, 10, 231. Gunther, F. 4., Barklcy, J . H., Blinn, K. C., and Ott, D. E., Stanford Res. Inst., Pesticide Res. de Faubert Maundcr, M. J., Egan, H., and Roburn, J . , AnaZyst, 1964, 89, 157. Mcnn, J . J., Patchett, G. G., and Batchelder, G. H., in Zweig, G., op. cit., Volume 2, p. 545. Goulden, R., Goodwin, E. S., and Davies, L., Analyst, 1963, 88, 941. Gunther, F. A., Blinn, R. C., and Kohn, G. K., Nature, 1962, 193, 573. Bohcmen, J . , Langer, S. H., Perrett, R. H., and Purnell, J . H., J .Clzein. SOL., 1960, 2444. Minyard, J . P., and Jackson, E. R., J . Ass. OH. Agrir. Chem., 1963, 46, 843. Watts, J . O., and Iilein, A. I<., Ibid., 1962, 45, 102. Beckman, H. F., and Bevenue, A . , J . Chromat., 1963, 12, 109. Bevenue. A.. in Zweig. G.. o B . cit.. Volume 1. D. 210. 19, 51. lishers Inc., New York and London, 1955, p. 248. 8, 399. 1961. worths, London, 1958, p. 361. No. 12, 1960. No. 3, 1960. Chromatography,” iicadcmic Press Tnc., Sew York, 1961, p. 333. Applicd Spectroscopy-, 1962. Spectroscopy, lB63. Bull., 1962, 2, No. 2.March, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 173 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 140. 150. 151. 152. 153. 154. 153. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170.171. 172. 173. 174. 176. 176. 177. 178. 179. 180. 181. 182. 183. 184. 186. 186. 187. 188. 189. 190. 191. 192. 193. 194. 195. 196. 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. Robinson, J . , and Richardson, A., Chenz. & I f i d . , 1063, 1460. Bonelli, E. J., Hartmann, H., and Dimick, I<. P., J . Agric. F d Chem., 1964, 12, 333, Klein, 11. I<., and IYatts, J . O., J . A s s . Off. A g r i c . Cheni., 1964, 47, 311. Segal, H. S., and Sutherland, M. L., in Gunther, F. *4., op. c i f . , Volume 5, 1964, p. 73. Ebing, ITr., and Henkel, H. G., J . Gas Chronzat., 1964, 2, 207. Henkel, H. G., and Ebing, If-., Ibid., 1964, 2, 215. Hannon, C. I., Angclini, J., and ivalford, R., I b i d . , 1963, 1, 7, 27. Benfield, C. A., and Chilwull, E. D., rl?iuZyst, 1964, 89, 475.Burke, J., J. A s s . Ojf. Agrir. Chew., 1963, 46, 198. Taylor, M. P., J . Chronzat., 1963, 9, 28. Shoemakc, G. R., Lovelock, J . E., and Zlatkis, A., Ibid., 1963, 12, 314. Scott, K. P. IT:., and Grant, L). M", ,d~iuIysf, 1964, 89, 170. Franc, J., arid Pulichajlova, S., J . Chvoniut., 1963, 12, 22. Klein, A. K., Watts, J. O., and namico, J . X., J . A s s . Ofl. Agric. Chenz., 1963, 46, 165. Beckman, H. F., and Rerkcnkotter, P., Awalyt. Chem., 1963, 35, 242. Kalls, J . \Y., Ibid., 1960, 32, 332. Beroza, M., and Acree, F., jiin., J . z4s.s. Ofl. .4gric. Chew., 1964, 47, 1. Beroza, &I., and Sarmiento, K., ,4iiaEyt. Chewz., 1!)64, 36, 1744. RIinyard, J . P., and Jackson, E. R., J . i l g r i c . Fd Chew., 1965, 13, 50. Franc, J . , and Iiolouskova, \'., J.Chrotmt., 1965, 17, 221. Grou, E., and Balif, G., Ilevln'. Chinz., 1962, 13, 682. Schmitt, K. A., in Zweig, G., op. cit., I'olume 1, p. 281. Guinn, V. P., and Schmitt, R. -I., in Gunther, F. A,, op. cit., Volume 5, 196.1, p. 148. Gajan, K. J . , i n Gunther, P. X., op. rzt., \-oliimc 5, 1964, p. 80; Volumc 6, 1964, p. 75. Brazcc, R. D., Trans. Anter. SOC. Agrir. E ~ g r s , 1963, 6, 163. Gajan, R. J., and Link, J . , J . - 4 s ~ . Off. Agric. Chew., 1964, 47, 1119. ., Iloczn. Claewz., 1952, 26, 281. ., and Krzenminska, A . , Cheinia AnaZit., 1960, 5, 611. Fehcr, F., and Monien, H., %. AwuZyt. Chent., 1964, 204, 19. Davidek, J . , and Janicek, G., Ibid., 1963, 194, 431. Fukami, H., Kimura, H., and Nakazima, JI., Boty~i-Kujukz4, 1953, 18, 51; Chenz. Abstr., 1953, Iieller, H., Hochweber, M., and Halban, H., Helv.Clzinz. A c t u , 1946, 29, 512. Davidek, J., and Janicck, G., Experientia, 1961, 17, 473. Swineyoel, 0. A., J. S. Afr. Chenz. Iizst., 1962, 15, 88. liosmatyi, E. S., Klzina. Sredstva. Zaslz. Rust., 1963, 134; Cheni. Abstr., 1965, 62, 4518. Kosmatyi, E. S., and Shlyapak, S. I., Vop. Pitan., 1963, 22, 83; Chem. Abstr., 1964, 60, 1035. Needham, f'. H., Anulyst, 1960, 85, 792. Dewey, J. E., J . A g r i c . F d Cheun., 1958, 6, 274. Sun, Y. P., in Zweig, G., op. cit., Volume 1, p. 399. Phillips, 1%'. F., in Zweig, G., op. cit., Volume 1, p. 471. Bass, RI. H., and Make, G. H., J . Econ. E?zt., 1964, 57, 665. McDonald, S., Caw. J. Zool., 1962, 40, 719. Davidow, B., and Sabatino, F. J,, J . Ass. 08. Agric. Cliewt., 1954, 37, 902.Dewey, J . E., and Parker, B. L., J. ECOX Etat., 1964, 57, 821. Gerolt, P., Bull. Ent. Res., 1957, 48, 311. Phillips, IV. F., Bowman, M. C., and Schultheisz, K. J., J . Agric. F d Chenz., 1962, 10, 487. Earle, N. W., Pankaskie, J . E., and Sun, Y. P., J . A s s . Uff. A g r i c . ChewL., 1959, 42, 586. Mulla, M. S., Axelrod, H., and Isaak, L. 1V., ;Ilosqiiito News, 1961, 21, 21.6. Henncbcrry, T. J . , hIcGovern, \V. L., Yeomans, A. € I . , and Mason, H. C., J. Econ. E+it., 1964, Sun, Y. P., and Sun, J . Y., Ibid., 1952, 45, 26. Laug, E. P., J . Pharmac. Exp. Ther., 1948, 93, 277. Edwards, C. A., Beck, S. D., arid Lichtenstein, E, P., J . Ecoii. E n t . , 1957, 50, 622. Sun, Y. l'., and Sanjean, J., I b i d . , 1961, 54, 841. Harwood, K. I?., and Xrcckul, S., Science, 1060, 131, 1369.Bombosch, S., 11Idt. b i d . Ruiidiist. Ld-it. Forstw., Uerlitt, 1956, 85, 113. liiigemagi, U., and Terriere, L. C., J . Ecoii. E d . , 1963, 56, 343. Blinn, R. C., and Gunther, F. A , , i,iz Gunther, F. A . , op. c i f . , I.'olume 2, 1063, p. 99. Bruce, li. H., in Zweig, G., op. cat., Volume 1, p, 149. Frehse, H., Hofclzeizbr. Buyer PjlSchzttz-Nachr., 1963, 16, 182. Hlinn, R. C., Gunther, F. X., and Mulla, RI. S., J . Econ. Ent., 1960, 53, 1129. Johns, T., and Uraithwaite, C. H., zit Gunther, F. -4., op. cit., Volume 5, 1964, p. 45. lireuger, H. ti., and Volkmann, I-I., %. Phys., 1963, 173, 78. Chcn, J . Y. T., J . A s . Off. A g r i c . Chem., 1965, 48, 380. Sparagana, M., and Mason, IT;. B., Analyt. Chewa., 1962, 34, 242.Giuffrida, L., J. A s s . 08. A g r i c . Client., 1965, 48, 364. MacDougall, D., i+t Gunthcr, F. A,, op. cit., Volume 5, 1964, p. 119; and in Zweig, G., op. czt., Hornstein, I., J . Agric. F d Chew.., 1958, 6, 32. McKinley, W'. P., and llahon, J. H., J . A s s . Off. Agric. C'kem., 1959, 42, 725. Moynihan, P., and O'Colla, P., Chem. 6- I n d . , 1951, 407. 47, 9180. 57, 769. Volume 1, p. 167.174 BEYNON AND ELGAR: ANALYSIS FOR RESIDUES OF [A%a&St, VOl. 91 San Antonio, J. P., J . Ass. Off. Agric. Chem., 1960, 43, 721. Egan, H., and Evans, W. H., Paper presented at the 18th International Congress of Pure and Evans, W. H., Analyst, 1962, 87, 569. Block, R. J., Durrum, E. L., and Zweig, G., “A Manual of Paper Chromatography and Paper Electrophoresis,” Academic Press, New York, 1958.Zweig, G., Chromat. Rev., 1964, 6, 110. McKinley, W. P., in Zweig, G., op. cit., Volume 1, p. 227. Mitchell, L. C., J . Ass. Ofl. Agric. Chem., 1952, 35, 920. -, Ibid., 1953, 36, 1183. -, Ibid., 1954, 37, 530. -, Ibid., 1954, 37, 996. -, Ibid., 1956, 39, 484. -, Ibid., 1956, 39, 891. -, Ibid., 1957, 40, 294. -, Ibid., 1957, 40, 1034. --, Ibid., 1958, 41, 781. -, and Mills, P. A., Ibid., 1960, 43, 748. Mitchell, L. C., Ibid., 1962, 45, 682. ---, Ibid., 1963, 46, 988. Graham, S. O., Science, 1963, 139, 835. Krzeminski, L. F., and Landmann, W. A., J . Chromat.. 1963, 10, 515. Conkin, R. A., in Gunther, F. A., op. cit., Volume 6, 1964, p. 136. Ganshert, H., Waldi, D., and Stahl, E., in Stahl, E., Editor, “Thin Layer Chromatography,” Walker, K.C., and Reroza, M., J . Ass. Off. Agric. Chem., 1963, 46, 250. Kovacs, M. F., Ibid., 1963, 46, 884. Morley, H. V., and Chiba, M., Ibid., 1964, 47, 306. Abbott, D. C., Egan, H., and Thomson, J., J . Chromat., 1964, 16, 481. Yamamura, J., Chiba, &I., Obara, S., and Suzuki, S., Kagaku Keisatsu Kenkyaisho Hokoku, 1962, Wood, T., Nature, 1955, 176, 175. Petrowitz, H. J . , Chemikerzeitung, 1961, 85, 867. Katz, D., J. Chromat., 1964, 15, 269. Salo, T., Salminen, I<., and Fiskari, K., 2. Lebensmittel unters. u.-Forsch., 1962, 117, 369; Chem. Baumler, J., and Rippstein, S., Helv. Chim. A d a , 1961, 44, 1162. Kawashiro, I., and Hosogai, Y., Shokuhin Eiseigaku Zasshi, 1964, 5, 54; Chern. Abstr., 1964, Purdy, S. J., and Truter, E. V., Lab. Pract., 1964, 500; Analyst, 1962, 87, 802.Truter, E. V., “Thin Film Chromatography,” Cleaver-Hulme Press, 1963. Major, A., and Barry, H. C., J . A s s . Off. Agric. Chem., 1960, 44, 202. Fisher, R., Parson, I)., and Morrison, G., Nature, 1948, 161, 764. Ryhage, R., Analyt. Chem., 1964, 36, 759. Kaufman, C. W., and Jackson, H. W., Proc. 16th Int. Dairy Congr., Copenhagen, 1962, Section C, Wells, C. E., Bureau By-Lines (published by the U.S. Department of Health, Education and Lichtenstein, E. P., and Schulz, K. K., J . Agric. Fd Chem., 1965, 13, 57. Lichtenstein, E. P., Myrdal, G. R., and Schulz, K. R., Ibid., 1965, 13, 126. Stewart, D. K. R., Chisholm, D., and Fox, C. J . S., Can. J. PI. Sci., 1965, 45, 72. Uecker, G. C., Bruce, 137. N., and Bigger, J . H., J . Econ. Ent., 1965, 58, 266.Hascoet, M., and Adam, N., Phytiat.-Phytopharm., 1964, 13, 59. Henderson, J . L., in Gunthcr, F. A., op. f i t . , Volume 8, 1965, p. 74. Porter, P. E., in Zweig, G., 09. cit., LTolume 2, p. 1. Gutenmann, W. H., and Lisk, D. J., J . Agric. Fd Chem., 1963, 11, 301. Williams, S., Mills, P. A., and Mcl)owell, R. E., J . Ass. Off. Agric. Chem., 1964, 47, 1124. Esselborn, \T7., and ICrebs, K. G., Pharm. Ztg ver. ApotlzekerZtg, 1982, 107, 464. Bridges, R. G., Harrison, A., and Winteringham, F. P. W., Nature, 1956, 177, 86. Cohen, S., J. Ass. Off. Agric. Chem., 1959, 42, 549. Schechter, M. S., and Hornstein, I., Analyt. Chem., 1952, 24, 544. Hornstein, I., and Sullivan, W. N., Ibid., 1953, 25, 496. Hornstein, I., J. Ass. Off. Agric. Chem., 1953, 36, 367. Klein, A. K., Ibid., 1954, 37, 576. Reith, J. I?., Chern. Weekbl., 1953, 49, 689. Hornstein, I., Analyt. Chem., 1952, 24, 1036. -, J. A,ss. Off. Agric. Chem., 1954, 37, 623. Lichtenstein, E. P., Beck, S. D., and Schulz, K. R., J . Agric. Fd Chem., 1956, 4, 936. Friestad, H. O., A d a Pharmac. Tox., 1961, 18, 351. Klein, A. I<., J . Ass. Off. Agric. Chem., 1956, 39, 700. Fujii, M., Sato, H., Tsuji, K., and Sugawara, M., Bull. Natn. Hyg. Lab., Tokyo, 1954, 72, 115; Braid, P. E., and LeBoef, J., Analyt. Chem., 1957, 29, 1625. Applied Chemistry, Montreal, August, 1961, Section C3-26, p. 262. Springer-Verlag, 1965, p. 359. 15, 321; Chem. Abstr., 1963, 58, 14635. Abstr., 1963, 57, 12957. 61, 6262. p. 560. Welfare, Food and Drug Administration), 1962, 4, 67. Chem. Abstr., 1955, 49, 6009. 208. 209. 210. 211. 212. 213. 214. 215. 216. 217. 218. 219. 220. 221. 222. 223. 224. 225. 226. 227. 228. 229. 230. 231. 232. 233. 234. 235. 236. 237. 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249. 250. 251. 252. 253. 254. 255. 256. 267. 258. 259. 260. 261. 262. 263. 264. 265. 266. 267. 268. 269. 270. 271.March, 19661 CHLORINATED INSECTICIDES AND ACARICIDES 175 272. 273. 274. 275. 276. 277. 278. 279. 280. 281. 282. 283. 284. 285. 286. 287. 288. 289. 290. 291. 292. 293. 294. 295. 296. 297. 298. 299. 300. 301. 302. 303. 304. 305. 306. 307. 308. 309. 310. 311. 312. 313. 314. 315. 316. 317. 318. 319. 320. 321. 322. 323. 324. Paulig, G., Dt. LebensmittRdsch., 1960, 56, 223. Gunther, F. A., Blinn, R. C., and Barnes, M. M., J . Agric. F d Chem., 1957, 5, 198. Higgons, D. J., and Iiilbey, D. W., Chem. & I n d . , 1954, 1359; J . Sci. F d Agric., 1955, 6, 441. Bowery, T. G., in Zwcig, G., op. cat., Volume 2, p. 49. Kutschinski, A. H., and Luce, E. N., Analyt. Chem., 1952, 24, 1188. Erwin, W., Paper presented at the 1st Annual Meeting of the Pacific Branch, Entomological Beckman, H., and Bevenue, A,, J . Agric. F d Chem., 1964, 12, 183. Harris, H. J., Ibzd., 1955, 3, 939. Margot, A., and Stammbach, K., in Zweig, G., op. cit., Volume 2, p. 65. Blinn, R. C., Gunther, F. A., and Kolbezen, M. J., J . Agric. F d Chem., 1954,2, 1080. \\'heatley, G. A., Aqzn. Appl. Biol., 1965, 55, 325. Grzenda, A. R., Nicholson, H. P., Teasley, J . I., and Patric, J . H., J . Econ. Ent., 1964, 57, 616. Coulson, D. M., Sta9zford Res. Inst., Pesticide Res. Bull., 1962, 2, No. 4, p. 1. Lipke, H., and Chalkley, J., Bull. W l d Hlth Org., 1964, 30, 57. Hunter, C. G., Robinson, J., and Richardson, A., Brit. Med. J . , 1963, 221. Dale, 11:. E., and Quinby, G. E., Science, 1963, 142, 593. Hayes, IV. J., Dale, W. E., and Le Breton, R., Nature, 1963, 199, 1189. Williams, S., J . A s s . Off. Agrac. Chem., 1964, 47, 815. Miskus, R., in Zweig, G., op. cit., Volume 2, p. 97. Rosenthal, I., Frisone, G. J., and Gunther, F. A., J . Agric. F d Chem., 1957, 5, 514. Gordon, C. F., Haines, L. D., and Martin, J . J., Ibid., 1963, 11, 84. Gordon, C. F., and Schuckert, K. J., in Zweig, G., op. cit., Volume 2, p. 263. Zweig, G., Pye, E. L., and Peoples, S . George, D. A., Fahey, J . E., and Walker, I<. C., Ibid., 1961, 9, 264. Hughes, J. T., Analyst, 1961, 86, 756. Eiduson, H. P., J . A s s . 08. Agric. Chem., 1961, 44, 183. Gunther, F. A., and Blinn, R. C., J . Agric. F d Chem., 1957, 5, 517. Cueto, C., jun., and Hayes, W. J., jun., Ibid., 1962, 10, 366. Roburn, J., Chenz. & Ind., 1963, 1555. Korte, F., Ludwig, G., and Vogel, J., Justus Liebigs A n n l n Chem., 1962, 656, 135. Ludwig, G., Wies, J., and Korte, F., L i f e Sciences, 1964, 3, 123. Hardee, D. D., Gutenmann, W. H., Lisk, D. J., Gyrisco, G. G., and Edmonds, C. M., J . Econ. Ent. Morley, H. V., and Chiba, M., Can. J . PI. Sci., 1965, 45, 209. Harvey, H. E., and Harvey, W. E., N. 2. J . Sci., 1963, 6, 3. Porter, P. E., in Zweig, G., op. cit., Volume 2, p. 143. Barnes, W. W., and Ware, G. W., J . Econ. E n t . , 1965, 58, 286. Cassil, C. C., and Drummond, P. E., Ibid., 1965, 58, 356. Graham, J . R., Yaffe, J., Archer, T. E., and Bevenue, A., in Zweig, G., op. cit., Volume 2, p. 507. Terranova, A. C., and Ware, G. W., J . Econ. Ent., 1963, 56, 596. Terriere, L. C., in Zweig, G., op. cit., Volume 2, p. 209. Rusk, H. W., and Fahey, J. E., J. Agric. F d Chern., 1961, 9, 263. Hardee, D. D., Gutenmann, 11'. H., Kennan, G. I., Gyrisco, G. G., Lisk, D. J., Fox, F. H., Trim- berger, G. W., and Holland, R. F., J . Econ. E n t . , 1964, 57, 404. Bowery, T. G., in Zweig, G., op. cit., Volume 2, p. 245. Lowen, W, K., Cluett, M. L., and Pease, H. L., in Zweig, G., op. cit., Volume 2, p. 303. Cluett, M. L., Lowen, W. I<., Pease, H. L., and Woodhouse, C. A., J . Agric. F d Chem., 1960, 8, 277. Gunther, F. A,, and Blinn, R. C., 09. cit., p. 337. Fullmer, 0. H., and Cassil, C. C., J . Agric. Fd Chem., 1958, 6, 906. Cassil, C. C., and Yaffe, J., in Zweig, G., 09. cit., Volume 2, p. 473. Gunther, F. A., Blinn, R. C., and Rarklcy, J . H., J . Agric. F d Chem., 1959, 7, 104. Graupner, ,4. J., and Dunn, C . T,., Ibid., 1960, 8, 286. Dunn, C. L., in Zweig, G., 09. cit., Volume 2, p. 623. Nicolov, N. F., and Doncv, L., Z h . Analit. Khim., 1963, 18, 532. Witt, J. M., Bagatella, G. F., and Percious, J. K., Stanford lies. Inst., Pesticide Res. Bull., 1962, Received August 24th, 1965 Society of America, Lake Tahoe, California, 1953. J . Agric. Fd Chem., 1963, 11, 72. 1964, 57, 583. 2, No. 1.