41 The Pesticides Monitoring Journal is published quarterly under the auspices of the WORKING GROUP, Subcommittee on Pesticides, President's Cabinet Committee on the Environment, and its Panel on Pesticide Monitoring as a source of information on pesticide levels relative to man and his environment. The WORKING GROUP is comprised of representatives of the U. S. Departments of Agricul- ture: Defense; the Interior; Health, Education, and Welfare: State; and Transportation. The Pesticide Monitoring Panel consists of representatives of the Agricultural Research Service, Consumer and Marketing Service, Federal Extension Service, Forest Service, Depart- ment of Defense, Fish and Wildlife Service, Geological Survey, Federal Water Pollution Con- trol Administration, Food and Drug Administration, Environmental Health Service, Depart- ment of Defense, National Science Foundation, and Tennessee Valley Authority. Publication of the Pesticides Monitoring Journal is carried out by the Division of Pesticide Community Studies of the Food and Drug Administration. Pesticide monitoring activities of the Federal Government. particularK in those agencies represented on the Pesticide Monitoring Panel which participate in operation of the national pesticides monitoring network, are expected to be principal sources of data and interpretive articles. However, pertinent data iti siiinniarizcd form, together with interpretive discussions, are invited from both Federal and non-Federal sources, including those associated with State and community monitoring programs, universities, hospitals, and nongovernmental research institutions, both domestic and foreign. Results of studies in which monitoring data play a major or minor role or serve as support for research investigation also are welcome: however, the Journal is not intended as a primary medium for the publication of basic research. Manu- scripts received for publication are reviewed by an Editorial Advisory Board established by the Monitoring Panel. Authors are given the benefit of review comments prior to publication. Editorial Advisory Board members are: Reo E. Duggan, Food and Drug Administration. Chairman Anne R. Yobs, Food and Drug Administration Andrew W. Breidenbach, Environmental Health Service Thomas W. Duke, Fish and Wildlife Service William F. Stickel, Fisli and Wildlife Service Milton S. Schechter, Agricultural Research Service Paul F. Sand, Agricultural Research Service Mention of trade names or commercial sources in the Pesticides Monitoring Journal is for identification only and does not represent endorsement by any Federal agency. Address correspondence to: Mrs. Sylvia P. O'Rear Editorial Manager PESTICIDES MONITORING JOURNAL Food and Drug Administration 4770 Buford Highway, Bldg, 29 Chamblee, Georgia 30341 ^ ^ CONTENTS Volume 4 June 1970 Number 1 EDITORIAL Change in sponsorship for the Pesticides Monitoring Journal. George L. Hutton RESIDUES IN FISH, WILDLIFE. AND ESTUARIES Insecticides in the Big Bend National Park Howard G. Applegate Occurrence of pesticides in whales Page Allen A. Wolman and Alfred J. Wilson, Jr. PESTICIDES IN WATER Copper sulfate in flooded cranberry bogs. 1 1 Karl H. Deubert and Irving E. Demoranville Pesticide residues in Hale County, Texas, before and after ultra-tow volui?ie application of malathion 14 G. O. Guerrant, L. E. Fetzer, Jr., and J. W. Miles PESTICIDES IN SOIL DDT moratorium in Arizona — agricultural residues after 1 year 21 G. W. Ware, B. J. Estesen, C. D. Jahn, and W. P. Cahill APPENDIX Chemical names of compounds mentioned in this issue 25 EDITORIAL Change in Sponsorship for the Pesticides Monitoring Journal Readers will see reflected in this issue a change in organizational support for the Pesticides Monitoring Journal. The stage was set for this change by a White House announcement of November 20, 1969, stating that the Federal Committee on Pest Control (FCPC) would be replaced by a working group of a cabinet-level Committee on Pesticides. Rapid develop- ments early in 1970 resulted in the announced organi- zation, with certain name redesignations. Thus, the FCPC was replaced by the Working Group of the Subcommittee on Pesticides of the Cabinet Committee on the Environment. The FCPC (originally the Federal Pest Control Review Board) was formed in 1961 through the agreement of the Secretaries of USDA, USDI, DHEW and DOD. Its role was purely advisory. It operated by consensus, with the deliberateness inherent in that approach. No credit was sought, lest it detract from the pride of accomplishment by participating agencies; if there was an exception to this policy of reticence, it was the FCPC's name on this Journal. The White House announcement described the general functions of the Working Group as provision of day-to- day coordination and the development of program and policy proposals. Its charter will be published in the Federal Register. The organizational position of the Working Group is obviously quite different from that of the former FCPC. While too early for a definitive assessment of its effectiveness, a direct pathway for action has been afforded. The Working Group has to date taken several forthright actions. One of the first was a review of the status of the Pesticides Monitoring Journal and consideration of its continued support. Based on well-documented data presented by Mrs. Sylvia O'Rear, Editorial Manager, the Working Group agreed to sponsor continued publi- cation of the Journal. Also, the former Subcommittee on Monitoring is to be continued as the Panel on Pesticide Monitoring of the Working Group; its first task will be a presentation to the Working Group of a somewhat revised National Pesticide Monitoring Program. The importance of monitoring for pesticides has been evidenced in a number of reports, including two issued in 1969: the Report of the Committee on Persistent Pesticides (Jensen) and the Report of the Secretary's Commission on Pesticides and Their Relationship to Environmental Health (Mrak). Work is underway for a global network for environmental monitoring under the auspices of the International Biological Program. Public and Congressional interest in the quality of the environment is more acute than ever before. The Working Group will make every effort to be responsive to this climate and to support the objectives of the Pesticides Monitoring Journal. George L. Mutton Chairman Working Group Vol. 4, No. 1, June 1970 RESIDUES IN FISH, WILDLIFE, AND ESTUARIES Insecticides in the Big Bend National Park Howard G. Applegate ' ABSTRACT Soil, vegetation, birds, rodents, and lizards from the Big Bend National Park, Texas, were analyzed for insecticides. All samples from campground areas had significant concen- trations of insecticides: however, in other areas sampled, only migratory birds were found to have significant concen- trations. Introduction National parks are unique areas in which to study the movement of pollutants. The State of Texas presented to the U. S. Government 707,895 acres of land in the Big Bend Region of Texas. The area was then desig- nated as the Big Bend National Park and closed to all commercial enterprises in 1944. Prior to this time the area had been used for cattle grazing. Although no records are available, based on present day operations of ranches in the region, it can be assumed that few pesticides were used in the area. No chlorinated hydro- carbons have been used in the Park since its establish- ment in 1944. Malathion was used in limited amounts at two sites in the Park (Boquillas and Castolon) in 1966. Since chlorinated hydrocarbons are ubiquitous, it seemed of interest to determine if these as well as other compounds could be found in the Big Bend National Park. Publication of these data now will establish a base that will permit future investigations to show either an increase or decrease in pollution. I izards. birds, and rodents and samples of soil and vegetation were collected from the following sites in the Park (Fig. 1): Boquillas and Chisos Basin camp- grounds, Castolon, Croton Spring, Glenn Spring, Mav- Deparlment of Plant Sciences, Texas A&M Universitv, College Sta- tion, Tex. 77843. erick. Persimmon Gap, and Tornillo Flats. Another collection site, Lajitas, was located approximately 3 miles west of the Park. The sites were selected both to be representative of various ecosystems in the Park and to provide coverage of the entire Park. All collec- tions were made in June, July, and August of 1968. The nearest areas using insecticides are all in Mexico. They are: Santa Helena (76.3 acres), San Carlos (12.5 acres), Paso San Antonio (47.5 acres). Alamos San Antonio (57.5 acres), and Boquilla San Isidro (7.3 acres). These are all located 5-10 miles southwest of the Lajitas and Castolon sites with the exception of Santa Helena which is situated just 1 mile west of Castolon (Fig. 1). The following pesticides have been applied to cotton fields in these areas: methyl parathion, ethyl parathion, malathion, ozinphosmethyl, and DDT. The nearest area using insecticides in the United States is Redford, Tex., 80 miles northwest of the Park. Materials and Methods Birds were captured by shooting, and lizards and rodents were trapped. Soil samples were gathered by taking the top 2.5 cm of 1 square meter of ground surface; all rocks and organic debris were discarded and the remain- ing soil mixed. Plants were cut at the soil line, their leaves stripped and placed in bags. Each sample was placed in an ice chest as quickly as possible after collection. The samples were taken to Presidio, Tex., (120 miles) 3-5 days later where they were placed in deep freezers and held for 10-14 days until processing. Just prior to processing, each sample was thoroughly mixed (soil) or diced (flora and fauna). The samples were then weighed into two equal portions of 10 g each. Tissues from several organisms were pooled to obtain the required weight. One portion was spiked by adding Pesticides Monitoring Journal 10 ml of hexane containing 5 ppm fw/v) of the insecticides that were expected to be found; 10 ml of hexane was added to the remaining portion. Both portions were then treated identically. A total of four individual samples of each specimen type were prepared from each site for analysis by gas chromatography. The following extraction procedures were used (3): SOIL Ten grams of soil was placed in a flask, and 50 ml of hexane: acetone solution (9:1 v'v) added; after 1 hour on a wrist shaker, the solution was filtered off and the soil re-extracted with another 25 ml of the solution; the solutions were combined, concen- trated in a Kuderna-Danish concentrator, cleaned up by sweeping through a Kontes codistiller, and then made to volume (1 ml of liquid = 10 g of sample). Extraction efficiency based on spiked samples ranged from 81-929^ for DDT; 90-95 "^f for DDE: 85-94^r for TOE; 87-95^7^ for methyl parathion; and 93% for ethyl parathion. FIGURE 1. — Map of Big Bend National Park (Some of the areas in Mexico are oversized relative to other map features to sitow their shape and location.) BIG BEND NATIONAL PARK c> a a a Name Elevation (Fi) Name Elevation iFl) Lajitas 2200 Chisos Mountains 6705-7835 Castolon 2734 Chilicotal Mountain 4104 Maverick 2745 Talley Mountain 3800 Croton Spring 3504 Mariscal Mountain 3940 Chisos Basis campcround 5560 Santa Helena 2734* Glenn Spring 2605 San Carlos 2800' Solis Spring 2135 Paso San Antonio 2800 « Boquillas campgrou nd 1873 Alamos San Antonio 2800' Tornillo Flats 2814 Boquilla San Isidro 2800* Persimmon Gap 2971 * Approximate elevation Vol. 4, No. I.June 1970 Ten grams of meat was ground with 10 g of anhydrous sodium sulfate; the mash was extracted first with benzin and then with acetonitrile; the benzin layer was drawn off and the mash re- extracted three more times with acetonitrile (the bottom layer was drawn off each time and combined with the henzin; the combined extract was extracted three times with 2% sodium chloride; the benzin layer was then filtered through anhydrous sodium sulfate, concentrated in a Kuderna-Danish concen- trator, cleaned up by sweeping through a Kontes codistiller, and then made up to volume (1 ml of liquid = 10 g of sample). Extraction efficiency based on spiked samples ranged from 85-90% for DDT; 89-95% for DDE: 84-94% for TDE; 85- 95% for methyl parathion; and 85-93% for ethyl parathion. LEAVES Ten grams of leaves was blended with ethylacetate, filtered through glass wool, and the filtrate treated with 5 g of Nuchar; after filtering, crystalline sodium chloride was added to the filtrate and the top layer concentrated in a Kuderna-Danish con- centrator, cleaned up by sweeping through a Kontes codistiller, and then made up to volume (1 ml of liquid ^ 10 g of sample). Extraction eflficiency based on spiked samples ranged from 93-95% for DDT; 94-95% for DDE: 90-94% for TDE; 91- 95% for methyl parathion: and 89-93% for ethyl parathion. Solvents were cleaned up by the methods of Burke and Giuffrida (4). All values reported in this paper were corrected for percentage loss during extraction and clean-up. Two gas chromatographs were used in this study. The MicroTek 2500R was equipped vAih a nickel-63 detector. The operating parameters were: nitrogen gas flow — 50 ml /minute: temperatures — column 190 C, inlet 200 C, outlet 230 C, detector 270 C; power supply voltage, 54 v: pulse width, 9 microseconds; pulse rate, 30 microseconds. The column was '4" x 6', glass, packed with a 50:50 (w/w) mixture of 7% OV-17 and 9% QF-1 on 80^100 acid-washed, silane-treated Chromosorb W. The Barber-Colman was equipped with an automatic injection device. The device and all operating conditions have been described previously (2). Quantification for both instruments was by use of a digital integrator. In a 2-jLil injection, 0.05 ppm could be quantified. Samples to be analyzed by mass spectrometry were collected off the gas chromatograph. After establishing retention times for the compounds, the column was detached from the detector. A sample was injected and, at the appropriate time, a glass-wool collecting device placed on the open end (8). The samples were washed from the glass wool with hexane. The hexane was evaporated to near dryness and collected in capillary tubes. A model 21-1108 Consolidated Electrodynamics Corporation Mass Spectrometer was used for analysis of standard and collected samples. Thin-layer chromatography was used as an additional confirmatory tool. Extracts were spotted or streaked on silica gel HP 254 and hexane:ether (100:1 v/v) was used as the solvent. Compounds were located under ultraviolet radiation and their Rf's compared to those of standards on the same thin-layer plate. Silica gel containing the sample was scraped off, eluted with hexane, and the eluate injected into the gas chromato- graphs, both "as is" and spiked with standards. Results The data for soils are presented in Table 1 . Lajitas had by far the highest concentrations. Boquillas campground, Castolon, and the Chisos Basin campground had lower concentrations: all three sites had similar values. Soils from other sites had only trace concentrations of the insecticides. TABLE 1. — Concentrations of insecticides in surface soil Residues in PPM i Site Methyl Para- thion Para- thion DDE TDE DDT BoquilLis 0.49 0 1.00 0.99 0.81 Castolon 0 15 0 0.84 0.37 0.70 Chisos Basin 0 0 0.99 0.48 2.39 Croton Spring 0 0 0 0 0 Glenn Spring 0 0 O.OI 0.01 0.01 Lajitas 6.34 0 8.55 8.04 10.47 Maverick 0.19 0 0.14 0 0.11 Persimmon Gap 0.01 0.01 0.01 0.01 0.01 Tornillo Flats 0.01 0 0.03 0 0.04 ' Results are the means of four samples per site. The data for vegetation, leatherstem leaves, Jatropha dioica var. dioica Sesse ex Cerv. are presented in Table 2. The same pattern of distribution appeared as for the soils. Lajitas had the highest concentrations followed by Boquillas campground, Castolon, and the TABLE 2. — Concentrations of insecticides in leatherstem Residues in PPM > Site Methyl Para- thion Para- thion DDE TDE DDT Boquillas 0.04 0.01 0.99 0.88 1.15 Castolon 0 0 0.91 0.80 1.07 Chisos Basin 0 0 0.93 0.80 0.84 Croton Spring 0 0 0 0 0 Glenn Spring 0 0 O.OI 0.03 0.06 Lajitas 0.09 0 3.09 2.07 2.08 Maverick 0.01 0 0.51 0.03 0.04 Persimmon Gap 0 0 O.OI 0.02 0.01 Tornillo Flats 0.01 0 0.03 0 0.04 Results are the means of four samples per site. Pesticides Monitoring Journal Chisos Basin campground with closely grouped values. The remaining sites had trace amounts except for Mavericic which contained 0.51 ppm DDE. The data for the rodents are for muscle tissue only (Table 3). Several species of rodents were collected: Perognathits penicillatus Woodhouse, Sit^modon hispidus Say and Ord, Citellus variegatus Erxleben, and Citellus mexicanus Erxleben. Not unexpected, the pattern of distribution followed the patterns for soils and leather- stem. Lajitas had by far the greatest concentrations Closely grouped with lesser values were Boquillas camp- ground, Castolon, and the Chisos Basin campground. The remaining sites had only trace amounts. Data for concentrations in whole lizards are presented in Table 4. The following species were collected; Uta TABLE 3. — Concentrations of insecticides in muscle tissue of rodents Residues in PPM Site Number of Methyl Specimens Parathion Parathion DDE TDE DDT Boquillas P. penicillatus 3 1.24 0.04 1.05 1.10 1.37 C. mexicanus I 1.20 0 0.95 1.16 1.43 Castolon P. penicillatus 2 0.34 0 1.25 1.35 1.06 C. mexicanus 1 0.40 0 1.27 1.40 1.06 S. hispidus 1 0.39 0 1.24 1.30 1.16 Chisos Basin 5. hispidus 2 0 0.11 1.66 1.86 1.00 C. variegatus 2 0.06 0 1.65 1.94 1.12 Croton Spring P. penicillatus 3 0.88 0 0.08 0.01 0.01 C. mexicanus 1 0 0 0 0 0 Glenn Spring P. penicillatus 2 0 0 0 0 0.01 C. mexicanus 2 0.11 0 0.16 0 0 Lajitas P. penicillatus 3 3.19 0 3.26 3.99 4.31 C. mexicanus 1 3.23 0 3.00 3.81 4.13 Maverick S. hispidus 2 0 0 0.09 0.01 0.04 C. variegatus 1 0.01 0 0 0 0.04 C. mexicanus 1 0 0 0 0 0 Persimmon Gap P. penicillatus 2 0.07 0 0.04 0 1.39 C. variegatus I 0 0 0.06 0.04 0 C. mexicanus 1 0 0 0.05 0.04 0 Tornillo Flats P. penicillatus 4 0 0 0.01 0.01 0.06 TABLE 4. — Concentrations of insecticides in whole lizards Residues IN PPM Sfte Number of Methyl Specimens Parathion Parathion DDE TDE DDT Boquillas C. tigris 3 0.55 0.01 0.93 0.77 0.84 V. ornatus 1 0 0 1.05 0.69 0.82 Castolon U. stansburiana 4 0.07 0 0.77 0.54 0.16 Chisos Basin C. tigris 2 0.51 0 0.70 0.58 0.66 C. septemvittatus 1 0 0.01 0.61 0.63 0.62 V. ornatus 1 0 0 0.69 0.71 0.57 Croton Spring U. stansburiana 4 0 0 0 0.03 0.01 Glenn Spring C. tigris 2 0.05 0 0.04 0.06 0.04 C. texanus 2 0.05 0 0.10 0.04 0 Lajitas U. ornatus 3 0.70 0.10 1.61 1.40 1.50 C. texanus 1 0.52 0 1.69 1.50 1.32 Maverick C. tigris 2 0.01 0 0.27 0.18 0 U. ornatus 1 0 0 0 0 0.64 C. septemvittatus 1 0 0 0 0 0 Persimmon Gap U. stansburiana 2 0 0 0.13 0 0 C. septemvittatus 1 0.01 0 0 0 0 C. tigris 1 0 0 0 0 0 Tornillo Flats C. texanus 3 0 0 0.12 0 0.18 C. tigris 1 0 0 0 0 0 Vol. 4, No. 1, June 1970 TABLE 5 — Concentrations of insecticides in bird muscle Residues in PPM Site Number of Methyl Specimens Parathion Parathion DDE TDE DDT Boquillas N. borealis — M > 2 7.54 0 4.64 0.31 2.29 Dendrocia spp. — M 1 7.00 0 4.44 0.58 2.73 C. squamata — P J 1 0.01 0 0.01 0 0.01 Castolon A. bilineala — P 4 0 0 0.09 0.14 0.19 Chisos Basin P. simmla—P 4 0.03 0.03 0.11 0.03 0.03 Croton Spring C. mexicanus — P 2 0 0 0.19 0.04 0.08 A. uUamarina — P 2 0 0 0 0 0.12 Glenn Spring C. mexicanus— -P 4 0 0 0.03 0 0.01 Lajitas P. pyrrhonota — M 1 0.04 0.04 10.80 9.93 7.35 P. sinuala — P 2 0 0 0.92 0 0.33 A. hilineata — P 1 0.02 0.02 0.72 0.43 0.41 Maverick /. parisorttm — M 1 0.11 0.21 1.05 10.89 9.37 A. bilineata — P 3 0.08 0.12 0.54 0.46 0.47 Persimmon Gap P. sinuala— V 1 0 0 0.01 0 0 C. mexicanus — P 1 0 0 0 0 0.08 A. ultramarina — P 1 0 0 0 0 0 C. squamata — P 1 0.01 0 0 0 0 Tornillo Flats Dendrocia spp. — M 3 0.11 0.01 0.91 0.52 2.57 C. mexicanus — P 1 0 0 0.11 0 0.12 1 M — Migrant; P — Permanent. stanshuriana Baird and Girard, Urossaurus onuitus Baird and Girard, Cophosaunis texanus Trosche, Cneinido- phorus tigris. and Cnemidophonis septemvittatiis. Once again, specimens from Lajitas had the greatest concen- trations. Specimens from Boquillas campground, Cas- tolon, and the Chisos Basin campground had similar concentrations which were less than those found at Lajitas. Specimens from the remaining sites had only trace amounts. Data for concentrations in bird muscle are presented in Table 5. Migrant species contained the higher con- centrations; those collected were NuttaUoniis borealis Swainson, Dendroica spp., Petrochelidon pyrrhonota Vieillot, Icterus parisoriim Bonaparte. Those species which were permanent residents in the Park had low concentrations; they were Amphispiza bilineata Cassin, Pyrrhiilo.xia sinuala Bonaparte. Carpodactts mexicanus Muller, Aphelocoma ultramarina Bonaparte, and Calli- pepla squaniata Vigors. Results for resident birds showed that Lajitas. the Boquillas and Chisos Basin campgrounds, Castolon, as well as Maverick had organ- isms with higher concentrations of insecticides than the remaining sites. Discussion In this study very few specimens were gathered over a very large area. Obtaining and storing enough ice to keep the samples frozen until they could be placed in deep freezers was the chief limitation. Some of the collecting sites were 100 miles (round trip) from the nearest ice source. The desert temperatures were over 100 F in most afternoons during the collecting period. When the sites were selected, it was anticipated that Lajitas and Castolon would have the highest concen- trations due to their proximity to sprayed areas. It was also felt that all other sites would show trace concen- trations if any insecticide residues were found at all, based on the distances of the sites from areas of insecticide application and orography of the study Park areas. Our data show that Lajitas or Castolon usually con- tained more insecticides per sample than all other sites combined. Concentrations found at Boquillas camp- ground and the Chisos Basin campground were unex- pected. Although the elevation and ecosystem of the Boquillas campground is similar to Lajitas and Castolon, Boquillas is located some distance from the cotton- growing areas in Mexico where insecticides were used. As can be seen from Fig. 1, the Chisos Basin camp- ground is literally a basin with a rim 2,000 feet above it that would serve as a shield from [jesticides applied 3,000 feet below. All four sites are used by people. They either contain camping sites (the Park areas) or a trading post (Lajitas). The Chisos Basin is heavily used during the warm months, while Boquillas and Castolon are used during the colder months. Lajitas is used all year long with heavy tourist use in the summer months. It would seem, therefore, that the presence of insecticides can be linked to the presence of people and not to drift from cotton fields. Pesticides Monitoring Journal It is possible that the insecticides found were applied by campers. The amounts needed to be used by the campers to reach the DDT soil concentrations reported here can be calculated if it is assumed that the insecticide values are characteristic for the total camping area; that an acre of soil 3 inches deep weighs 1 million pounds (a figure commonly used in agronomy texts); and that a pressurized spray can contains 0.9% DDT. (The last assumption is based on a survey made at local markets.) The size of the three camping sites and the number of persons using each one during the camping season of June, July, and August 1968 are as follows: Chisos Basin — 10 acres and 29,288 persons; Castolon — 6.5 acres and 738 persons; Boquillas — 15 acres and 13,895 persons. The following data were calculated: Chisos Basin con- tained 17,458 g of DDT in the soil which would require 4,367 cans containing 0.9%, or 1.4 cans per person. Castolon contained 5,630 g of DDT in the soil which would require 1,408 cans containing 0.9% DDT, or 1.9 cans per person. Boquillas contained 19,050 g of DDT in the soil which would require 4,762 cans con- taining 0.9% DDT, or 0.3 cans per person. No infor- mation is available on family groups versus lone campers or trailer users versus open-air campers. No data are available on the people who visited the Lajitas trading post (population 15). Undoubtedly, several hundred tourists per year stop while on their way to or from the ghost town of Terlingua. Lajitas is situated in a cul-de-sac opening to the southwest. The prevailing winds are southerly and southeasterly. While it is possible that drift from the cotton fields blew over the mesas to Lajitas, it does not seem probable. Lajitas is used by Mexicans; however, it is not an authorized Mexican port of entry for insecticides. Unauthorized movement of insecticides across the bor- der may have occurred; if so, then haste in handling the sacks at night may have resulted in spillage. The amounts of insecticides found at Lajitas are similar to those found adjacent to cotton fields at Presidio, Tex. (7,5,6.7). Concentrations found at Boquillas camp- ground, Castolon, and the Chisos Basin campground are similar to those found 3 miles from the cotton fields at Presidio, while concentrations at the other sites in the Big Bend National Park are similar to those found more than 9 miles from the cotton fields at Presidio. The relative proportions of TDE to DDE and DDT are also similar to those found in the Presidio Basin. The differences in concentrations between migratory and resident birds in the Park may indicate that the migratory birds ingested the pesticides elsewhere. How- ever, the listed migratory species are predominantly insectivorous while the species listed as resident are predominantly seed eaters. Thus, the concentration Vol. 4, No. 1, Jiwe 1970 differences may reflect dietary habits. If this is true, then the compounds could have been ingested in the Park. The data showing no species differences in insecticide concentrations in the lizards and the rodents supports data gathered from the Presidio Basin. Previous studies have shown that organisms occupying similar niches within an ecosystem had similar insecticide concentra- tions when similar tissues were compared (5,6). Care must be used in interpreting data from whole organ- isms; whole gravid female lizards had higher insecticide concentrations than whole non-gravid female lizards. However, when muscle versus muscle comparisons were made, there were no differences between a gravid and a non-gravid female lizard. Results of an earlier study showed that higher concentrations in the whole gravid lizards were due solely to insecticide accumula- tions in the lipids of the developing eggs (5). Acknowledgments This is Technical Article No. 8155, Texas Agricultural Experiment Station, College Station. The research was supported by Grant AP 28, National Center for Air Pollution Control and Grant CC 272, National Com- municable Disease Center. I wish to thank Chief Ranger Rov Allen and Chief Naturalist Roland H. Waver, Big Bend National Park for collecting permits, help in selecting the sampling sites, and critical reading of the manuscript. See Appendix for chemical na paper. of compounds mentioned in this LITERATURE CITED ( n Applegale. H. G. 1966. Pesticides at Presidio. II. Vegeta- tion. Tex. J. Sci. 18:266-271. 12) Applegate, H. G. and G. Chillwood. 1968. Automation of pesticide analysis. Bull. Environ. Contamination Toxicol. 3:211-226. (3) Biirchfield, H. P. and D. E. Johnson. 1965. Guide to the Analysis of Pesticide Residues. U. S. Dep. of Health, Educ. and Welfare, Public Health Service, Office of Pesticides. Washington. D. C. 20204. '4) Burke. J. and L. Giiiffrida. 1964. Investigations of elec- tron capture gas chromatography for the analysis of multiple chlorinated pesticide residues in vegetables. J. Ass. Agr. Chem. 47:326-342. i5} CuUey, D. D. and H. G. Applegate. 1967. Pesticides at Presidio. IV. Reptiles, Birds, and Mammals. Tex. J. Sci. 19:301-310. 16} Ciilley. D. D. and H. G. Applegate. 1967. Insecticide concentrations in wildlife at Presidio, Texas. Pesticides Monit. J. 1:21-28. (7) Laliser, C. and H. G. Applegate. 1966. Pesticides at Presidio, III. Soil and Water, Tex. J. Sci. 18:387-395. IS) Miimma. R. O. and T. R. Kantner. 1966. Identification of halogenated pesticides by mass spectroscopy. J. Econ. Entomol. 59:491-492. Occurrence of Pesticides in Whales Allen A. Wolman ' and Alfred J. Wilson, Jr.° ABSTRACT Organochlorinc pesticides were found in tissue samples of brain, blubber, and liver from 6 of 23 gray whales, Eschrich- tius robustus, and in each of 6 sperm whales, Physeter catodon. collected near San Francisco, Calif., in 1968 and 1969. Concentrations of DDT or its metabolites ranged up to 0.36 ppm in blubber tissue of gray whales, and 6.0 ppm in blubber tissue of sperm whales. The highest dieldrin concentrations were 0.075 ppm in gray and 0.019 ppm in sperm whales. Introduction These chemicals, although they mainly affect the nervous system, may after long-term accumulation cause sterility and mortality in adults or mortality among progeny, as observed in pelicans and cormorants (2). They may interfere with steroid hydroxylation, resulting in the lowering of calcium deposition in eggshells (6); interact in polyunsaturated fatty acid metabolism, resulting in riboflavin deficiency {13); and cause degenerative changes in rat liver tissue (10). This report records the amounts of DDT and its metabolites and dieldrin found in the tissues of gray whales, Eschrichtius robustus, and sperm whales, Physeter catodon. Residues of the chlorinated hydrocarbon insecticides, DDT and dieldrin, have been found in various forms of wildlife, mainly in fatty tissues (5. 7); these com- pounds are resistant to chemical breakdown by digestive and physiological processes in mammals, birds, and fish (5). Pesticide residues have been found in blubber, brain, and other body tissues of marine mammals in Antarctica {4. 12) and in gray seals, Halichoerus grypus; common seals. Phoca vituUna: and harbor porpoises, Phocoena phocoena. from the coasts of Scotland (7). In addition, juvenile and adult harp seals, Phoca groen- landica. on the Canadian Atlantic coast contained pesti- cides, principally of the DDT group (7), as did porpoises examined by Wilson in Florida {unpublished data). Anas and Wilson (/) found DDT and its metabolites and dieldrin in brain and liver samples from the northern fur seal, Callorhinus ursinus. Koeman and van Gen- deren (8) found 9.6 to 27.4 ppm of DDT and 0.07 to 2.30 ppm of dieldrin in harbor seals in the Netherlands. ' Bureau of Commercial Fisheries. Marine Mammal Biological Labora- tory, Seattle. Wash. 98115. = Bureau of Commercial Fisheries. Biological Field Station. Gulf Breeze, Fla. 32561. Sampling Samples of brain, blubber, and liver tissues were col- lected from 23 gray whales taken during migration sea- sons in March-April 1968, December 1968-January 1969, and March 1969. Similar samples from six sperm whales were taken during May and November 1968. All of the collections were made off San Francisco, Calif. About 100 g of each tissue was collected from each animal. Tissues were frozen from the time of collection until time of analysis. Maturity and reproduc- tive condition were determined by histological examina- tion of the testes and mammary glands and examination of ovaries for corpora lutea and corpora albicantia. A nalytical Procedures Liver, brain, and blubber tissues were analyzed for BHC, heptachlor, aldrin, heptachlor epoxide, toxaphene, meth- oxychlor, dieldrin, endrin, and the o.p' and p,p' isomers of DDE, DDD (TDE), and DDT. Tissues were thawed and mixed with anhydrous sodium sulfate in a blender. The mixture was extracted for 4 hours with petroleum Pesticides Monitoring Journal ether in a Soxhlet apparatus. Extracts were concentrated and partitioned with acetonitrile. The acetonitrile was evaporated just to dryness and the residue eluted from a Florisil column (9). Sample extracts were then identi- fied and quantified by gas chromatographs equipped with electron capture detectors. Column packing and operating parameters were as follows: Columns: 5' x ''s". glass, packed with 3^^ DC-200, 5% QF-1. and a 1:1 ratio of 3% DC- 200 and 5% QF-1, all on 60/80 Gas Chrom Q Temper- Detector 210 C ature: Injector 210 C Oven 190 C Carrier: Prepurified nitrogen at a flow rate of 40 ml minute A few samples were analyzed using thin-layer chro- matography. Laboratory tests gave the following re- covery rates: p.p'-T)DE. 80-85^7^: p.p'-UDD. 92-95%; p,p'-DDT, 91-95 '^r; dieldrin, 85-90^r. The lower limit of sensitivity was 0.010 ppm (mg/kg, wet weight). Data in this report do not include a correction factor for percentage recovery. Polychlorinated bi- phenyls (PCB's) reported by Holden and Marsden (7), were not detected in our samples. All values reported were calculated on a wet weight basis. Discussion DDT and its metabolites were found in 4 males and 2 females of the 23 (26% ) gray whales examined (Table 1). All four of the whales taken during spring 1968 contained residues at levels ranging up to 0.058 ppm. In contrast, only 2 of 19 whales sampled during the 1968-69 migration contained pesticides, but DDT and its metabolites were about 5 times more concentrated in these 2 whales; the highest level was 0.36 ppm. One male and one female gray whale had trace amounts of DDE in the brain tissue. The liver tissue of all the gray whales was free of pesticides. Dieldrin was found in all the spring 1968 samples but was lacking in the 1968-69 migration series. Gray whales feed during the summer in the Bering and Chukchi Seas, principally in large areas of shallow water with an abundant benthos. Their food is primarily benthic amphipods, mostly Ampelisca macrocephala, in addition to a number of benthic isopods, mysids, mol- lusks, polychaetes, and hydroids (14, 15). While mi- grating between their northern summer grounds and their winter calving grounds off Baja California and during their stay on the wintering grounds they eat vir- tually no food (//). Body weight greatly decreases dur- ing the migration, resulting in a great use of stored body fat. Because they fast during migration, gray whales Vol. 4, No. I.June 1970 undergo considerable physiological stress, which is ac- centuated in lactating cows, whose calves suckle for about 6 months. All of the sperm whales sampled contained DDT and metabolites in concentrations ranging from 0.010 to 6.0 ppm (Table 2). All had DDT or its metabolites in each of the three kinds of tissue examined. Concentrations ranged up to 0.12 ppm in brain tissue, 6.0 ppm in blub- ber, and 0.35 ppm in liver. Dieldrin was found in two samples of blubber taken in May 1968 but was absent in all samples from November 1968. Sperm whales are in all the world oceans; they feed mainly on squid (mostly the larger species, such as Moroteuthis rohustus off California), octopuses, and deepwater fishes, such as sharks, skates, hake, and to a lesser extent, rockfish. In gray whales and sperm whales the concentrations of pesticides were highest in the blubber. Since a relatively high proportion of the body weight of whales is in the form of blubber, the animals may contain high total amounts of pesticides. For instance, the estimated weight of gray whale No. 1969-35 is 14,300 kg, and possible blubber proportion is 29% (/6), or 4,147 kg. At a concentration of 0.36 ppm DDE in its blubber (Table 1 ), the whale might carry about 1.5 g of DDE in this tissue. See Appendix for chemical names of compounds mentioned in this paper. LITERATURE CITED (1) Anas, R. E. and A. J. Wilson, Jr. 1970. Organochlorine pesticides in fur seals. Pesticides Monit. J. 3(4): 1 98-200. (2) Anderson. D. W.. J. J. Hickcy, R. W. Rischrough, D. F. Hughes, and R. E. Chrislenscn. 1969. Significance of chlorinated hydrocarbon residues to breeding pelicans cormorants. Can. Field-Natur. 83(2):9 1-112. (31 Gakstatter, J. H. 1967. The uptake from water by sev- eral species of freshwater fish of p.p'-DDT, dieldrin, and lindane; their tissue distribution and elimination rate. Diss. Abstr. B27, 11, 3820. 14) George. J. L. and D. E. H. Frear. 1966. Pesticides in the Antarctic. J. Appl. Ecol. 3 (Suppn:155-167. (5) Greenwood, R. J., Y. A. Greichus, and E. J. Hugghins. 1967. Insecticide residues in big game mammals of South Dakota. J. Wildlife Manage. 31:288-292. (6) Mickey, J. J. (editor). 1969. Peregrine falcon popula- tions, their biology, and decline. Univ. Wisconsin Press. Madison, Wis.. 596 p. (7) Holden. A. V. and K. Marsden. 1967. Organochlorine pesticides in seals and porpoises. Nature 216(5122): 1274-1276. (8) Koeman. J. H. and H. van Genderen. 1966. Some pre- liminary notes on residues of chlorinated hydrocarbon insecticides in birds and mammals in the Netherlands. J. Appl. Ecol. 3 (Suppl.):99. (9) Mills. Paul A.. J. H. Onley, and R. A. Gailher. 1963. Rapid method for chlorinated pesticide residues in nonfatty foods. J. Ass. Offic. Agr. Chem. 46:186-191. (10) Phillips. W. E. ]. 1963. DDT and the metabolism of Vitamin A and carotene in the rat. Can. J. Biochem. Physiol. 41:1793-1802. (in Rice, D. W. and A. A. Wolman. The gray whale Eschrichtius robustus: Life history and ecology. J. Mammal. Spec. Publ. 3 (in press). (I2i Sladcn. W. J. L.. C. M. Menzie. and W . L. Reichel. 1966. DDT residues in Adelie penguins and a crab- eater seal from Antarctica. Nature 210(5037):670-673. (IS) Tinslcy. I. J. 1966. Nutritional interactions in dieldrin toxicity. Agr. Food Chem. 14:563-565. (M) Tomilin. A. C. 1937. Kity dal-nego Vostoka [The whales of the far east]. Uch. Zap. Mosk. Gos. Univ., Ser. Biol. Nauk, 12:119-167. (Russian with English summary.) f/5j Zenkovich. B. A. 1937. The food of far-eastern whales. C. R. (Dokl.) Acad. Sci. URSS, 16(4):23 1-234. (16) 1937. Esche o serom kalifomisskom kite (Rhachiancctes glaucus Cope, 1864). [More on the gray California whale (Rhachiancctes glaucus Cope, 1864).] Vestnik Akad. Nauk SSSR. Dal'nevostochnyi Fil., 23: 91-103. TABLE 1. — Pesticides in brain, bhibber, and liver tissues of migrating gray whales Field Length (me- Sex Repro- ductive Residues in ppm (mg/kg, WET WEIGHT)! Direction of Date COL- DDT DDD DDE Dieldrin MIGRATION AND a K s « SEASON COLLECTED NUMBER LECTED ters) condition z ffl p: Z s O; z ^ z n ft I 3 u I 3 > < 3 > < 3 > m m .3 a n ,J o o .3 n n u Northbound Spring 1968 1968-57 3-11-68 10.68 d- Immature 0 0.026 0 0 0.034 0 0 0.040 0 0 0.051 0 1968-58 3-11-68 11.75 cT Mature 0 0.022 0 0 0.034 0 0 0.041 0 0 0.044 0 1968-60 4- 2-68 10.88 cT Immature 0 0.033 0 0 0.029 0 0 0.046 0 0 0.055 0 1968-64 4- 5-68 9.40 ? Immature 0 0.028 0 0 0.058 0 0 0.046 0 0 0.075 0 Southbound Winter 1968-69 1968-245 12-21-68 12.85 9 Mature 0 0 0 0 0 0 0 0 0 0 0 0 1968-247 12-22-68 11.30 d- Mature 0 0 0 0 0 0 0 0 0 0 0 0 1968-252 12-29-68 13.20 9 Mature 0 0 0 0 0 0 0 0 0 0 0 0 1968-256 12-30-68 11.85 d- Mature 0 n 0 0 0 0 0 0 0 0 0 0 1969-2 1- 2-69 12.15 d- Mature 0 0 0 0 0 0 0 0 0 0 0 0 1969-13 1- 5-69 13.30 9 Mature 0 0.13 0 0 0.091 0 0.011 0.30 0 0 0 0 1969-22 1- 7-69 11.60 9 Mature 0 0 0 0 0 0 0 0 0 0 0 0 1969-23 1- 7-69 13.40 9 Mature 0 0 0 0 0 0 0 0 0 0 0 0 1969-24 1- 8-69 12.15 9 Mature (:> 0 0 m 0 0 (2> 0 0 (2) 0 0 1969-26 1- 8-69 11.30 9 Immature 0 0 0 0 0 0 0 0 0 0 0 0 Northbound Spring 1969 1969-30 3- 2-69 11.32 d Immature 0 0 0 0 0 0 0 0 0 0 0 0 1969-31 3- 2-69 10.92 d Mature 0 0 0 0 0 0 0 0 0 0 0 0 1969-32 3- 5-69 12.37 d Mature 0 0 0 0 0 0 0 0 0 0 0 0 1969-33 3- 9 69 11.92 d Mature 0 0 0 0 0 0 0 0 0 0 0 0 1969-34 3- 9 69 12.44 d Mature 0 0 0 0 0 0 0 0 0 0 0 0 1969-35 3-10 69 11.83 d- Mature 0 0.13 0 0 0.19 0 0.011 0.36 0 0 0 0 1969-44 3-15 69 12.65 9 Mature 0 0 0 0 0 0 0 0 0 0 0 0 1969-45 3-15 69 11.93 9 Immature 0 0 0 0 0 0 0 0 0 0 0 0 1969-48 3-16-69 12.92 9 Immature 0 0 0 0 0 0 0 0 0 0 0 0 1 0 = pesticides not detectable (<0.010 ppm). ^ Brain tissue not sampled. TABLE 2. — Pesticides in brain, blubber, and liver tissues of sperm whales Date col- lected Length (me- ters) Sex Repro- ductive con- dition Residues in ppm (mg/ko, wet weight)i DDT DDD DDE Dieldrin number z i CQ > .J z i n I > 3 2 i m i 2 z i g g > 1968-119 5- 9-68 12.19 d" Mature 0.026 2.6 0.042 0.010 0.83 0.066 0.12 6.0 0.33 0 0.016 0 1968-120 5-15-68 14.94 d Mature 0 0.86 0.11 0 0.22 0.029 0.014 0.74 0.15 0 0.019 0 1968-228 11- 7-68 10.26 9 Mature 0 1.2 0.044 0 0.49 0.10 0.059 4.0 0.35 0 0 0 1968-229 11- 7-68 10.77 9 Mature 0.013 1.3 0 0 0.48 0.053 0.054 3.3 0.23 0 0 0 1968-232 11- 7-68 11.20 9 Mature 0.014 2.1 0.028 0.012 0,59 0.042 0.073 4.4 0.19 0 0 0 1968-233 11- 7-68 10.92 9 Mature 0.022 1.9 0.016 0 0.50 0.056 0.074 3.3 0.18 0 0 0 0 = not detectable (<0.010 ppm). 10 Pesticides Monitoring Journal PESTICIDES IN WATER Copper Sulfate in Flooded Cranberry Bogs^ Karl H. Deubert and Irving E. Demoranville ABSTRACT Cranberry bogs are treated with copper sulfate to control algal growth. In order to assess possible water pollution after release of treated floodwaler into streams and ponds, the rate at which copper disappeared from the water after treatment was monitored in two separate bogs. In both bogs the concentration of copper 25 hours after application was higher than expected due to smaller volumes of floodwater. During the first 6 days after treatment, copper concentrations decreased rapidly, and after 10 days about 95% of the copper had disappeared. When floodwater was released about 4 weeks after treatment, the concentration of copper was at the same level found in the water prior to treatment. Introduction Cranberry bogs are potential sources of water pollution. Dieldrin and DDT have been shown to be very per- sistent in bog soil (2, 4), and adsorption of dieldrin on organic matter suggests that floodwater may remove small quantities of this insecticide from bogs with or- ganic colloids (/). The use of copper sulfate in cranberry bogs as an algicide is common practice. Growers apply copper sulfate at a rate of 4 Ib/acre-foot of water. This amount is in- tended to yield about 0.4 ppm copper, an amount which is toxic to fish although the limiting concentrations may vary widely. Since the fate of copper sulfate in floodwater on cran- berry bogs was not known, a monitoring study was car- ried out to examine the disappearance of copper from water and assess the importance of this compound as a potential water pollutant. This study was conducted un- der practical field conditions. 1 From the University of Massachusetts, Cranberry Experiment Station, East Wareham, Mass. 02538. Vol. 4, No. 1, June 1970 Sampling Areas Two bogs, located near East Wareham, Mass., were chosen for the monitoring study. Bog 1, a 10-acre bog planted in 1882, was divided by earth dikes into 4 bays. One bay, 2 acres in size, was treated with copper sulfate on April 22, 1969. Floodwater was taken from a nearby 1 8-acre pond and, after the treatment was finished, re- leased into the same pond on May 20, 1969. Bog 2, a 2-acre bog about as old as Bog 1 , received the water from a river. It was treated on April 22, 1969. Flood- water was released on May 27, 1969, into the same pond into which the water from Bog 1 was drained. No special application or flooding techniques were used; application of the copper sulfate and depth of floodwater were the same as for a normal treatment. Neither of the bogs was level, resulting in a variation of between 4 and 13 inches in the depth of the water. The average depth was estimated to be 6 to 7 inches. In each bog two sampling sites were chosen in shallow water and two sampling sites in the deeper water. Sampling Methods Water samples were taken in acid-washed 800-ml glass jars. For surface water samples, the jars were submerged until half of the opening was about Va inch below the water surface. To obtain subsurface water samples, covered jars (glass cover and rubber seal) were hori- zontally submerged to the bottom. The covers were then removed to fill the jars and put back in place before the jars were raised out of the water. This procedure al- lowed a distinction between surface and subsurface water only; the water was not deep enough to justify the use of different depths as reference points. 11 The first samples were taken prior to the treatment. Sampling was continued 1, 3, 6, 8. 10, and 28 days after treatments. The last samples were taken I day before the bogs were drained. Pond water samples were ex- amined at drainage and 1 and 3 days after. Application of Copper Sulfate The hogs were first flooded: then copper sulfate was ap- plied at the rate of 4 lb 'acre-foot of water. Distribution was made by placing the appropriate amount of copper sulfate in a burlap bag and dragging the bag through the water. This simple method assures a rather uniform distribution without excessive initial fixation by the soil. A nalytical Procedures All water samples were analyzed within 15 to 20 minutes after they were taken. The samples were filtered through Whatman No. 44 paper to remove most suspended solids. Extraction and quantitation of copper was carried out using "bathocuproine" (7). At low residue levels (<0.04 ppm) 150 ml instead of 100 ml of water was extracted. Standards were prepared as described in APHA Stand- ard Methods for the Examination of Water and Waste Water (7). Recovery from 10 flood water samples forti- fied to 0.3 ppm copper and then filtered through What- man No. 44 was 94 ± 0.8%. Residue data reported in this paper are the means of two separate analyses and are corrected for the percentage of recovery. Results and Discussion Results of the monitoring studies on both bogs are given in Table 1 . The pond and river water, prior to flooding and treatment in the bogs contained 0.02 ppm copper. The same amount was present 1 and 3 days after the bogs were drained back into the pond water. At least during the first 3 days after treatment, the cop- per concentration in the treated floodwater will generally be greater than 0.4 ppm (the amount the applied dosage is intended to yield) for two reasons. First, the depth of water on a bog may vary considerably due to variations in grade, some bogs being as much as 2 feet higher at one end. The grower applies copper sulfate in relation to the depth of the water (4 lb acre-foot of water) and not in relation to the surface area (4 lb/acre). Secondly, the volume of water may be reduced somewhat due to evaporation. In the present study the copper concentration 24 hours after application was about twice as much as expected. After 6 days, however, about 91% of the copper dis- appeared leaving a concentration of 0.07 ppm in the floodwater. Ten days after treatment only 0.04 ppm copper was found, indicating that more than 95% of the copper had been "fixed." The diff'erence between levels in surface and subsurface water suggested that copper was adsorbed onto soil particles and that copper may disappear faster from the subsurface water. TABLE 1 . — Disappearance of copper from floodwater Bog and Site No. Before treatment Copper Residues in ppm Days after treatment 12 SURFACE WATER Bog 1 Site 1 Site 2 0.03 0.02 0.90 0.77 0.61 0.42 0.12 0.09 0.11 0.05 0.04 0.05 0.02 0.02 Bog 2 Site 1 Site 2 0.02 0.02 0.77 0.79 0.52 0.49 0.10 0.11 0.05 0.06 0.05 0.05 0.02 0.02 Mean 0.02 0.81 0.51 0.10 0.06 0.05 0.02 SUBSURFACE WATER Bog 1 Site I 0.03 0.98 0.09 0.06 0.05 0.04 0.02 Site 2 0.02 0.87 0.11 0.05 0.08 0.05 0.02 Bog 2 Site 1 0.02 0.99 0.05 0.06 0.06 0.03 0.02 Site 2 0.02 0.94 0.07 0.07 0.05 0.03 0.02 Mean 0.02 0.94 0.08 0.06 0.06 0.04 0.02 Mean (both water 0.02 0.87 0.29 0.08 0.06 0.04 0.02 levels) PESTicroES Monitoring Journal The findings on the disappearance of copper from flood- water were in agreement with data obtained by other workers. Riemer and Toth (5) found that 90-100% of the copper applied as copper sulfate in various concen- trations to small ponds disappeared from surface water 10 days after the treatment. Tobia and Hanna (6) re- covered only 1 ppm copper sulfate from unfiltered Nile water 6 hours after addition of 20 ppm CuSOo. The question remains whether copper, adsorbed on bog soil, can be considered as a potential pollutant which may contaminate streams and ponds in small amounts. According to Lucas {3) organic soils held copper tena- ciously in the zone of placement. The equivalent of 48 lb/ acre of copper sulfate was found in the upper 8 inches of organic soil 5 years after the application of 50 lb/ acre. Tobia and Hanna (6) obtained only 4.2 and 2.0 ppm copper in 1,000 ml of water leached through 30 g of soil (2.6% C) containing 24 and 30 ppm copper. They stated that organic matter and soil reaction rather than clay content correlated with the retention of copper by soil. Since organic matter con- trols the sorptive capacity of bog soil (/), it can be as- sumed that copper adsorbed on soil particles is released only in small quantities. A cknowledgment This study was supported by Hatch Project 293. The technical help of Mr. R. Alberghini is appreciated. LITERATURE CITED (/) Deubert, K. H. 1970. Soil characteristics influencing dieldrin adsorption. Proc. of the Cornell 1970 Waste Manage. Conf. In press. (2) Deubert. K. H. and B. M. Zuckerman. 1969. Distribu- tion of dieldrin and DDT in cranberry bog soil. Pesti- cides Monit. J. 2; 172-1 75. (3) Lucas, R. E. 1948. Chemical and physical behavior of copper in organic soils. Soil Sci. 66:1 19-129. (4J Miller. C. W. 1966. Dieldrin persistence in cranberry bogs. J. Econ. Entomol. 59:905-906. (5) Riemer. D. N. and S. J. Toth. 1967. Behavior of copper sulfate in small ponds. Northeastern Weed Contr. Conf. 21:534-540. 16} Tobia. S. K. and A. S. Hanna. 1958. Effect of copper sulfate added to irrigation water on copper status of Egyptian soils. I. Amount of copper retained by soils. Soil Sci. 85:302-306. (7) American Public Health Association, Inc. 1967. Stand- ard methods for the examination of water and waste water. New York, N.Y. Vol. 4, No. 1, June 1970 13 Pesticide Residues in Hale County, Texas, Before and After Ultra-Low Volume Aerial Application of Malathion ^ G. O. Guerrant, L. E. Fetzer. Jr., and J. W. Miles ABSTRACT In ultra-low volume aerial sprayings of malathion for con- trol of arthropod-borne encephalitis in 1967 in Hale County, Texas, the amount of malathion deposited was determined by measuring the amount found on exposed filter papers: the average concentration found was 1.5 mg/ft', or 65% of the applied dosage. The maximum concentration found in environmental waters was 0.5 ppm malathion, which de- composed with a half -life of 0.5 to 10 days, depending upon pH. Methods for collecting, transporting, and determining malathion residues are described. The chlorinated hydrocarbon residues present in the spray area were monitored also and found to consist mainly of DDT and BHC isomers. Most of the sampled waters con- tained less than I ppb of the individual chlorinated pesticides. Introduction Malathion was used against Culex tarsalis mosquitoes in selected communities in Hale County in northwestern Texas from June through August 1967 to reduce the transmission of encephalitis virus to humans. This is an area of high incidence of human infection from July through September after the buildup of the mosquito population following spring rains and agricultural irriga- tion. To evaluate the effectiveness of ultra-low volume (ULV) aerial application of malathion in reducing vector mosquito populations and the occurence of human encephalitis, the towns of Plainview and Abernathy were ^ From the Technical Development Laboratories, Laboratory Division, National Communicable Disease Center. Health Services and Mental Health Administration, Public Health Service, U. S. Department of Health, Education, and Welfare, P. O. Box 2167, Savannah, Ga. 31402. 14 selected for study, with Petersburg as a control. Mala- thion low-volume concentrate (95%) was applied at the rate of 3.0 fluid oz/acre by the U. S. Air Force Tactical Air Command, Special Aerial Spray Flight, 4500th Air Base Wing, Langley Air Force Base, Va. The applica- tion was made from a C-123 cargo aircraft flying at an altitude of 150 feet at an air speed of 150 mph. Dur- ing the test local municipal agencies applied 3% BHC- 5% DDT dust as an independent routine mosquito con- trol measure. In addition, the usual crop protection pesticides were being used since the area is predominantly agricultural. Mitchell et al. (4.5) have described the environmental features of Hale County and reported on the effect of malathion ULV sprayings on the mosquito populations and on western encephalitis virus activity. The primary objective of the study reported here was to measure the actual amount of malathion deposited and to determine the distribution and persistence of malathion in the treated areas. Other pesticides were also measured to obtain the total pesticide load in the environment before and after the malathion application. Water sources were selected for sampling since this represents an important route of introducing pesticides into higher vertebrates. To insure accuracy of data obtained, studies were also carried out to develop efficient methods for collecting, storing, and transporting samples. This included a study of the decomposition rate of malathion in field waters; laboratory data on the rate of hydrolysis at various pH values are presented. Prespray samples for this study were taken on June 12, 1967; other samples were taken after the ULV sprayings on June 16, 21-22, July 26-27, and August 24, Pesticides Monitoring Journal 1967. Fig. 1 shows the corporate limits of the three municipalities sampled with sample sites numbered. These sites are identified and described in Table 1. FIGURE 1. — Sample sites within Plainview, Abeniathy, and Petersburg, Tex. -Description of water sample collection sites shown on Fig. I LOCATION OF EXPOSED FILTER PAPERS XXX PETERSBURG ABERNATHY Measurement of Malathiou Application Rate Assessment of the mosquito kill and its effectiveness in reducing viral activity is dependent upon a knowledge of the dispersal and deposition of the maiathion appli- cations. The aerial application equipment and procedures used had been developed and tested in previous usage (7,2). The output of insecticide from the spray system was determined and adjusted by selection of appropriate orifice size and number to deliver at a known rate. This delivery rate was made compatible with the height and speed of the aircraft to give a coverage of 3.0 oz of maiathion per acre. Spraying was done only during periods of optimum atmospheric conditions such as with wind currents below 10 mph and during periods of tem- perature inversions. Prior to spraying, filter papers 15.0 cm in diameter were attached to plywood panels and distributed to prime locations within the area where they were exposed with Vol. 4, No. I.June 1970 PLAINVIEW 1 Intermittent stream 7.2 2 Intermittent stream 7.4 3 Excavation pond 7.8 4 Borrow pit 8.0 5 Irrigation reservoir 9.5 6 Stock water tank 8.5 7 Stock water trough 8.6 8 Excavation pond 7.7 9 Stock water tank 7.7 10 Fishpond 8.1 11 Intermittent stream 7.6 12 Playa 7.9 13 Playa 6.8 14 Stock water tank 7.8 15 Intermittent stream 7.6 16 Stock water trough 7.6 17 Lake 7.4 I Auto tire 8.2 II Auto tire 8.0 ABERNATHY 18 Lake 7.4 19 Stock water tank 7.9 20 Stock water tank 8.5 21 Lake 7.6 22 Lake 7.4 23 Lake 7.6 24 Lake 7.7 25 Stock water tank 8.9 PETERSBURG 26 Playa 7.4 27 Playa 7.4 28 Stock water tank 29 Stock water tank 8.2 30 Playa 7.6 31 Borrow ditch 7.4 32 Stock water tank 8.0 surfaces parallel to the ground. Arrangement of the filter papers is sketched in Fig. 1; these were located along existing roads with the sampling pattern along hnes parallel and also at right angles to the spraying flight pattern. Immediately after spraying, the papers were in- dividually collected and stored in 6-oz prescription bot- tles which were shipped to Savannah, Ga., for analysis. At the laboratory 50 ml of hexane was added to each bottle. The bottles were shaken and set aside overnight to extract the maiathion for analysis by gas chromato- graphy. A 1-^1 sample was analyzed by using a Varian Aerograph Series 204b gas chromatograph equipped with a sodium thermionic detector. Separation was on a 5' X Vs" SE-30 column on Chromosorb W, 60/80 mesh, DMCS, at 190 C. Concentrations were determined from peak area measurements. To determine the effect of storage on the malathion- exposed filter papers, tests were devised to study decom- position under simulated field conditions. Maiathion in hexane was added to a series of papers; the solvent was permitted to evaporate; and the dried papers were stored. 15 Samples were analyzed at selected times, and decom- position was found to be at the rate of 6% per day. The results showed that with the field samples, less de- composition would have occurred if solvent had been added to each sample when it was collected rather than when it arrived at the laboratory. Since this was not done, appropriate corrections were made for malathion decomposition during transportation to the laboratory. Assuming aerial ULV application at the rate of 3 oz/acre, the theoretical concentration of malathion de- posited was 2.39 mg/ft-. The average concentration found following two aerial sprayings on June 16 and 21-22, 1967, was 1.5 mg/ft-, or 65% of the application rate: the remaining 35% was presumed to have dissi- pated as a non-condensing vapor. This indicates good distribution in the test area. The data are presented in Table 2 and range from a low of 0.28 mg/ft^ to a high of 4.39 mg/ft2. TABLE 2. — Malathion found on filler papers exposed during aerial ULV spraying in Texas, June 1967 Sample No. Malathion (mg/ft=) PLAINVIEW A 1.14 B 1.93 C 1.47 D 0.67 2.35 E 1.33 4.39 F 1.52 1.98 G 1.51 2.42 H 1.31 I 2.09 J 3.66 Overall mean 1.37 2.60 S.D. 0,39 1.06 ABERNATHY K 1.66 1.10 L 1.82 1.32 M 0.86 0.60 N 0.85 0.28 O 0.55 P 0.63 1.19 Q 1.93 1.98 Overall mean 1.29 1.00 S.D. 0.57 0.58 Decomposition of Malathion in Water Malathion levels were lower than expected in water samples collected immediately after sprayings on June 16 and 21-22, 1967. This finding indicated that hydro- lysis had occurred. To reduce malathion hydrolysis during sample shipment, samples taken in July and August were acidified to pH 4 by adding two drops of glacial acetic acid to each 8-oz sampling bottle at the time of collection. The stability of malathion in water was studied further by determining the hydrolysis rate of known samples 16 after storage at various pH values. Aqueous solutions containing 2.25 ppm malathion in Clark and Lubs buffer mixtures at pH values of 2, 4, 6, 7, 8, and 10 were pre- pared. Samples were analyzed immediately and at 1, 2, 5, 9, 13, and 20 days after treatment. Malathion was determined by hexane extraction with a vortex mixer and subsequent analysis on a MicroTek Model 220 gas chromatograph equipped with a phosphorus-specific flame photometric detector. Separation was on 3% OV-17 on 60/80 Chromosorb W, A.W., DMCS-treated, aluminum column, 2' x Va", at 225 C, and a nitrogen carrier flow of 100 ml/ minute. Malathion concentrations found at the various times are shown in Fig. 2. The data show that malathion is stable at pH 2 with essentially 100% recovery at the end of the 20-day test period. At pH 4 and pH 6 after 20 days, the recovery of malathion was 72% and 63%, respectively. The decomposition rate increased further with increasing pH. At pH 7 and 8, only 16% and 11%, respectively, of the malathion remained after 20 days. At pH 10 malathion began to decompose almost immediately, and only 60% remained after 1 hour. FIGURE 2. -Decomposition of malathion in water with time at various pH values pH2-ie5 DAYS-* TIME IN DAYS A logarithmic plot of concentration remaining versus time yields a straight line for each pH as shown, indicat- ing that decomposition is a first-order reaction. Assum- ing decomposition follows the straight-line relationship shown, and interpolating or extrapolating for pH values of 2, 4, 6, 7, 8, and 10, malathion half-lifes found were 185, 45, 26, 8, 6, and 0.1 days respectively. These data are the decomposition rates determined at laboratory temperatures of 23 C and do not necessarily represent field and sample storage temperatures. Field water tem- peratures would be expected to be reasonably constant and to approximate average ambient temperatures; how- ever, the temperature of surfaces might exceed ambient Pesticides Monitoring Journal temperatures in certain instances by 20 C. This tem- perature difference can also occur in confined storage during shipment and can result in a greater decomposi- tion rate. At the pH values for field sites shown in Table 1 (6.8-9.5) malathion has a half-life of from 0.5 to 10 days at 23 C. At the highest pH values found in the field significant changes in concentration occurred within the first hour after application. Even the field samples with the lowest pH would be expected to undergo ap- preciable decomposition in 1 or 2 days. Data from repetitive samplings of field waters after spraying are consistent with this finding. Results of these tests indicate that acidifying the July and August samples to pH 4 was an effective measure for reducing malathion hydrolysis. Because of the hy- drolysis occurring in the June water samples, these results are not included in this paper. Malathion in Field Waters The concentrations of malathion in the water samples taken immediately after the spraying on July 26-27, 1967, are presented in Table 3. These samples were col- TABLE 3. — Malathion residues in waters in Hale County, Texas, 4 hours after spraying on July 26-27, 1967 Site No. MALAXmON (PPM) PLAINVIEW 1 .11 2 .018 3 .011 4 .026 5 .000 6 .002 7 .007 8 .018 9 .006 10 .010 11 .017 12 .020 13 .005 14 .007 15 .010 16 .010 17 .014 ABERNATHY 18 .061 19 .025 20 .51 21 .025 22 .068 23 .082 24 .12 25 .06 PETERSBURG 26 .000 27 .000 28 .000 29 .000 30 .000 31 .000 32 .000 Vol. 4, No. 1, June 1970 lected approximately 4 hours after spraying. The gen- eral procedure for collecting samples was to locate a pool of water which was completely exposed to the aerial spraying operation, usually a livestock watering tank, intermittent stream, excavation pond, or borrow pit. Each sample was taken in an 8-oz prescription bottle to which 2 drops of glacial acetic acid had been added. Although the water samples contained very little sus- pended matter, they were passed through a plug of pesticide-free glass wool prior to extraction. The samples were analyzed for malathion by extracting 100 ml of water with three consecutive 25-ml portions of purified «-hexane. The /j-hexane extracts were com- bined and concentrated to 5 ml for gas chromatographic analysis. A modified Barber-Colman Series 5000 gas chromatograph, a Hamilton injection block, and an Ionics tritium electron capture detector operating at 20v DC were employed. The separating column was 3' X 1/8 " packed with 5% OV-17 on 100/120 mesh Anachrom SD. Temperatures were 195, 205, and 230 C for the column, detector, and injector, respectively. Nitrogen carrier gas flow was 30 ml/minute with 70 ml/ minute detector purge. Under the conditions em- ployed, the detector was sensitive only to malathion and two BHC isomers. This detector response appeared unusual, but repeated tests of the usual chlorinated hydrocarbons such as DDT, DDE, heptachlor, hepta- chlor epoxide, aldrin, and dieldrin, gave no response. This is the specific Lovelock design (3) of the electron capture detector. It was operated, however, with nitro- gen carrier in the DC mode rather than with argon- methane carrier and pulsed electrical input. The minimum detectable quantity of malathion was 0.0005 ppm. The hexane extraction method, as used for recovery of malathion, was verified by extracting known additions of malathion from water under the same conditions. Only four samples contained malathion concentrations high enough (.082, .11, .12, and .51 ppm) for possible kill of mosquito larvae. The highest concentration (.51 ppm) was in a sample from a livestock watering tank in southeastern Abernathy. Samples collected in Peters- burg, the control area, had no residues. Water samples after the spraying of Plainview on August 24, 1967, were collected along with control samples from Petersburg at 4-, 24-, and 48-hour intervals. Results of analyses of these samples are tabulated in Table 4. These samples were analyzed on a MicroTek gas chrom- atograph. Model MT-220, employing a phosphorus- sensitive flame photometric detector. The separating column consisted of 2.5% E301 plus 0.25% Epon 1001 on Chromosorb W, 100/120 mesh, 4' x ^/ie" O.D. at a nitrogen flow of 100 ml/minute, at 160 C. The detection limit with the Flame photometric detector was 0.001 ppm. 17 TABLE 4. — Malathion residues in waters in Hale County, Texas, after August 24, 1967, spraying Site Malathion (ppm) No. 4 HOURS 24 HOUItS 48 HOURS Plainview 2 0.15 <0.00I <0.001 5 0.015 < 0.001 <0.001 7 0.021 0.006 0.001 8 0.50 <0.001 <0.001 11 < 0.001 13 0.021 0.001 <0.001 14 0.037 <0.001 <0.001 16 0.029 <0.001 I 0.059 0.027 <0.001 II 0.082 0.013 0.006 Paint carton 0.233 Petersburg 28 <0.001 <0.001 < 0.001 30 < 0.001 <0.001 < 0.001 31 0.001 <0.001 0.001 Samples designated by Roman numerals (Table 4) were taken from automobile tires located in southern Plainview. The tires, with water added, had been placed at these sites for sampling to determine the effectiveness of the spray treatment upon such prime mosquito- breeding sites. The sample containing the highest con- centration (0.50 ppm) was collected from an excava- tion pond 4 hours after spraying. This concentration, identical to the maximum found 4 hours after the July spraying, decreased to less than 0.001 ppm after 24 hours. The maximum concentration found after 24 hours (0.027 ppm) was from the No. I tire sample, and this amount decreased to less than 0.001 ppm after 48 hours. The maximum concentration found after 48 hours was 0.006 ppm and occurred in the No. II tire sample. Although different gas chromatographs and conditions have been used for the analysis of malathion as reported in Tables 2, 3 and 4, the results are directly comparable. Chlorinated Pesticides in Field Waters As part of the overall investigation, water samples were also analyzed for chlorinated pesticides. The average amounts of pesticides found in the samples collected on June 12, 16, and 21-22 are tabulated in Table 5: analyses of August samples are presented in Table 6. One hundred milliliters of each sample was extracted three successive times with 25-ml portions of petroleum ether. The extracts were combined and concentrated to 5-ml volume for analysis by gas chromatography. The following equipment and conditions were used: A Mi- croTek Model 2500R gas chromatograph with s. 5'7( QF-I column 6' x '4" O.D. on 90/100 mesh Anakrom ABS (Analabs, Inc., Hamden, Conn.) at 181 C and 100 ml/ minute nitrogen flow with a tritium electron capture detector in the DC mode; a modified Barber-Colman TABLE 5. — Chlorinated pesticides in waters in Hale County, Texas, June 1967 Residues in PPB > Site BHC Hepta- No. P,P' DDT o,p' DDT P.P' DDE o.p' DDE P.P' DDD DlEL- DRIN Al- DRIN Hepta- CHLOR CHLOR EPOXIDE c ;3 7 A 1 1. .3 .4 .2 .3 .3 .4 .4 2 1. 1. 1. .8 .7 .03 .2 .2 .02 .2 .09 3 .2 .2 .2 5 .2 .2 .1 .3 6 .1 .1 .1 .4 .1 .1 .02 7 .2 .1 1. .5 .5 .4 8 1. .2 .9 .8 1. .6 .4 .4 .05 9 .1 .1 2. .1 .8 .7 .3 .1 .4 10 .3 .2 .5 .1 .4 .05 .06 11 1. 1. .3 .6 .5 .02 .9 .2 12 1. 3. .5 .5 .3 .5 .2 .02 .2 13 .6 .3 .7 .3 .7 .6 .5 .2 .5 14 .1 .03 .02 .2 .1 I .1 15 1. .6 5. .4 .3 .3 .2 .2 .2 16 .4 .2 .01 .1 .4 .2 .1 .1 17 3. 1. 1. .9 .5 .2 .2 .2 .4 18 .6 1. .1 .1 .6 .2 3 .2 19 .02 .01 .6 .4 .3 .1 20 .5 .5 1. .8 .07 .6 .09 .07 .02 .4 21 .2 .1 .3 .3 .6 .5 .1 2. .5 22 .2 .3 .4 .3 .2 .3 .1 .2 .1 23 .3 2. 1. 1. 2. .4 .3 .2 .3 .02 .2 .06 24 .4 .4 .3 .3 3. .3 1. .02 .2 25 1. .7 .6 .6 .6 .2 .2 .1 26 .1 .1 .6 .3 .4 27 .04 .02 .08 .02 .6 .2 2 .1 .02 28 .5 .2 .2 29 .01 .01 .01 .6 ,2 .02 30 1. .6 .1 .3 .5 .1 .2 .02 31 1. .3 3. .4 .1 .9 .6 .02 32 .4 .2 2. 1. ,1 .6 .2 .4 I .1 .05 .09 .01 1. .7 .1 1.5 .3 .3 .3 II .3 .2 .2 .7 .6 .8 .1 .1 ' Averaged results of samples taken on June 12, 16, and 21-22, 1967. 18 Pesticides Monitoring Journal TABLE 6. — Chlorinated pesticides found in water. Hale County, Texas, August 1967 Residues in PPB i Site BHC No. P.P' DDT o,p' DDT P.P' DDE o.p' DDE P.P' DDD DlEL- DRIN Hepta- a 3 7 \ CHLOR I A .2 .1 .1 ,1 .4 .1 .1 2 .06 .2 .02 .02 .5 .3 .1 01 .4 5 .1 .2 .06 .06 .5 .05 .4 7 .2 .3 .1 .1 .2 .2 .3 .1 .1 8 .7 .2 3. Present 2 .3 .1 .2 13 .1 .1 .2 .4 Present .5 14 .1 .04 .04 .04 .4 2 .1 .1 .02 16 .1 .1 .05 .05 Present .05 .09 .03 28 .04 ,06 .03 .1 30 .3 31 .4 .3 .3 .6 .02 1 Averaged results of samples taken at 4-. 24-. and 48-hour intervals. Series 5000 gas chromatograph with a Hamilton injec- tion port, an Ionics high-temperature nickel-63 electron capture detector, an Infotronics Model EA-1 electro- meter, and a MicroTek Model No. 630410 variable pulse electron capture power supply. Separation was obtained on a 4' X 4-mm I. D. glass column of 3% GC grade GE SE-30 on 100/120 mesh Chromosorb W, DMCS acid washed (Applied Science Laboratories, State Col- lege, Pa.), at 190 C and 60 ml/minute argon-methane carrier. The detector was operated in the pulsed mode at 30v, 1 microsecond pulse every 100 microseconds. The limiting sensitivity varies somewhat for the different compounds and was approximately 0.01 ppb for p.p'- DDT. The extraction procedure was checked by use of water spiked with known quantities of DDT isomers. The primary reason for using the two chromatographs with different columns, detectors, and operating condi- tions was to confirm the pesticides present; each result reported is the average of four determinations. There were few unidentified chromatographic peaks. Many samples contained an appreciable amount of elemental sulfur which interfered with the aldrin determination on the SE-30 column. The identification of pesticides pres- ent was based on relative retention times. Reference standards were prepared of the persistent chlorinated pesticides. In addition, the pesitcides used by local municipal agencies plus the crop protection pesticides applied by aerial applicators were included. The re- tention times and peak heights of these standards were used to determine the pesticides present in the field sam- ples. All water samples contained residues of DDT and BHC and related isomers. The concentrations found in one-third of the samples were less than 1 ppb. Summary The malathion application rate by aerial ultra-low vol- ume (ULV) spraying can be reliably measured by a chemical method involving the exposure of filter paper panels for subsequent extraction with hexane and anal- ysis of the recovered malathion by gas chromatography. Vol. 4, No. 1, June 1970 Malathion is subject to hydrolysis in water. The rate is pH-dependent, and field water samples must be acidi- fied to reduce decomposition during transport to the laboratory. The concentration half-life varies from 8 days in neutral solution to 2' 2 hours in alkaline solution at pH 10. The maximum malathion concentration found in the environmental waters of Hale County, Texas, after field ULV spraying was 0.5 ppm, and complete decomposition occurred in 1 day. Chemical data indicate that ULV spraying of 3 oz/acre of malathion produces negligible malathion contamination of the area waters. The chlorinated pesticides present in the waters of the environment of Hale County, Texas, during ULV mala- thion spraying has been determined. All waters con- tained residues of DDT and BHC and related isomers. The concentraitons found in one-third of the samples were less than 1 ppb. A cknowledgments This project was made possible through cooperative ef- forts with the Arboviral Disease Section, National Com- municable Disease Center; U. S. Air Force: Texas State Department of Health; and Plainview-Hale County Health Department. Staff of the Technical Development Laboratories — Messrs. J. W. Kilpatrick, W. E. Dale, James D. Jones, E. P. Hill, and Dr. M. T. Shafik — assisted with the plan- ning and in the analysis and interpretation of data; Messrs. J. W. Pickett and L. N. Thomas assisted in the field sampling. The use of trade names and commercial sources is for identification purposes only and does not constitute endorsement by the Public Health Service or by the U. S. Department of Health. Education, and Welfare. See Appendix for che paper. al names of compounds mentioned in this 19 LITERATURE CITED of gases and vapors. Anal. Chem. 33:162-178. {4} Mitchell. C. J.. R. O. Hayes, Preston Holden, H. R. Hill. ,,,„., . , , ,,, ,„,, „ _, -c .• r ""d T. B. Hughes, Jr. 1969. Effects of ultra-low volume (Ij Kilpatrick, J. W. 1967. Performance specifications for .... t „ i .u- • u i /- . t t ... , . , .... r ■ .■ J r applications of malathion in Hale County, Texas. I. ultra-low volume aerial application of insecticides for ,,, . , ,..• .•■.•.. j j . 1 T. . ^ . -!£• /». N on o.< Western encephalitis virus activity m treated and un- mosquito control. Pest Contr. 35 (May):80-84. ,^^^,^j ^^^^^ j ^^^ g^, 6(2):155-162. {2) Kilpatrick, J. W. and C. T. Adams. 1967. Emergency ^jj Mitchell, C. J.. J. W. Kilpatrick, R. O. Hayes, and H. W. measures employed in the control of St. Louis encepha- Currv. 1970. Effects of ultra-low volume applications of litis epidemics in Dallas and Corpus Christi, Texas, 1966. malathion in Hale County, Texas. 11. Mosquito popula- Calif. Mosq. Contr. Ass. Proc. 35:53. tions in treated and untreated areas. J. Med. Ent. 7(1): (3) Lovelock, J. E. 1961. Ionization methods for the analysis 85-91. 20 Pesticides Monitoring Journal PESTICIDES IN SOIL DDT Moratorium in Arizona — Agricultural Residues After 1 Year ^ G. W. Ware, B. J. Estesen, C. D. Jahn, and W, P. Cahill ABSTRACT Generally, the DDT moratorium in Arizona, begun in 1969, l>as been only moderately successful during its first year because of some apparent agricultural use of DDT in the three major irrigated areas. Alfalfa residues declined slightly, while beef fat residues remained the same. Both irrigated and desert soil residues also remained relatively stable .slwwing little decline. Introduction Arizona's 1-year moratorium on the agricultural use of DDT was established by the Board of Pesticide Control in January 1969. At the same time they requested that the residue laboratory of the Department of Entomology, University of Arizona, monitor the primary irrigated areas for changes in residues of DDT and its related de- gradation products (DDTR). The reasons for such monitoring were ( 1 ) to determine whether violations oc- curred and (2) to measure the decline of environ- mental DDT residues following 20 consecutive years of agricultural use. Because of our previous experience (/) we chose to sample growing alfalfa, as well as the upper 10 inches of soil in these alfalfa fields, on a quarterly basis and also beef fat from selected feed lots following the cyclic changes in beef feeding practices. Sampling Methods Alfalfa and soil samples were collected from the three major irrigated areas — the Salt River Valley around ^ From the Department of Entomology, The University of Arizona, Tucson, Ariz. 85721. Vol. 4, No, 1, June 1970 Phoenix, Pinal County, and the Yuma mesa and valley. Desert soil samples were also collected adjacent to and northwest and southeast of the three irrigated areas. Ten alfalfa fields and four desert soils were sampled from each of the three areas in January, May-June, Septem- ber, and December of 1969. In addition, an earlier study (/) was continued on the 60-mile Maricopa County east-west transect known as Baseline Road, providing a reference standard and continuity to the moratorium monitoring back to 1966. Alfalfa samples were hand-harvested 2 inches above the soil surface. Each sample consisted of 20 subsamples varying between 0.5 and 1 .0 ft^ in area, totaling approxi- mately 3 lb of green alfalfa. Each soil sample consisted of twenty 10-inch deep plugs, 1 inch in diameter, taken randomly with a standard soil sampler. Beef fat samples, 50 to 100 g in size, were removed with a scalpel from the left kidney of carcasses after chilling 24 hours in the slaughterhouse. A nalytical Methods Green alfalfa samples were extracted with solvent as previously described (/). Screened soil samples con- sisting of 25-g aliquots were extracted in a Soxhlet ap- paratus for 16 hours with an azeotropic solvent, hexane and acetone (14:59). After filtering, the extracts were washed, dried through sodium sulfate, and refrigerated. A rapid, on-column extraction cleanup method for an- imal fat (2) was used for the beef fat samples. This method handles a maximum of 0.8 g of fat tissue in the single-step procedure and permits a large number of samples to be processed in a day. 21 An aliquot of alfalfa or soil extract was cleaned on a 4-inch colum of activated Florisil topped with '4 inch of sodium sulfate. The extract was eluted with 250 ml of 6'^'r diethyl ether in hexane, reduced in volume by evap- oration, and adjusted to 5 ml in a glass-stoppered centri- fuge tube for analysis by electron capture gas-liquid chromatography. Recovery standards and analytical reagent blanks were carried through the extraction and cleanup procedures for each day's analyses. Recoveries were consistently 90 to 100%; however, these correc- tions were not applied to the data presented. The mini- mum sensitivity of the analytical method was arbitrarily set at 0.02 ng for p.p'- and o.^'-DDT and DDE. The relative sensitivities were .001 ppm for alfalfa, .003 ppm for soil, and .060 ppm for beef fat, based on a minimum sample size and 6 lA extract injected into the chromato- graph. Analytical confirmatory tests were conducted on a random basis. Because of interfering peaks from toxa- phene encountered in the September and December alfalfa samplings, in the confirmatory tests all extracts were dehydrohalogenated after Florisil cleanup and measured only as o,p'- and p.p'-DDE according to the methods described by Cahill et al. (5). Results and Discussion The analytical results of alfalfa, soil, and beef fat sam- plings during the moratorium are presented in Tables 1-8, as total DDTR. Table 1, showing the continued Baseline Road study indicates a general reduction in alfalfa residues each year, with the exception of residues found in fields 2 and 11 in December 1969. The higher DDTR levels in this sampling indicate a probable agri- cultural use of DDT in that vicinity subsequent to Sep- tember 8, the date of previous sampling. Data on alfalfa residues from Maricopa County (Table 2) show two fields with high levels in December. These results are similar to those for Baseline Road, which is also in Maricopa County. In Pinal County, alfalfa from three fields sampled in September and one sampled in December had high residues (Table 3). In Yuma County, alfalfa from five fields sampled in September and one in December had high residues (Table 4). It seems apparent from these data that DDT was used for agricultural purposes in both Pinal and Yuma Counties prior to the September 6 sampling. TABLE 1. — DDTR residues in green alfalfa along Baseline Road, Maricopa Coiinly, Arizona 1Q67-69 Field Residues in PPM No. 1967 1968 1969 Mar. May Aug. Nov. Feb. Sept. Dec. Apr. Sept. Dec. 2 .055 .018 .305 .094 .220 .045 .016 .038 .127 3 .019 .048 .283 — — .052 .020 .027 .025 4 .052 .021 .170 .165 .070 .120 .064 .027 .038 .051 5 .039 .025 .077 — .060 .044 — .020 .091 6 .019 .016 .277 .116 — — — .023 .035 .060 8 .039 .028 .794 .455 — — .161 — — — 9 .045 .031 — .076 — .017 .034 .055 10 .063 .032 .350 .187 .110 .092 — .029 .054 .095 11 .043 .017 .453 .283 .176 .580 .091 .029 .064 .189 12 .011 .012 .299 .114 — .077 .042 .005 .025 .055 13 .072 .023 .606 — — — .065 — — — Average .042 .025 .404 .213 .113 .175 .068 .021 .037 .083 TABLE 2. — DDTR residues in green alfalfa during 1969 DDT moratorium, Maricopa County, Arizona TABLE 3. — DDTR residues in green alfalfa during 1969 DDT moratorium, Pinal County, Arizona Field No. Jan. May Sept. Dec. 1 .087 .021 .042 .075 2 .303 .025 .062 .049 3 .102 .021 .078 .059 4 .107 .020 .047 .043 5 .049 .012 .030 .067 6 .113 .027 .064 .122 7 .082 .033 .034 .052 8 .125 .084 .056 — 9 .085 .029 .044 .073 10 — .011 - .123 Average .117 .028 .051 .074 Residues in PPM Field No. Jan. May Sept. Dec. .047 .006 .042 .047 .047 .011 .031 .063 .142 .012 .187 — .231 .051 .076 .095 .092 .006 .130 .024 6 .038 .004 .058 .012 7 .079 .011 .118 .052 8 .068 .007 .071 .029 9 .054 .014 .068 .027 10 .078 .011 .077 .171 Average .088 .013 .086 .058 22 Pesticides Monitoring Journal In Maricopa County, residues in soils from both alfalfa fields and desert sites showed a relatively stable picture, with the exception of fields 6 and 8 (Table 5). In Pinal and Yuma Counties also, residues in these soils remained constant throughout the year (Tables 6 and 7). The stability of these DDTR soil residues was unexpected since our soil insecticide field plots indicate a residual half-life of approximately 2 years for DDT under irri- gated cultivation conditions and slightly longer for desert plots. DDTR residues in beef fat. shown in Table 8. indicate no change from the pre-moratorium levels. The wide variation in residues between feed lots is a reflection of the feed types and geographic feed sources in Arizona. TABLE 4. — DDTR residues in green alfalfa during 1960 DDT nwratorium. Yiinui County, Arizona TABLE 6.— DDTR residues in soils during 1969 DDT moratorium, Pinal County, Arizona Field Residues in PPM No. Jan. June Sept. Dec. 1 .047 .012 .373 .048 2 .039 .010 .098 .043 3 .049 .014 .256 .079 4 .057 .009 .093 .041 5 .057 .017 .545 .095 6 .044 .009 .317 .049 7 .059 .016 .241 .106 8 .036 .002 .045 .032 9 .021 .003 .056 .016 10 .046 .015 .074 .054 Average .046 .011 .210 .056 TABLE 5.— DDTR residues in soils during 1969 DDT moratorium, Maricopa County, Arizona Alfalfa Residues in PPM Field No. Jan. May Sept. Dec. 1 0.54 0.34 0.57 0.53 2 1.54 2.22 1.88 2.04 3 0.59 I 1.33 1.60 1.81 4 0.74 0.71 0.68 0.64 5 0.44 0.23 0.35 0.26 6 3.92 3.70 3.41 5.05 7 1.22 1.22 1.22 1.43 8 4.08 4.34 14.54 5.21 9 2.41 2.14 2.32 2.35 10 1 .34 0.48 0.26 0.30 Average 1.58 1.67 1.68 1.96 Desert Site No. 1 0.13 0.07 0.41 0.39 2 0.35 0.18 0.60 0.26 3 0.67 0.16 0.11 0.21 4 12.51 1.14 3.50 2.90 Average 0.92 0.39 1.16 0.94 ^ No sample — this figure represents the mean of the values shown for samples collected in the other 3 months. Vol. 4, No. I.June 1970 Alfalfa Field Residues in PPM No. Jan. May Sept. Dec. 1 4.64 2.79 4.08 3.47 2 1.48 1.66 1.79 2.07 3 2.89 12.89 3.06 2.72 4 2.35 2.75 2.12 2.30 5 0.33 0.74 0.50 0.54 6 0.12 0.13 0.14 0.11 7 2.68 2.17 2.75 2.62 8 0.14 0.13 0.17 0.12 9 1.06 0.90 1.07 1.21 10 1.16 1.46 1.28 1.42 Average 1.69 1.56 1.70 1.66 Desert Site No. 1 0.15 0.05 0.14 0.14 2 0.32 0.10 0.42 0.18 3 0.05 0.06 0.14 0.14 4 1.06 0.67 1.14 1.26 Average 0.40 0.22 0.46 0.43 No sample — this figure represents the me samples collected in the other 3 months. of the values shown tor TABLE 7. -DDTR residues in soils during 1969 DDT noratorium, Yuma County, Arizona Alfalfa Field No. Residues in PPM Jan. June Sept. Dec. 1 0.16 0.17 0.16 0.10 2 0.60 0.62 0.64 0.64 3 1.72 1.79 1.52 1.67 4 1.25 1.37 1.42 1.19 5 0.91 1.09 0.67 1.20 6 1.29 1.13 1.28 1.11 7 1.85 1.47 1.51 1.65 8 0.07 0.07 0.06 0.08 9 0.00 0.02 0.01 0.00 10 0.31 0.24 0.23 0.21 Average 0.82 0.80 0.75 0.79 Desert Site No. J 0.38 0.27 0.30 0.27 2 0.07 0.07 0.03 0.05 3 0.06 0.04 0.03 0.04 4 0.01 0.02 0.00 0.02 Average 0.13 0.10 0.09 0.10 TABLE 8. -DDTR residues in beef fat from selected Ari- zona feed lots Feed Lot Residues in PPM No.i Nov. 1968 May 1969 Dec. 1969 1 2 6 9 12 1.34 1.07 1.15 0.80 0.47 0.74 1.27 0.77 0.14 0.48 0.59 1.93 0.75 0.71 0.84 Average 0.97 0.68 0.96 Average of five animals per feed lot. 23 Despite the apparent agricultural use of DDT in all three major irrigated areas, the DDT moratorium would be considered moderately successful. It appears that the most frequent use of DDT was in Yuma County, fol- lowed by Pinal and Maricopa Counties. Alfalfa fields and adjacent desert soil residues are higher in Maricopa and Pinal Counties than Yuma County by a factor of two. Residues in green alfalfa from three counties were generally lowered by the moratorium and in the same order of magnitude; residues in beef fat remained at the pre-moratorium levels. See Appendi: paper. tor chemical names of compounds mentioned in this This study is a contribution to Regional Project W-45, "Residues of Selected Pesticides— Their Nature. Distribution, and Persistence in Plants, Animals and the Physical Environment." University of Ari- zona Agricultural Experiment Station journal series No. 1601. LITERATURE CITED (1) Ware. G. W., B. J. Eslesen ami W. P. Cahill. 1968. Pesticides in soil — an ecological study of DDT residues in Arizona soils and alfalfa. Pesticides Monit. J 2(3)- 129-132. (2) Cahill, W. P., B. ]. Eslesen and G. W. Ware. 1970. A rapid on-column extraction-cleanup method for animal fat. Bull. Environ. Contamination Toxicol, (in press). (3) Caliill. W. P., B. J. Eslesen and G. W. Ware. 1970. De- termination of DDT in the presence of toxaphene resi- dues. Bull. Environ. Contamination Toxicol, (in press). 24 Pesticides Monitoring Journal APPENDIX Chemical Names of Compounds Mentioned in This Issue ALDRIN BHC COPPER SULFATE DDD (TDE) (including its isomers and dehydrochlori- nation products) DDT (including its isomers and dehydrochlorination products) DIELDRIN AZINPHOSMETHYL HEPTACHLOR HEPTACHLOR EPOXIDE MALATHION METHYL PARATHION PARATHION Not less than 95% of 1.2,3,4,10,10-hexachloro-1.4,4a.5,8,8a-hexahydro-l,4-cHdo-exo-5,8-dimethanonaphthalene 1,2,3,4,5,6-hexachlorocyclohexane, mixed isomers CuSO, • H„0 l,l-dichloro-2,2-bis(p-chlorophenyl)ethane; technical TDE contains some o,p'-isomer also 1 , 1 -dichloro-2,2-bis ( p-chlorophenyl ) ethylene l,l.l-trichloro-2,2-bis(p-chlorophenyl)eth3ne; technical DDT consists of a mixture of the p,p'-isomer and the o.p'-isomer (in a ratio of about 3 or 4 to 1 ) Not less than 85% of l,2,3.4.10,10-hexachloro(6,7-epoxy-l,4.4a.5.6,7,8,8a-octahydro-l,4-cndo-«o-5,8-dimethano- naphthalene 0,0-dimethyl 5(4-oxo-l ,2,3-benzotr]azin-3 ( 4H i -> Imethyl ) phosphorodithionate 1, 4,5,6,7, 8,8-heptachloro-3a,4,7,7a-tetrahydro-4,7-methanoindene l,4,5,6,7,8,8-heptachloro-2,3-epoxy.3a,4,7,7a-tetrahydro-4,7-methanoindan diethyl mercaptosuccinate, 5-ester with 0.0-dimethyl phosphorodiihioate 0,0-dimethyl O-p-nitrophenyl phosphorothioate 0,0-diethyl O-p-nitrophenyl phosphorothioate Vol. 4, No. I.June 1970 25 Information for Contributors The Pesticides Monitoring Journal welcomes from all sources qualified data and interpretive information which contribute to the understanding and evaluation of pesticides and their residues in relation to man and his environment. The publication is distributed principally to scientists and technicians associated with pesticide monitoring, research, and other programs concerned with the fate of pesticides following their application. Additional circulation is maintained for persons with related in- terests, notably those in the agricultural, chemical manu- facturing, and food processing industries; medical and public health workers; and conservationists. Authors are responsible for the accuracy and validity of their data and interpretations, including tables, charts, and refer- ences. Accuracy, reliability, and limitations of the sampling and analytical methods employed must be clearly demonstrated through the use of appropriate procedures, such as recovery experiments at appropriate levels, confirmatory tests, internal standards, and inter- laboratory checks. The procedure employed should be referenced or outlined in brief form, and crucial points or modifications should be noted. Check or control samples should be employed where possible, and the sensitivity of the method should be given, particularly when very low levels of pesticides are being reported. Specific note should be made regarding correction of data for percent recoveries. Preparation of manuscripts should be in con- formance to the Style Manual for Biological Journals, American Institute of Biological Sciences, Washington, D. C, and/or the Style Manual of the United States Government Print- ing Office. An abstract Cnot to exceed 200 words) should accompany each manuscript submitted. All material should be submitted in duplicate (original and one carbon) and sent by first-class mail in flat form — not folded or rolled. Manuscripts should be typed on 8V2 x 11 inch paper with generous margins on all sides, and each page should end with a completed para- graph. All copy, including tables and references, should be double spaced, and all pages should be num- 26 bered. The first page of the manuscript must contain authors' full names listed under the title, with affiliations, and addresses footnoted below. Charts, illustrations, and tables, properly titled, should be appended at the end of the article with a notation in text to show where they should be inserted. Charts should be drawn so the numbers and texts will be legible when considerably reduced for publication. All drawings should be done in black ink on plain white paper. Photographs should be made on glossy paper. Details .should be clear, but size is not important. The "number system" should be used for litera- ture citations in the text. List references alpha- betically, giving name of author/s/, year, full title of article, exact name of periodical, volume, and inclusive pages. Pesticides ordinarily should be identified by common or generic names approved by national scientific so- cieties. The first reference to a particular pesticide should be followed by the chemical or scientific name in parentheses — assigned in accordance with Chemical Abstracts nomenclature. Structural chemical formulas should he used when appropriate. Published data and information require prior approval by the Editorial Advisory Board; however, endorsement of published in- formation by any specific Federal agency is not intended or to be implied. Authors of accepted manuscripts will receive edited typescripts for approval before type is set. After publication, senior authors will be provided with 100 reprints. Manuscripts are received and reviewed with the under- standing that they previously have not been accepted for technical publication elsewhere. If a paper has been given or is intended for presentation at a meeting, or if a significant portion of its contents has been published or submitted for publication elsewhere, notation of such should be provided. Correspondence on editorial matters or circulation mat- ters relating to official subscriptions should be addressed to: Mrs. Sylvia P. O'Rear, Editorial Manager, Pesti- cides Monitoring Journal. Division of Pesticide Com- munity Studies, Food and Drug Administration. 4770 Buford Highway, Bldg. 29, Chamblee, Ga. 30341. Pesticides Monitoring Journal The Pesticides Monitoring Journal is published quarterly under the auspices of the WORKING GROUP, Subcommittee on Pesticides, President's Cabinet Committee on the Environment, and its Panel on Pesticide Monitoring as a source of information on pesticide levels relative to man and his environment. The WORKING GROUP is comprised of representatives of the U. S. Department of Agricul- ture; Defense; the Interior; Health, Education, and Welfare; State; and Transportation. The Pesticide Monitoring Panel consists of representatives of the Agricultural Research Service, Consumer and Marketing Service, Federal Extension Service, Forest Service, Department of Defense, Fish and Wildlife Service, Geological Survey, Federal Water Quality Administration, Food and Drug Administration, Environmental Health Service, Department of Defense, National Science Foundation, and Tennessee Valley Authority. Publication of the Pesticides Monitoring Journal is carried out by the Division of Pesticide Community Studies of the Food and Drug Administration. Pesticide monitoring activities of the Federal Government, particularly in those agencies repre- sented on the Pesticide Monitoring Panel which participate in operation of the national pesticides monitoring network, are expected to be principal sources of data and interpretive articles. How- ever, pertinent data in summarized form, together with interpretive discussions, are invited from both Federal and non-Federal sources, including those associated with State and community monitoring programs, universities, hospitals, and nongovernmental research institutions, both domestic and foreign. Results of studies in which monitoring data play a major or minor role or serve as support for research investigation also are welcome; however, the Journal is not intended as a primary medium for the publication of basic research. Manuscripts received for publication are reviewed by an Editorial Advisory Board established by the Monitoring Panel. Authors are given the benefit of review comments prior to publication. Editorial Advisory Board members are: Reo E. Duggan, Food and Drug Administration. Chairman Anne R. Yobs, Food and Drug Administration Andrew W. Briedenbach, Environmental Health Service Thomas W. Duke, Fish and Wildlife Service William F, Stickel, Fish and Wildlife Service Milton S. Schechter, Agricultural Research Service Paul F. Sand, Agricultural Research Service Mention of trade names or commercial sources in the Pesticides Monitoring Journal is for identification only and does not represent endorsement by any Federal agency. Address correspondence to: Mrs. Sylvia P. O'Rear Editorial Manager PESTICIDES MONITORING lOURNAL Food and Drug Administration 4770 Buford Highway, Bldg. 29 Chamblee, Georgia 30341 CONTENTS Volume 4 September 1970 Number 2 RESIDUES IN FOOD AND FEED Page Toxaphene and DDT residues in ladino clover seed screenings 27 T. E. Archer Chlorinated hydrocarbon residues in the milk supply of Ontario, Canada 3 1 R. Frank, H. E. Braun, and J. W. McWade Residues of chlorinated hydrocarbons in soybean seed and surface soils from selected counties of South Carolina 42 W. R. McCaskill, B. H. Phillips, Jr., and C. A. Thomas PESTICIDES IN PEOPLE Serum organochlorine pesticide levels in people in southern Idaho 47 Michael Watson, W. W. Benson, and Joe Gabica RESIDUES IN FISH, WILDLIFE, AND ESTUARIES Significance of DDT residues in fishes from the estuary near Pensacola, Fla 51 David J. Hansen and Alfred J. Wilson, Jr. Chlorinated hydrocarbon pesticides in representative fishes of southern Arizona 57 Donald W. Johnson and Sam Lew Pesticide residues in channel catfish from Nebraska 62 N. P. Stucky PESTICIDES IN SOIL "Apparent" organochlorine insecticide contents of soils sampled in 1910 67 B. E. Frazier, G. Chesters, and G. B. Lee PESTICIDES IN WATER Pesticides in surface waters of the United States — a 5-year summary, 1964-1 James J. Lichtenberg, James W. Eichelberger, Ronald C. Dressman, and James E. Longbottom APPENDIX Chemical names of compounds mentioned in this issue 87 RESIDUES IN FOOD AND FEED Toxaphene and DDT Residues in Ladino Clover Seed Screenings^ T. E. Archer ABSTRACT Ladino clover seed crop screenings field-contaminated with DDT and its analogs and toxaphene were analyzed for these compounds in a composite sample and in 13 separate frac- tions of the composite sample. The total residue for DDT and its analogs was 20.0 ppm and 21.1 ppm for toxaphene. The residues were excessively high for use of the plant mate- rial as animal feed. Two fractions, representing 29% of the composite sample, contained 74% of the DDT and 70% of the toxaphene. These fractions were ladino clover chaff which was 19% of the composite and contained 9.1 ppm DDT and 9.4 ppm toxaphene and soil which was 10% of the composite and contained 5.8 ppm DDT and 5.4 ppm toxaphene. Introduction Large quantities of pesticides have been applied to ladino clover and alfalfa seed crops during their pro- duction. The seed-cleaning processes after harvest re- sult in such by-products as straw, chaff, weed, and other seeds which constitute a source of animal feed. These have been used either separately or in feedstuff mixtures although recommendations are usually against such use. The quantitative measurement of combinations of Ara- mite, DDT, toxaphene, and endrin has previously been reported in specific animal feeds (i). Various workers ' From the Department o( Environmental Toxicology, University of California, Davis, Calif. 95616. Vol. 4, No. 2, September 1970 have studied the analytical characteristics of toxaphene after treatment with sulfuric-fuming nitric acid mixtures during residue determinations (6,7,8). The decontam- ination of malathion residues from ladino clover seed screenings has also been reported (2). The persistence of methyl parathion residues on sunflower seeds was investigated (5); it was found that if allowable residues were set as low as 0.1 ppm, the minimum interval be- tween methyl parathion treatments of 0.5 and 1.0 lb/ acre and harvest would be 7 and 14 days, respec- tively. The present investigations were undertaken to determine the levels of DDT and its analogs and toxaphene in a composite sample of field-contaminated ladino clover seed crop screenings and also in separate components. If the contaminated components are identified, proce- dures could possibly be developed to eliminate them and thus lower the pesticide levels in the seed screenings to make a more acceptable animal feed. Methods and Materials SAMPLING PROCEDURES The commercial seed screenings (13.0% moisture) were obtained from a ladino clover seed crop which had been treated with DDT and toxaphene for pest control during field growth. An application of 1 lb/ acre DDT plus 2 lb/ acre toxaphene was made on May 28 and another Of IVi lb/ acre DDT plus 3 lb/ acre toxaphene on July 5. 27 SAMPLE FRACTIONATION A 100-g composite sample of the seed screenings was weighed and sieved by shaking over stacked screen sieves ranging from 16 to 100 mesh until the sample was separated into seven crude fractions, including that material which was collected on the bottom pan of the stacked screens. These 7 fractions were further sep- arated with the aid of an illuminated circular magnifier (Luxo Magnifier, Luxo-Lamp Corp., Port Chester, N.Y.) into 13 individual samples for analysis. The fractions were identified and the percent weight of each fraction in relation to the total composite sample was calculated (Table 1). The individual fractions were then extracted and analyzed for DDT and its analogs and toxaphene. SAMPLE EXTRACTION AND CLEANUP In addition to the fractions mentioned above, 10 g of a composite sample and 10-g portions of separated fractions, in triplicate, were individually extracted by refluxing for 30 minutes with 100 ml of solvent con- sisting of 90 ml of benzene, 1 0 ml of ethyl alcohol, and 0.5 ml of 12n hydrochloric acid. The reflux extraction was performed 3 times, and the solvent was pooled and concentrated in vacuo at 50-60 C prior to cleanup. The solvent extracts were cleaned up on Florisil (acti- vated at 270 C for 3 hours). DDT and its related degradation products (DDTR) and toxaphene were eluted from the Florisil with 390 ml of 30% diethyl ether and 70% pentane, and recoveries exceeded 90%. The extracted plant material was treated with ethanolic potassium hydroxide and analyzed; the residues found were summed with those found in the solvent extracts previously treated with ethanol alkalki (4). DETECTION AND DETERMINATION OF PESTICIDES All chemicals used were reagent grade. The DDTR pesticides were recrystallized analytical standards, and the toxaphene analytical standard was obtained from Hercules Powder Co., Wilmington, Del.; the reagent grade solvents were redistilled shortly before use. Gas- liquid chromatography (GLC) and thin-layer chroma- tography (TLC) procedures were employed routinely, either separately or in combination. No decomposition of the toxaphene applied to the GLC column after treatment with ethanolic potassium hydroxide (3) was evident. No decomposition of the DDT was evident on the GLC column, and its retention time on the GLC and Rf value on the TLC were different from those of either DDD or DDE. The gas chromatograph was the Varian Aerograph Model 1200 equipped with an elec- tron capture detector and a Leeds and Northrup Speed- omax W 1-mv recorder with a chart speed of V2 inch per minute. Areas under the peaks were measured with a polar planimeter. The chromatographic column, an 8 ft. X 4 mm LD. Pyrex glass coil packed with 60/80 28 mesh silylated Chromosorb W, acid washed, was coated with 5% Dow 710 silicone fluid and 5% SE-30 silicone gum rubber. Nitrogen carrier gas (50 psi, 20 cc/ min- ute) and a column temperature of 220 C gave the best results and was used in these experiments. A practical method sensitivity was established at 0.01 ppm. Re- coveries through the entire procedures of extraction, cleanup, and analysis exceeded 90%. All residue data are expressed on a dry-weight basis. Thin-layer chromatography was employed for screening and in combination with GLC as an analytical method. Silica gel H (0.5 mm layer thickness) adsorbent, de- veloping solvent ( 100% pentane), and the silver nitrate: 2-phenoxyethanol color test (9) as a dip solution were employed. For quantitative work, chromatogram areas containing the unknowns were extracted from the silica gel with pentane after comparing Rf values (0.41, DDE; 0.26, DDT; 0.14, DDD; and 0.10 to 0.20, toxaphene) with those of parallel standard pesticide tracers, and the extracts were analyzed by GLC. The toxaphene extracts from TLC were treated with ethanolic potas- sium hydroxide (i) prior to analysis by GLC for the purpose of increasing the analytical method sensitivity and to eliminate interference from DDTR on the GLC. Recoveries were within the limits previously described. Results and Discussion Table 1 shows the levels of DDTR and toxaphene found in the fractionated 100-g sample. Fraction A (Buckhorn) was the largest component (53.4% by weight) and contained relatively low levels of pesticide (0.67 ppm DDTR and 4.26 ppm toxaphene). Fractions B through E comprised approximately 41% of the total composite sample and contained 94% of the DDTR and 85% of the toxaphene. Fraction C (soil, 10.1% by weight) contained 29% of the DDTR and 26% ol the toxaphene residues present in the composite sample. The remaining fractions (F through M) made up ap- proximately 6% of the total composite sample; and although in some of these individual fractions th£ pesticide residues were high, their levels contributec very small amounts to the total levels in the composite sample. Table 2 shows the residue levels of DDD, DDE, anc DDT and their sums found in individual 10-g fraction; after adjusting the levels according to the percentage; by weight of each fraction in relation to a composite sample. Generally, in each fraction, the largest per centage of the DDTR residues was DDT followed b) DDD and DDE. The total residues in fractions A through M were DDD, 1.89 ppm; DDE, 1.26 ppm DDT, 13.60 ppm, with a sum of 20.00 ppm DDTR in eluding 3.20 ppm found as DDTR in the extracted plan material. Pesticides Monitoring Journai TABLE 1. — Levels of DDTR and loxaphene in fraction components of ladino clover seed crop screenings [All residues are reported on a dry-weight basis; ND = not detectable] I^RACTioN Identification Code Letter % OF Total Weight Residues in PPM Buckhorn (Plantago lanceolala) Chall — landino clover Soil Ladino clover seeds Ryegrass (Lolium spp.) Watergrass (Echinochloa crusgalU) Fall dandelion (Leontodon autumnalis) Knotweed (Polygonum ariculare) Naked watergrass (Echinochloa crusgalU) Oxtongue (Picris eschioides) Bermudagrass (Cynodon dactylon) Storkbill (Erodium cicutarium) Miscellaneous — unidentified Weed Legume Legume Grass Grass Weed Weed Grass Weed Grass Weed 53.4 19,3 0.32 0.13 0.04 0.01 0.67 47.08 57.28 47.28 19.42 10.75 15.76 2.94 47.06 2.32 31.36 14.75 29.40 4.26 48.00 53.00 27.40 28.30 9.39 9.90 1.65 70.40 11.80 0.77 1.25 22.50 TABLE 2. — Levels of DDTR in individual 10-g fractions and the composite of these fractions. Levels are adjusted according to the percentage of each plant fraction in a composite sample (Table I). [All residues are reported on a dry-weight basis; ND = not detectable] DDTR IN Sum Code Letteri DDD (PPM) Total Residue DDE (PPM) Total Residue DDT (PPM) Total Residue DDTR (PPM) Extracted Screenings (PPM) Total DDTR (PPM) A ND _ 0.0121 9.5 0.1140 90.5 0.1261 0.2270 0.3531 B 1.1500 15.1 0.8220 10.8 5.6700 74.1 7.6420 1.4400 9.0820 C 0.3150 6.5 0.1670 3.4 4.3600 90.1 4.8420 0.9370 5.7790 D 0.2170 8.8 0.1800 7.3 2.0700 83.9 2.4670 0.3030 2.7700 E ND — 0.0150 1.7 0.8630 98.3 0.8780 0.2020 1.0800 F 0.0566 23.3 0.0200 8.2 0.1670 68.5 0.2436 0.0182 0.2618 G 0.0477 19.3 0.0243 10.0 0.1710 70.7 0.2430 0.0140 0.2570 H 0.0088 5.8 0.0004 2.7 0.0124 91.5 0.0216 0.0052 0.0268 I 0.0810 29.1 0.0123 4.4 0.1850 66.5 0.2783 0.0458 0.3241 J 0.0025 40.9 0.0005 8.6 0.0031 50.5 0.0061 0.0013 0.0074 K 0.0136 36.1 0.0074 19.5 0.0168 44.4 0.0378 0.0029 0.0407 L 0.0007 15.2 0.0019 41.2 0.0017 36.6 0.0043 0.0013 0.0056 M 0.0004 13.5 0.0006 21.2 0.0018 65.3 0.0028 ND 0.0028 Letter codes and sample identification (Table 1). TABLE 3. — Levels of loxaphene in individual 10-g fractions and the composite of these fractions. Levels are adjusted according to the percentage of each plant fraction in a composite sample (Table 1). [All residues are reported on a dry-weight basis; ND = not detectable] Residues in PPM Code TOXAPHENE IN Sum TOXAPHENE Extracted Total Screenings TOXAPHENE A 2.28 ND 2.28 B 9.37 ND 9.37 C 5.36 ND 5.36 D 1.61 ND 1.61 E 1.32 0.26 1.58 F 0.23 ND 0.23 G 0.16 ND 0.16 H 0.01 ND 0.01 I 0.42 ND 0.42 J 0.04 ND 0.04 K ND ND ND L ND ND ND M ND ND ND Total 20.80 0.26 21.06 ' Letter codes and sample identification (Table 1 ) . Vol. 4, No. 2, September 1970 Table 3 shows the residue levels of loxaphene found in individual fractions and adjusted according to the per- centages by weight of each fraction in relation to a composite sample. Approximately 96% of the loxa- phene residues were in fractions A through E with a total adjusted residue on the composite sample basis of 2L06 ppm. Table 4 contains data for comparing the total residues of DDTR and loxaphene found in fractions A through M with the total residues of the pesticides obtained by direct analyses on a composite seed crop screenings sample. The level of DDTR residues obtained in the composite sample by direct analysis was 22.9 ppm as compared to 20.0 ppm in the adjusted total residue for the fractionated sample (Table 2). The loxaphene residue obtained in the composite sample by direct analysis was 20.4 ppm as compared to 21.1 ppm in the adjusted total residue for the fractionated sample (Table 2). 29 TABLE 4. — Comparison of the levels of toxaphene and DDTR in ladino clover seed crop screenings as determined by direct analysis of a composite sample with calculated adjusted levels as determined from analyses on 13 fractions of a composite sample [All residues are reported on a dry-weight basis; ND = not detectable] Residues in PPM Sample DDT DDE ODD DDTR DDTR IN Extracted Seed Screenings Sum Total DDTR Toxa- phene Toxaphene IN Extracted Seed Screenings Sum Total Toxa- phene Composite ladino clover seed screenings Fractionated sample of screenings reported as composite 16.8 13.6 1.6 1.3 2.3 1.9 20.7 16.8 2.2 3.2 22.9 20.0 18.1 20.8 2.3 0.3 20.4 21.1 Percent total residue recovered in fractionated screenings sample with respect to that residue found in the composite sample 87.3% 103.4% If 2 fractions of the 13 (ladino clover chaff and soil) were eliminated from the composite sample, the re- maining 71% of the sample would contain only 26% of the total DDTR and 30% of the toxaphene. The remaining fractions would be much more acceptable as animal feed. However, complete removal of all contaminants from the feed would be ideal, and investi- gations are currently in progress for developing proce- dures for practical removal of all contaminants from the feed by physical and chemical methods. A cknowledgments The technical assistance of Mr. James Stokes of the Department of Environmental Toxicology and the per- sonnel of the Cal Crop Improvement Program of the University of California is gratefully acknowledged. See Appendix for chemical names of compounds mentioned in this paper. LITERATURE CITED (/) Archer, T. E. 1968. Quantitative measurement of com- binations of aramite, DDT, toxaphene. and endrin in crop residues. Bull. Environ. Contamination Toxicol. 3:71. (2) Archer, T. E. 1969. Decontamination of malathion from ladino clover seed screenings. J. Poultry Sci. 48:2075. U) Archer, T. E. and D. G. Crosby. 1966. Gas chromatog- raphy measurement of toxaphene in milk, fat. blood, and alfalfa hay. Bull. Environ. Contamination Toxicol. 1:70. (4) Crosby, D. G. and T. E. Archer. 1966. A rapid ana- lytical method for persistent pesticides in proteinaceous samples. Bull. Environ. Contamination Toxicol. 1:16. (5) Borough, H. W ., N. M. Randolph, and G. H. Wimbish. 1966. Persistence of methyl parathion residues on sun- flower seeds. Bull. Environ. Contamination Toxicol. 1:86. (6) Erro, F., A. Revenue, and H. Reckman. 1967. A method for the determination of toxaphene in the presence of DDT. Bull. Environ. Contamination Toxicol. 2:372. (7) Kawano, Y., A. Revenue, H. Reckman, and F. Erro. 1969. Studies on the effect of sulfuric-fuming nitric acid treatment on the analytical characteristics of toxaphene. J. Ass. Oflfic. Agr. Chem. 52:167. (8) Klein, A. K. and J. D. Link. 1970. Elimination of inter- ferences in the determination of toxaphene residues. J. Ass. Oflfic. Agr. Chem. 53:524. (9) Mitchell, L. C. 1958. Separation and identification of chlorinated organic pesticides by paper chromatography. XI. A study of 114 pesticide chemicals: technical grades produced in 1957 and reference standards. J. Ass. Oflfic. Agr. Chem. 41:781. 30 Pesticides Monitoring Journ.\l Chlorinated Hydrocarbon Residues in the Milk Supply of Ontario, Canada^ R. Frank, H. E. Braun, and J. W. McWade ABSTRACT A province-wide survey of chlorinated hydrocarbon insecti- cide levels in the fluid milk of Ontario was carried out be- tween November 1967 and lune 1969. Composite samples were collected from bulk tankers, each representing an aver- age of 76 producers. With a total collection of 1,651 sam- ples, each of 27,000 producers in the Province was sampled at least once. DDT plus its metabolites and dieldrin were present in almost all milk fat tested; lindane was found in only 8% of the samples and heptachlor epoxide in 3%. Milk from 20 producers contained levels in butterfat which ex- ceeded the administrative tolerances of 0.1 ppm for dieldrin and heptachlor epoxide and 1 ppm for lindane and DDT plus metabolites. Seventeen of these cases were dieldrin vio- lations; three were DDT or DDD; and one was due to lin- dane. These high residues were traced to contaminated feed in seven cases and to improper spraying or use of contami- nated spray in five cases; in four herds, the source of the insecticide could not be found. The disappearance of residues from butterfat, after removal of the source, was an expo- nential function of time. This disappearance was not clearly evident in herds where multiple sources of low-level con- tamination occurred. The average levels of chlorinated in- secticide residues in fluid milk of the Province were 0.134 ppm DDT and its metabolites, 0.031 ppm dieldrin, 0.005 ppm lindane, and 0.001 ppm heptachlor epoxide. Introduction The Canadian province of Ontario ranks first in the Nation with respect to milk production and milk con- sumption. Approximately 6.7 billion pounds of milk are produced per annum with a farm value of about $260 million (i). This represents 930 lb per capita 1 From the Provincial Pesticide Residue Testing Laboratory, Ontario Department of Agriculture and Food, Guelph, Ontario, Canada. Vol. 4, No. 2, September 1970 for the 7.15 million inhabitants (census of 1967). Ex- ports of milk and dairy products amount to approxi- mately $14 million. Milk supplies have been monitored for chlorinated hy- drocarbon insecticides in several countries. Clifford et al. (3.4) and Duggan (5) have reported levels in fluid milk and dairy products in the United States. Bro-Rasmussen et al. (2) have published residue levels in Danish milk and butter. Since no comprehensive data on chlorinated hydro- carbon insecticide levels in milk were available for the province of Ontario, a monitoring survey was initiated to determine residue levels and to attempt to correlate these levels with the particular type of farming opera- tion. Ontario can be divided into southern and northern parts. The southern part, which represents only 13% of the Province, is largely devoted to agriculture; the northern section which is part of the Great Canadian Shield, has only a few small pockets of agriculture around the industrial areas. The great majority of the territory is covered by forests, lakes, rock outcrops, or tundra. From an agricultural point of view, the southern part can be divided into southern, western, central, and eastern regions (Fig. 1), each differing in the type of agriculture practiced (Table 1). The area most in- tensively used for agriculture is in the southern region where high-value crops such as tobacco, corn, soybeans, fruit, and vegetables are produced. Some cash crops are grown close to Lake Huron in the western region, but the mainstay of agricultural production in this region is livestock and, in particular, beef production. The central region has some cash crop farming along Lakes Ontario and Simcoe, but the largest area is devoted to tourism, forestry, and mixed fanning. The eastern 31 FIGURE 1. — Map of the Province of Ontario divided into five regions N^.^ ^^i^ NORTHERN V^ Q vx e. b c c r^^v^^^ -'^*7*"^Ju^''^ \ 0 ^""^x^. ^'N^H.toJl \ — *«i;;^"'- U *JF ^k \ *'c \ ^^3f^ \ ^ X v5^ ^^ \ — y^/ ^ ^ \ V^A EASTERhK \^ ^ \^trr '>3^ £^_.._... 1 Iflo *?^^^S»<^ ^<^^^ \%\ "^ V I/) 4J / i- /Lc.\ ""'■' Jn/'IrnA A /\\ >V\' Vf d / * tJ ^-- y ^-^ !« \ y i\ 1 ' r — ^^ > .ry^"^^ V'^y 'J\ ,^vjw(^^9^? t HURON / 'Cr A,'''°^°"/C'^V^^^ J UrxuLfcci stales T5 /' , / > -j"^ J2l!isv^\/ :" '^ ^ 0 »J 1 /^ ^\o*'°"fd.^V>v*^^*^^-L t 1 '''SOI -^r^^;p\V'^l^^ ^«^ -^ <^_ZL- "T^^m JHOP* jT ^^ *7 ^ D 5 ^js: -' v/ "*^*-^ ^./-'y I'-A ^^^ 1 y •'^*' > .,---' /^ / Dl &TR.CT V ^ \ Ij essE^ \y^^ je.ct'" ^^ '/ NORTHERN ^.-^^ i.^^"" "^^^--^ SOUTHERN ONTARIO A^-^ 1^ f '•-.. ^.- \ ^^_0_^- COCH«*Nt^ I ^ ■*•', -^ A^F J ^4£^ NORTHERN ONTARIO region has a potential for increased cash crop produc- tion, but the main pursuit is dairying and raising Hve- stock. Northern Ontario has isolated pockets of mixed agriculture near the larger population centers. As a consequence of the differing types of agricultural production from area to area and region to region, it is to be expected that differences will be found in the pattern of use of pesticides. In 1968, a survey was car- ried out by the Ontario Department of Health, under the Pesticides Act {13), to discover the use pattern of chlorinated hydrocarbon insecticides in agriculture. The results of this survey are shown in Table 1. Almost 90% of the DDT and cyclodienes used in the Province was applied in the southern region. Unfortunately, the amounts only account for that used in agriculture and do not include the widespread nonagricultural applica- tion for forest and parkland protection and control of nuisance insects. Such uses for DDT have been wide- spread in the central and northern regions. A population of 926,000 dairy cows is distributed over the 5 regions on some 27,000 farms. Milk produced in 32 the southern region is utilized mainly as fluid milk or for butter production. The western region represents the largest butter-producing area, while the central re- gion represents the largest fluid milk-producing area. Milk from the eastern region is used primarily for the production of cheese and butter. Milk produced in northern Ontario is consumed mainly as domestic fluid milk (Table 2). Sampling Procedure The milk monitoring program was initiated by the Ontario Department of Agriculture and Food in No- vember 1967 and completed in June 1969. During this period, a total of 1,651 composite milk samples, repre- sentative of each of the 5 agricultural regions, was collected and analyzed. The survey was intended to include samples from all producers in Ontario who marketed fluid milk. Since the large number of pro- ducers made individual sampling impractical, 1 -quart composite samples, representing 8,000 to 20,000 lb of Pesticides Monitoring Journal TABLE 1. — Type of cropping, distribution of livestock and use of insecticides in the regions of Ontario (1968) SOUTHEKN Western Central Eastern Northern Total Total land area (acres x lO^) Total no. of farms (x 103) 5.2 36.7 6.5 31.7 10.2 17.0 7.3 18.7 191 5.8 220.2 109.9 Crop distribution: PERCENT Acreage (x 10") Cereal Legumes Com Hay Fruit Vegetables 28 85 64 16 76 52 40 14 20 31 11 24 14 0.5 8 17 12 16 14 0.5 8 26 1 5 4 0 0 10 0 3 2.7 0.4 1.3 3.4 0.02 0.11 Livestock distribution: PERCENT Numbers (x 10") CatUe Swine Sheep 22 32 19 40 49 35 15 12 20 18 6 15 5 1 11 3.2 2.0 0.26 Insecticide use: i PERCENT Pounds (X lOS) DDT + DDD Aldrin + Dieldrin 89.3 89.8 5.3 9.2 3.3 1.0 2.1 <0.1 302 32 1 Data obtained by request from the Ontaric^ Department of Health, Toronto. TABLE 2. — Population of dairy cows and dairy herd replacements and the annual production and utilization of milk in Ontario Region of Numbers (x 10=) Quantities produced (lb x W) Ontawo Dairy cows Replacements Butter Cheese Fluid milk Manufactured Southern Western Central Eastern Northern 207 281 138 258 42 58 72 40 74 9 287,894 1,243,163 173,087 546,234 42.675 55,754 81,437 199,237 584,888 26,289 480,570 202,219 801,008 237,637 154,146 110,542 50,046 286,025 92,101 15,878 Ontario (Total) 926 253 2,293,053 947,605 1,875,580 554,592 fluid milk from 5 to 20 producers along a truck route, were collected from bulk milk carriers at the time of delivery at their respective dairies or processors. With this sampling procedure, it was possible to reduce the number of samples required for analysis without sacrificing geographical identity. The 1 -quart milk samples were systematically collected by personnel of the Milk Commission, Ontario Department of Agricul- ture and Food, and delivered promptly to the Provincial Pesticide Residue Testing Laboratory under refrigerated conditions. Analytical Procedure All samples were processed and the butterfat removed immediately upon receipt at the laboratory. The milk was allowed to warm to room temperature, and after a thorough shaking, the butterfat was separated according to the procedure described by Moubry et al. (10). Ap- proximately 100 ml of milk was transferred to a 200-ml volumetric flask and, with constant mixing, the flask was filled to the neck with a detergent reagent consisting of 50 g sodium tetraphosphate plus 24 ml of Triton X-100 per liter of water. This mixture was placed in a water bath at 95 C until the clear butterfat layer separated into the neck of the flask. The butterfat was transferred into vials and held under refrigeration for subsequent analysis. Vol. 4, No. 2, September 1970 Chlorinated hydrocarbon insecticides were isolated from the butterfat by the one-step Florisil column method of Langlois et al. (9) with some minor modifications. Florisil, 60/100 mesh, activated commercially at 1200 F, was reheated at 135 C for a minimum of 24 hours. Upon cooling to room temperature, the adsorbent was partially deactivated by the addition of water at the rate of 5 ml per 100 g Florisil. It was then held in an air-tight container, and allowed to equilibrate during a 12-hour tumbling period. Cleanup was carried out in 25 mm x 300 mm Pyrex columns fitted with Teflon stopcocks and 350-ml glass reservoirs. The eluting solution consisted of a 1:4 mixture of dichlorometh- ane:hexane (v/v). All solvents employed were reagent grade chemicals which had been redistilled. Butterfat was fluidized by placing in a warm oven; 1.00 g was transferred with a dropping tube to 25 g of conditioned Florisil and mixed thoroughly until a free- flowing powder was obtained. Twenty-five grams of deactivated Florisil was poured into a chromatographic column to form the bottom half of the cleanup system. This was prewashed with 50 ml of a 1 : 1 mixture of dichloromethane and hexane. The butterfat-Florisil mix- ture was then introduced to form the top layer. The column was eluted with 300 ml of the 1 :4 dichlorometh- ane: hexane elution mixture at a percolation rate of ap- proximately 5 ml/ minute. Eluates were concentrated just to dryness with rotary vacuum at 45 C; the residue was redissolved in 5.0 ml of hexane and transferred to 33 a glass-stoppered tube for subsequent gas chromato- graphic analysis. A Varian Aerograph Model 1200 gas-liquid chromato- graph, equipped with a 250 mc tritium electron capture detector, was used for all quantitative assays. Operating parameters were as follows: Column: 5' x Vs" Pyrex, packed with 4% SE-30 -\- 6% QF-1 on Chromosorb W, preconditioned 72 hours at 225 C Temperature: Column 175 C Detector 200 C Injector 225 C Carrier gas: Nitrogen at 40 ml/ minute Injection volumes were kept constant at 5fil for both sample solutions and comparison standards. Where necessary, sample solutions were diluted until chromato- graphic responses were within the linear range of the detector. Since all analyses were carried out under isothermal and isobaric conditions, peak heights alone were used for quantitation. Qualitative residue confirmations in milk samples ex- hibiting abnormal quantitative and/or qualitative char- acterstics were accomplished by thin-layer chromatog- raphy and by chemical conversion procedures (7). Recoveries of chlorinated hydrocarbons by the analytical procedure were checked regularly by fortification di- rectly into the milk prior to butterfat separation. Aver- aged recoveries were as follows: lindane, 91%; hepta- chlor epoxide, 94%; dieldrin, 92%; p.p'-DDE, 94%; p.p'-DDD, 90%; and p.p'-DDT, 93%. The data pre- sented in this report do not include recovery corrections. Where the composite sample contained residues of 0.1 ppm dieldrin or heptochlor epoxide or 1.00 ppm DDT plus metabolites or lindane, the milk from each pro- ducer making up the composite sample was collected for analysis. When a producer had been singled out as producing milk that exceeded the above mentioned levels, an immediate investigation was undertaken on his farm to find the source of contamination. This in- volved sampling feed, litter, water, medicines, and sprays for analysis. When the source of contamination was established, corrective measures were introduced, and a regular check was kept on the milk supply to ensure a decline in the residual levels. Results THE PROVINCIAL MILK SURVEY A total of 1,651 bulk tankers representing 286 dairies, creameries, and cheese factories located in 214 towns and cities of Ontario were sampled and analyzed. Each sample represented an average of 16.3 producers or 34 10,900 lb of milk. The average total residue of chlori- nated hydrocarbons in the butterfat was 0.171 ppm, consisting of 0.134 ppm DDT and its metabolites, 0.031 ppm dieldrin, 0.005 ppm lindane, and 0.001 ppm heptachlor epoxide (Table 3). This total residue was slightly lower than the level of 0.227 ppm reported by Duggan (J) for milk sampled between 1963 and 1966 in the United States. The level for DDT and its metabolites was identical in both surveys at 0.134 ppm; levels of dieldrin, lindane, and heptachlor epoxide averaged two- thirds, one-half, and one-tenth, respectively, of the amounts reported in the U.S. survey. Separation of the Ontario data on a regional basis re- vealed that the highest average level of DDT and meta- bolites occurred in the southern region (Table 3) where the major use of DDT in agriculture is practiced (Table 1). In this region, DDT levels in butterfat ranged from 0.120 ppm in milk produced in counties where general farming practices were followed and increased up to 0.328 ppm in counties of intensive cash crop production. The counties with the highest use pattern of DDT in 1968 tended to have higher residues in the milk, but several exceptions were noted. DDT residues in the central region varied from 0.056 ppm in counties of little agricultural activity to 0.215 ppm in the counties of intensive fruit and vegetable production. Most coun- ties between these two extremes fell into the general order of their use pattern. In the eastern and western regions, where general mixed agriculture is pursued, the total DDT residue was slightly over 0.1 ppm with little significance in the differences between counties. The ranges in levels were 0.084 to 0.137 ppm in the eastern region and 0.058 to 0.148 ppm in the western region. The northern region of the Province exhibited the lowest residue levels of DDT, ranging from 0.018 to 0.087 ppm in butterfat. Few milk samples in the Province were free of DDT and its metabolites (Table 4). Only 0.6% of samp?js from the eastern region contained nondetectable residues of p,p'-DDE; in all other regions, 100% of the samples contained measurable amounts. Levels of p,p'-DDD were nondetectable in 0.3, I.O, and 7.1% of the milk samples from the central, eastern, and northern regions, respectively, and p,p'-DDT was nondetectable in 0.3, 0.8, and 1.4% of the samples from the same three regions. There were no detectable lindane residues in milk samples from the northern region. The average levels of lindane in the central, eastern, and southern regions were between .001 -.002 ppm in the butterfat; how- ever, levels were nondetectable in over 96% of all samples from these regions (Tables 3 and 4). Only 9 of 46 counties in these three regions produced milk that contained detectable amounts of lindane, and in no Pesticides Monitoring JoimN.\L TABLE 3. — Residues of chlorinated hydrocarbons found in Ontario-produced milk (November 1967-June 1969) Region of Ontario Number of SAMPLES Average level of chlorinated hydrocarbons in butterfat of milk (ppm) P,p'-DDE p.p'-DDD p,p'-DDT Total DDT Lindane DiELDRIN Heptachlor EPOXIDE Central Eastern Northern Southern Western 387 489 70 372 333 .075 .046 .023 .115 .057 .034 .023 .017 .037 .023 .037 .032 .022 .041 .034 .146 .101 .062 .193 .114 .001 .001 ND .002 .017 .030 .022 .024 .043 .037 ND ND ND <.001 .003 Ontario (Total) 1651 .070 .029 .035 .134 .005 .031 .001 TABLE 4. — Frequency distribution of insecticide residues in butterfat of milk in Ontario (November 1967-June 1969) Frequency by region of Ontario (%) Insecticide (Ranges in ppm) Central Eastern Northern Southern Western Ontario Total p,p'-DDE .00— .09 1 79.6 98.8 100 59.7 94.0 84.6 .10— .19 18.3 1.2 — 30.4 5.7 12.6 .20— .29 2.1 — — 6.7 0.3 2.1 .30— .39 — — — 2.7 — 0.6 .40+ — — — 0.5 — 0.1 p^'-DDD .00— .09 » 96.4 99,2 97,1 98,4 99.1 98.2 .10— .19 2.6 0.8 2,9 1,0 0.9 1.4 .20— .29 0.0 — — 0,2 — 0.1 .30— .39 0.3 — — 0.2 — 0.1 .40+ 0.7 — — 0.2 — 0.2 p,p'-DDT .00— .09 I 97.4 97,4 97,1 96,5 97.9 97.3 .10— .19 2.6 2,2 2,9 2,7 1.8 2.3 .20— .29 — 0,2 — 0.5 — 0.2 .30— .39 — — — 0,3 0,3 0.1 .40+ — 0,2 — — — 0.1 Dieldrin .00— ,04 1 92.0 98,6 97,1 66,9 84,7 87.0 .05— .09 7,2 1,2 2,9 31,2 13,8 12.0 .10— .20 0.8 0,2 — 1.6 1.2 0.8 .20+ — — — 0.3 0.3 0.2 Heptachlor .00 100 100 100 96.5 89.2 97.0 epoxide .01— .09 — — — 3.5 10.5 2.9 .10+ — — — — 0.3 0.1 Lindane .00 96.6 99.4 100 96,5 71.2 92.4 .01— .09 3.4 0.4 — 3.2 23.7 6.4 .10+ — 0.2 — 0.3 5.1 1.2 Nondetectable residue levels: p,p'-DDE — 0.6% in eastern region; p.p'-DDD — 0,3, 1,0. and 7,1% in central, eastern, and northern regions; p,p'- DDT — 0,3, 0,8 and 1,4% in central, eastern, and northern regions; and dieldrin — 0,3 and 0.6% in central and eastern regions. case was the residue level higher than 0.002 ppm in ithe butterfat. In the western region, only 5 of the 10 counties produced milk with detectable amounts of lindane, and levels ranged from 0.007 to 0.045 ppm. Lindane residues were believed to be caused by three main sources: (7) the use of lindane vaporizers in milk rooms, (2) contamination of animal diets with treated seed, and (3) direct application to dairy herds for insect control. The highest average dieldrin levels (0,043 ppm) were recorded from milk samples from the southern region where the majority of aldrin and dieldrin is used in agriculture. A range of 0.026 to 0.070 ppm was present in the 12 counties of this region; those counties with the highest levels were intensively involved in tobacco, fruit, and vegetable production. Residue levels of dieldrin in the western region varied from 0.021 to 0.055 ppm. In the central, eastern, and northern regions, dieldrin levels in butterfat ranged from 0.019 to 0.033, 0.018 to Vol. 4, No. 2, September 1970 0.035, and 0.015 to 0.040 ppm, respectively. In samples from the central and eastern regions, 0.3% and 0.6%, respectively, contained nondetectable dieldrin levels. In the central, eastern, and northern regions, respectively, 99.2, 99,8, and 100% of fat samples tested showed dieldrin levels below 0.10 ppm. Administrative toler- ances of 0.10 ppm in butterfat were exceeded by 1.9% of the samples from the southern region, 1.5% from the western region, 0.8% from the central region, and 0.2% from the eastern region. Milk produced in the southern and western regions contained only low amounts of heptachlor epoxide (Table 3). In the southern region, the average level was below 0.001 ppm, and milk originating from only two of 1 2 counties had detectable residues. These counties bordered on three counties in the western region where heptachlor had been used for turnip production and/or seed treatment of cereal grain. Both cull turnips and treated seed were suspected as having been used as feed 35 for dairy cattle causing the appearance of heptachlor epoxide in the milk. The use of treated grain for feed is a violation of the Feeds Act and Regulations (5) and the use of cull turnips is not recommended. In the southern and western regions, 97 and 89%, respectively, of the samples had nondetectable levels of heptachlor epoxide. Only 0.3% of samples in the western region contained residues that were at the 0.1 ppm level or greater, i.e., levels above the permitted administrative tolerances. Heptachlor epoxide did not appear in milk fat from the central, eastern, or northern regions. HERDS WITH HIGH RESIDUES Where a composite sample contained residues in the fat above 1.0 ppm DDT or lindane, or 0.1 ppm dieldrin or heptachlor epoxide, samples of milk from each producer contributing to the composite were collected for an- alysis. From 17 bulk tankers rechecked, a total of 20 individual producers were found delivering milk with residues above these levels. In five cases, the residues were between 0.10 and 0.20 ppm dieldrin. In these cases, the milk was resampled at 2-week intervals; and if the dieldrin residue was found to be declining, no further action was taken, and the source of contami- nation was not determined. There were 12 producers with butterfat containing between 0.33 and 3.17 ppm dieldrin, one producer with 1.72 ppm lindane, and three producers with 3.36 to 17.7 ppm DDT and metabolites (Tables 5, 6, and 7). In each of these cases, the producers were visited, and a total of 258 samples of feed, litter, water, spray, etc., were collected and analyzed. In addi- tion, 354 milk samples, including individual cow samples and composite herd samples, were collected for analysis. In seven cases, the residues were the result of ingestion of contaminated feed or litter. In five cases, cattle had been sprayed with an insecticide not registered for that purpose or one that was contaminated with a persistent chlorinated hydrocarbon insecticide. In four cases, the source was no longer present and could not be de- termined, and four additional cases were not in- vestigated. (Table 8). Studies with dairy herds which exhibited residues above the tolerance levels for total DDT and dieldrin indicated that the rate of decline of these insecticides was de- pendent upon the original level and generally followed a decay curve pattern. The length of time required for a decline to reach acceptable levels varied according to the initial level, the effectiveness of removing the source, and the insecticide causing the problem. The time TABLE 5. — Distribution of dieldrin in Herds 1 and 2 at time of discovery and the decline of these residues Dieldrin in Herd 1 Herd 2 MILK PAT (Ranges in ppm) No. OF cows Average PPM No. OF cows Average PPM .00— .49 12 0.14 14 0.34 .50— .99 2 0.71 11 0.73 1.00—1.49 10 1.27 6 1.17 1.50—1.99 8 1.70 2 1.67 2.00 & over 7 3.43 Dieldrin (ppm) Days after discovery Seven highest cows — Herd 1 Herd 2 Back fat Milk fat Hair Milk fat 0 _ 3.43 _ 0.95 9 — — — 1.04 20—26 — 2.30 — 0.75 47—53 — 1.34 0.15 — 71—75 2.58 0.94 0.05 0.53 96—99 — 0.68 0.05 0.37 124—133 0.60 0.41 0.02 0.23 140—145 — 0.24 — 0.31 165 — 0.12 — — 184 — 0.16 206—218 0.16 0.09 — — Biological half-life (days) 34 37 — 58 TABLE 6. — Decline of dieldrin residues in milk after discovery and removal of source Dieldrin residues (ppm) in herds (No.) Days after discovery 4 5 6 7 8 9 10 11 0 0.58 0.94 1.09 0.44 0.78 2.20 0.66 1.60 15 — 0.90 1.08 0.35 — — 20 — — — — _ 2.17 29 1.58 0.89 1.02 0.28 — 38 — 0.52 1.04 0.21 0.46 0.98 0.30 1.63 50 0.80 0.49 1.09 — 65 0.40 — — — — — 1.52 78 — 0.53 1.12 — — — — 0.75 85 0.33 0.47 — — 0.17 0.69 103 — 1,26 0.39 0.23 126 — 0.77 0.13 0.49 145 — 0.52 0.85 0.10 0.10 0.09 0.49 163 — 0.45 0.37 — — 187 — — — — — 0.19 200 — 0.40 0.29 — — 0.23 230 0.11 0.53 0.37 — 0.10 275 0.16 0.27 308 — — 0.05 322 0.22 0.18 — 364 — 0.23 — — — — — Biological half-life (days) 50 - - 38 38 30 - 40 36 Pesticides Monitoring Journal TABLE l.—Declin e in DDT and metabolites in Herds 12, 13, and 14 Content in BirrrERFAT (ppm) Days after DISCOVEKY p.p'-DDE p.p'-DTm p.p'-DDT TOTAL DDT HERD 12 0 6.20 10.10 1.40 17.7 8 4.90 5.90 1.70 12.5 13 4.70 5.21 1.50 11.4 21 4.50 5.00 1.50 11.0 29 4.51 3.98 1.29 9.78 41 3.26 3.05 1.03 9.34 61 3.90 3.35 1.25 8.50 77 6.04 4.28 1.23 11.55 90 5.63 3.45 1.33 10.41 105 5.79 2.60 0.95 9.34 125 5.65 1.60 0.93 8.18 140 3.12 0.87 0.48 4.47 216 3,29 0.28 0.47 4.04 246 4.27 0.24 0.54 5.08 300 1.84 0.05 0.28 2.17 425 1.07 0.06 0.18 1.31 482 0.23 0.02 0.04 0.29 530 0.09 0.06 0.07 0.22 HERD 13 0 0.66 0.85 4.29 5.80 43 0.41 0.15 1.98 2.54 155 0.22 0.06 0.57 0.85 HERD 14 1 0 0.11 2.71 0.54 3.36 24 0.34 0.34 0.40 1.08 41 0.17 0.44 0.48 1.09 41 (3 heifers only) 0.36 1.41 1.51 3.28 63 0.14 023 0.41 0.98 78 0.13 0.10 0.34 0.57 1 This producer was a can shipper, and it is conceivable that only one can was sampled. TABLE 8. — Sources of conlaminalion in dairy he ■ds with high residues of persistent insecticides Method of SoiniCE OF Number CONTAMINATING COhrrAMINATION CONTAMINATION OF HERDS INSECTICIDE By ingestion Treated seed grains in feed Feed grains stored in contaminated bins Hay and feed grain low level contamination 1 1 1 Aldrin, dieldrin Aldrin Dieldrin Hay 2 Dieldrin Sweet com silage 1 DDT, metabolites Sawdust litter 1 Dieldrin By dusting or spraying Contaminated spray (Ciodrin, Vapona) 2 Aldrin, dieldrin Contaminated wettable powder used as dust (Rotenone) 1 DDT Improper use 2 Lindane. DDT Unknown No source found 4 Dieldrin Not investigated 4 Dieldrin required was usually in excess of 200 days (Tables 5, The milk from each cow of the herd was analyzed 6, and 7). High lindane residues (>1.0 ppm) dropped (Table 5), and this revealed a wide range of levels that rapidly to acceptable levels in about 2 weeks. failed to resolve the possible means of contamination. Twelve animals with the lowest residues, averaging 0.14 Herds with High Dieldrin ppm dieldrin, were mostly heifers that had only recently Herd 1. The composite milk sampled from the bulk entered the herd. Hamburger made from calves born to tanker prior to delivery at the dairy revealed a dieldrin some of the contaminated animals contained 2.2 ppm content of 0.48 ppm. This high dieldrin residue was dieldrin on a whole-tissue basis. The seven animals with traced back to one producer who was contributing milk the highest dieldrin levels were removed from the herd with a level of 3.17 ppm dieldrin in the butterfat. An- for study. alysis of various samples in an attempt to locate the source of contamination revealed that dieldrin levels Milk, hair, and fat biopsy were analyzed from these were only .005 ppm in soil, .003 to .004 ppm in hay and seven animals, and the loss of dieldrin was found to be grass .0001 ppm in drinking water, and nondetectable an exponential function of time (Table 5 and Fig. 2). in corn, cereal, litter, and feed additives. It was con- The decline in milk fat from 3.5 to 0.1 ppm took almost cluded that none of these sources were responsible for 200 days on a diet with nondetectable or insignificant the extremely high dieldrin level present. levels of dieldrin. Two animals were dry at day 96, and Vol. 4, No. 2, Sept EMBER 1970 37 FIGURE 2. — Decline in dieldrin content of hair, milk fat, and fat biopsies taken from Herd I at varying intervals following discovery 3- HERD 1 Biopsy Fat Milk Fat . 7 Highest Cows 12- \ Hair J o 3 \ Milk Fat 32 Cows S '■ *'''* --.., \^ ■ ■ - -■ --^ --—___ 50 100 ISO 200 TIME IN DAYS one of these calved at day 218. The level of dieldrin at the last milking of one was 0.57 ppm, and the first test after calving was 0.56 ppm. During this time period, the residue in the milk of the other five cows had de- clined from 0.68 to 0.09 ppm. The average level of the 32 cows remaining on the farm declined slowly from 0.94 to 0.22 ppm in 120 days. Analysis of hair from the seven test animals showed a steady decline from 0.15 to 0.02 ppm in 80 days. Fat biopsies were taken on three separate occasions and showed the level of dieldrin in the body fat to be considerably higher than that present in the milk fat. Herd 2. Investigation of a composite bulk tanker sample containing 0.15 ppm dieldrin in butterfat located the source with one herd contributing a level of 0.95 ppm. Analysis of the feed revealed either trace levels (.004 ppm) or nondetectable levels of dieldrin in all except one concentrate which contained 0.01 ppm. Samples of wood shavings used for litter were found to contain dieldrin at levels from 2 to 15 ppm. This herd had been kept on a restricted diet; and, as a result, wood shavings were being consumed. These wood shavings were traced back to teak timbers which had originally been grown and treated in Columbia, South America, and shipped to Ontario. Table 5 shows the average levels after analysis of individual herd cows. The level of dieldrin in the butterfat rose until the wood shavings were removed as litter, and then a decline to 0.16 ppm occurred over a period of 184 days. Herd 3. An excessive level (0.35 ppm) was found in milk from only one producer; this level declined to 0.05 ppm in 46 days, and no investigation for a source was undertaken. Herd 4. A large bulk tanker was found to have a total cyclodiene level of 0.19 ppm. Upon investigation, milk samples from five producers were found to contain levels between 0.10 and 0.20 ppm, one at 0.33 ppm, and one at 0.58 ppm. After further investigation, no source of contamination could be found for the first six. In the 38 case of the producer with the highest level, it was found that seed grain which had been treated with aldrin and heptachlor was being used in the herd's diet. The con- centrate feed contained 2.6 ppm heptachlor, 1.91 ppm aldrin, and 0.50 ppm chlordane; and over a 30-day feed- ing period, the milk fat residues of dieldrin rose from 0.58 to 1.58 ppm. When the contaminated feed was re- moved from the diet, the dieldrin level dropped from 1.58 to .11 ppm in 201 days (Table 6). Herds 5 and 6. The composite sample contained 0.16 ppm dieldrin, and two producers were found with milk containing 0.94 ppm and 1.09 ppm, respectively. Initial investigation failed to locate a source of dieldrin, and the residue in the butterfat remained unchanged. Further investigation on both farms revealed sources of hay and feed, previously overlooked, that were con- taminated. On the first farm, hay and oats which. con- tained 0.05 and 0.004 ppm dieldrin, respectively, were being fed to the herd. Following the removal of these feeds, the level dropped from 0.89 to 0.49 ppm in 21 days and then leveled out for the next 200 days (Table 6). Further tests revealed low levels on pasture grasses ranging from 0.01 to 0.03 ppm on a dry-weight basis which were growing in soil that contained levels from 0.003 to 0.06 ppm. A further decline to 0.23 ppm oc- curred after clipping these pastures. On the second farm, the source was found to be hay that contained 0.06 ppm dieldrin and 0.07 ppm aldrin. This hay was removed, and hay that contained 0.003 ppm was substituted. On this latter hay. the milk residue decreased from 1.26 to 0.18 ppm in 219 days. Pasture grasses were found to contain 0.005 ppm on a dry-weight basis while soil contained 0.002 ppm. It was concluded that general con- tamination of many feeds on these two farms made it difficult to remove the source and obtain a more rapid disappearance of the residues. Herd 7. The composite milk sample from the bulk tanker contained 0.10 ppm dieldrin in the butterfat, and the producer's milk fat contained 0.44 ppm dieldrin. The investigation revealed that an insecticide spray, used to control flies, was contaminated with 780 ppm of aldrin plus dieldrin, 214 ppm of DDT, and 1 18 ppm of endrin. Analysis of a similar batch of spray from the manu- facturer failed to show contamination at the factory level. After removal of the spray, the residue declined from 0.44 ppm to 0.10 ppm in 145 days (Table 6). Herds 8 and 9. A composite sample containing 0.37 ppm dieldrin in the butterfat was found to be caused by two producers with milk fat residue levels of 0.78 and 2.2 ppm of dieldrin. After investigation, it was dis- covered that the dieldrin in Herd 8 came from spraying the herd with two organophosphorus insecticides that were contaminated with 15 ppm aldrin. When this spray- ing was discontinued, the residue level dropped from 0.78 to 0.10 ppm in 145 days (Table 6). Pesticides Monitoring Journal Herd 9, when originally studied, revealed a dieldrin residue level of 2.20 ppm in butterfat. Over a period of 103 days, this level dropped to 0.39 ppm. No source of dieldrin contamination could be located until the level was noted to rise to 0.49 ppm, when it was discovered that this herd had been placed on new pasture which contained 0.004 ppm dieldrin (dry-weight basis). After the herd was removed from this pasture, the butterfat residue levels declined to 0.09 ppm within 3 weeks (Table 6). Herd 10. The composite butterfat sample contained 0.16 ppm dieldrin, and the individual producer was delivering milk with 0.66 ppm in the butterfat. The investigation showed that hay, produced in 1967 and 1968, was contaminated with 0.01 ppm dieldrin: feed grains were contaminated with 0.01 and 0.02 ppm; and dried beet pulp had 0.01 ppm. The combination of all three sources was considered responsible for contribut- ing to the dieldrin residue. On removal of these feeds, the residue in the milk fat declined from 0.66 to 0.23 ppm in 103 days (Table 6). Herd 11. TTie bulk tanker contained butterfat with a residue of 0.14 ppm dieldrin, and the producer was found to be delivering milk with 1.60 ppm in the butterfat. Wheat and oat grains were found to be con- taminated with 0.5 to 0.6 ppm aldrin and 0.01 to 0.04 ppm dieldrin, respectively. The grains had been stored in wooden granary bins that had been treated with aldrin. Hexane washings of the bins resulted in solutions con- taining 26 ppm aldrin. It was concluded that the grain became contaminated as the result of volatilization of aldrin from the wood. When the feeding of this grain was discontinued, the residue in the butterfat declined from 1.60 to 0.10 ppm in 230 days (Table 6). Herds with High DDT and Melaboliies Herd 12. This herd was discovered after a bulk tanker was found with a total DDT level of 6.78 ppm. The investigation showed milk fat with 17.7 ppm at the farm level. Soil on the farm contained only 0.01 ppm; con- centrates and feeds contained from a trace to 0.01 ppm; and silage had only 0.03 ppm. Oat straw had the highest residue with 0.11 ppm. At a later date, a total source of ODD and DDT of 40 ppm was located on sweet corn silage. After removal of this silage from the feed, the level dropped from 17.7 to 0.22 ppm in 530 days (Table 7). The rise in the residue at 77 days was associated with a number of heifers and cows that calved. Sampling three of these animals at random revealed an average residue level of 21.3 ppm DDT plus metabolites in their milk fat. Following this rise, the decline in residue continued at an exponential rate (Table 7 and Fig. 3). Vol. 4, No. 2, September 1970 FIGURE 3. — Changes in DDT and its metabolites in milk fat following its discovery in Herd 12 16- HERD 12 14 \ Total Residue t'^ DDD a: \ r\ DDE ■^ 10 A \ DDT » 8 \ ■' \ * a /--x\ 4 \ !/'• \ ---^ •';..-•,' •. •-. •-,>- __^ -T77^ 2s. - 100 200 300 400 500 TIME IN DkVS Herd 13. A bulk tanker was located with 2.30 ppm, and the producer was found to be delivering milk with 5.80 ppm DDT and metabolites. An investigation re- vealed that the herd had been sprayed with technical DDT. The total DDT level declined from 5.80 to 0.85 ppm in 155 days while the respective drop in p,p'-DDT was from 4.29 to 0.57 ppm (Table 7). Herd 14. A producer was discovered delivering milk with 3.36 ppm DDT and metabolites in the butterfat. The investigation uncovered a rotenone wettable powder that contained 6.4 ppm DDT and DDD. This wettable powder had been used to dust a group of recently purchased heifers that were infested with lice. Among a total milking herd of 17 animals, only 3 had been treated with the contaminated powder, and the contribution from these 3 was sufficient to raise the residue level of DDT and its metabolites to an average level of 3.36 ppm. This level declined to 0.57 ppm over a period of 78 days. Herd with High Lindane Herd 8. In addition to a high dieldrin residue in the milk, a lindane level of 1.72 ppm was also present. This lindane residue was found to originate from spray treatment for lice control one day before sampling. The lindane residue dropped to 0.12 ppm in 41 days and 0.03 ppm in 84 days. Discussion The analytical data indicated that dieldrin residue levels in butterfat from milk produced in agricultural areas where aldrin and/ or dieldrin had been used extensively for insect control were only approximately twice as high as dieldrin residue levels in areas where little or 39 none of these cyclodienes were used. This would sug- gest that the cyclodienes have considerable mobility in the environment. The fact that aldrin was used almost to the exclusion of dieldrin would further suggest the high environmental mobility of aldrin. It is reported that aldrin is moderately volatile (vapor pressure 6 x 10-*' mm at 25 C) and hence could move into the atmosphere to explain the wide and fairly even distribu- tion of dieldrin residues in the butterfat of milk in the Province (8). The use of animal feed produced in other areas could partially explain this rather uniform background level of dieldrin but could not fully explain dieldrin presence in areas where dairy farms produced most of their own feed. As was expected, only dieldrin and no aldrin was detected in butterfat samples. The pattern of decline of high dieldrin residues in adipose tissue was similar to that reported by Moubry et al. (11,12). In the seven highest cows in Herd 1, the analyses of back fat samples indicated that dieldrin levels were considerably higher than those found in the butterfat, and the exponential disappearance of dieldrin was found to be slightly more rapid in the back fat. A biological half-life was calculated at 34 and 37 days, respectively, for back fat and milk fat (Table 5). In seven herds where sources of dieldrin were located and removed, biological half-lives ranged from 30 to 50 days (Tables 5 and 6). This wide range could reflect either the effectiveness of removal of the dieldrin source or the general background dietary level. The declines of residues in Herds 3, 5, 6 and 10 were complicated by the presence of multiple sources that were not iden- tified or removed simultaneously. Of the 1 1 herds investigated, 7 were located in the southern region, 3 in the western region, and 1 in the central region. This frequency closely followed the use pattern of aldrin and dieldrin in these respective regions (Table 1). In only one case was the use of aldrin by a producer connected to the presence of high dieldrin levels in milk. In all other cases, the farmer was un- aware of the presence of aldrin or dieldrin on the farm. In four of these instances, the pesticide was present in feed either as the result of drift or treated seed grain mixed with the feed. In one case, the dieldrin was present in wood-shaving litter and in two cases as the result of contamination of insect spray formulations. As a result of the general distribution and persistence of dieldrin in milk, whether from the use or misuse of aldrin and dieldrin, the Department of Health of On- tario in 1969 introduced a complete ban on the sale of these insecticides under the Pesticide Act, 1969 (13). DDE, DDD, AND DDT The largest agricultural use of DDT (89%) is confined to the southern region of the Province. However, the residue levels of DDT and its metabolites in milk from 40 this area were only three times higher than the levels in the northern region where less than 0.1% of this insecticide is used in agriculture. This suggests possible mobility in the environment; but, more probably, the residues in the northern region may result from the use of DDT for nonagricultural purposes, i.e., recreation and parklands, industrial expansion, and commercial forestry operations. It should be noted that, although the amount of DDT used for agricultural purposes was 10 times that of aldrin and dieldrin combined, the num- ber of herds which violated tolerance levels for DDT was only 3 as compared to 17 for dieldrin; however, the residue tolerance of dieldrin is only one-tenth that of DDT and metabolites. Two of these herds were located in the southern region and one in the western region. In two of these cases, high DDT residues re- sulted from direct application of DDT to dairy cows (in one instance as a contaminant in an approved in- secticide). The third case resulted from a feeding diet which contained corn silage with DDT plus metabolite levels of 40 ppm. The decline of DDT levels in Herd 12 initially followed a normal pattern and then exhibited a sudden rise. This was attributed to the introduction of milk from 10 freshly calved cows that were secreting much higher residues in their milk than the rest of the herd. If the beginning of decline is taken from this latter point, then the biological half-lives of DDE, DDD, DDT and total were calculated at 76, 37, 98 and 76 days, re- spectively. In Herd 13, which had been subjected to a DDT spray, these respective half-lives were 66, 47, 39 and 65 days; and in Herd 14, they were 85, 14, 52 and 63 days. The elimination of DDE appeared to take the longest time and that of DDD the shortest. Although disappearance rates of the DDT varied widely from one herd to another, the disappearance of the DDT plus its metabolites was remarkably similar with half-lives ranging from 63 to 76 days in the three herds. The half-life for DDT and its metabolites was ap- proximately twice that found for dieldrin. Approximately 8% of the milk samples analyzed con- tained detectable levels of lindane, and since these samples were localized to only a few areas, the lindane source was comparatively easy to locate. High lindane residues resulted primarily from use of treated seed grains as feed and by direct application for insect con- trol. Neither of these practices is permitted in dairy husbandry, and both are readily remedied by education and extension programs or by the threat of litigation. Only one herd was located (in the southern region) which exhibited levels of lindane residues at high enough concentration to study the half-life which was calculated to be about 14 days. Pesticides Monitoring Journal HEPTACHLOR EPOXIDE Heptachlor epoxide was detected in only 37c of the milk samples. These findings were confined to areas where heptachlor is or had been widely used for seed treatment and/or the cultivation of turnips. Since only small quantities of heptachlor have been used in On- tario agriculture, no problems of high residues were encountered. samples across the Province. A special thanks goes to Mr. E. H. Smith, Milk Commission, for coordinating and organizing the sample collection. Appreciation is expressed to Mrs. G. E. Bowers, Provincial Pesticide Residue Testing Laboratory, Ontario Department of Agriculture and Food, who confirmed residues by thin- layer chromatography and to Dr. D. J. Ecobichon, University of Dalhousie, Halifax, Nova Scotia, for col- lection and analysis of the fat biopsies from Herd 1. Conclusion See Appendix for chemical names of compounds mentioned in this paper. The residues of chlorinated hydrocarbon insecticides in Ontario-produced milk were at levels similar to or lower than those reported in surveys from other parts of the world (2,3.4.5). The low incidence of excessive resi- dues (20 per 27,000) indicate that the Ontario farmer practices insect control with caution. In only 2 cases were high residues caused by improper use of insecti- cides; the remaining 18 problems arose from the in- troduction of contaminants from outside the particular farm, i.e.. purchase of contaminated feed or contamina- tion of insecticide formulations. Lindane residues disappeared rapidly from contaminated animals with a biological half-life of approximately 2 weeks. Dieldrin and DDT and its metabolites exhibited lower rates with half-lives of approximately 6 weeks and 10 weeks, respectively. Background residue levels of DDT in milk could only partially be related to the use of DDT in agriculture and might be partially influenced also by nonagricul- tural applications of the insecticide. The general ap- pearance of dieldrin residue over all areas of the Province suggests a high environmental mobility for aldrin and/or dieldrin. Lindane and heptachlor epox- ide residues resulted primarily from localized use. The use of DDT in Ontario has recently been severely restricted, and a total ban on aldrin, dieldrin, and hepta- chlor became effective in 1969. These restrictions should be reflected shortly in a change in the general background residue levels in Ontario-produced milk. A cknowledgments The authors wish to thank Mr. G. A. McCague, Chair- man of the Milk Commission, Ontario Department of Agriculture and Food, Toronto, and the members of his branch who participated in the collection of milk LITERATURE CITED (/) Ontario DepartmenI of Agriciilliire and Food. 1968. Agricultural statistics for Ontario. Publ. 20. Parliament Buildings, Toronto. (2) Bio-Rasmussen, F., Sv. Dalgaard-Mikkehen, Th Jakob- sen. Sv. O. Koch F. Rodin, E. Uhl, and K. Voldum- Clausen. 1968. Examinations of Danish milk and but- ter for contaminating organochlorine insecticides. Resi- due Rev. 23:55-69. (i) Clifford. P. A. 1957. Pesticide residues in fluid market milk. Public Health Rep. 72:729. (4) Clifford. P. A., J. L. Bassen. and P. A. Mills. 1959. Chlorinated organic pesticide residues in fluid milk. Public Health Rep. 74:1109. (5) Duggan, R. E. 1967. Residues in food and feed. Pesti- cides Monit. J. 1:2-8. (6) Feed Act and Feed Regulations. 1960. Chapter 14, established by PC 1967-1072. Queen's Printer and Con- troller of Stationery, Ottawa, 1967. (7) Hamence, J. H., P. S. Hall, and D. J. Caverly. 1965. The identification and determination of chlorinated pesticide residues. Analyst 90:649. (S) Harris, C. R., and E. P. Lichtenstein. 1961. Factors affecting the volatilization of insecticidal residues from soil. J. Econ. Entomol. 54:1038-1045. (9) Langlois, B. E., A. R. Stemp, and B. J. Liska. 1964. Analysis of animal food products for chlorinated insec- ticides. J. Milk and Food Technol. 27:202. (10) Moubry, J. R., G. R. Myrdal, and H. P. Jensen. 1967. Screening method for the detection of chlorinated hydrocarbon pesticide residues in fat of milk, cheese and butter. J. A.O.A.C. 50:885. (;;) Moubry, R. J., G. R. Myrdal, and W. E. Lyle. 1968. Investigation to determine the respective residue amounts of DDT and its analogues in the milk and back fat of selected dairy animals. Pesticides Monit. J. 2:47-50. (12) Moubry, R. J.. G. R. Myrdal. and A. Sturges. 1968. Rate of decline of chlorinated hydrocarbon pesticides in dairy milk. Pesticides Monit. J. 2:72-79. (13) The Pesticides Act, 1967. Statutes of Ontario. Chapter 74, and Ontario Regulation 445/67. Printed and pub- lished by Frank Fogg, Queen's Printer, Toronto, 1969. Vol. 4, No. 2, September 1970 41 Residues of Chlorinated Hydrocarbons in Soybean Seed and Surface Soils From Selected Counties of South Carolina^ W. R. McCaskill,' B. H. Phillips, Jr.,= and C. A. Thomas' ABSTRACT A residue study of soybeans grown on soil witli a history of cotton production was initialed in 1967 to determine the chlorinated hydrocarbon residue levels in soybeans and if a relationship existed between the concentration of chlorinated hydrocarbons in soybeans and soils. Soybean and soil (0-6 inch depth) samples were collected in three counties from each of four selected soil regions. In addition, soybean sam- ples were collected from nine buying stations within the test areas. The residue levels of chlorinated hydrocarbons in soy- beans varied from 0.007-0.156 /ig/g (ppm) and in the soil from 0.000-3.582 /ig/g. There was no significant correlation between amounts of chlorinated hydrocarbons in soybeans and residues in soil. Soybeans from tlie nine buying stations contained a slightly lower average concentration of chlori- nated hydrocarbons than the soybeans from the test areas. Approximately 40% of the soybean samples contained de- tectable levels of aldrin, heptachlor, or chlordane. The study indicated that residues of DDT probably would not exceed present tolerance levels provided current pesticide recom- mendations are followed. Introduction Insects aifecting cotton production are usually con- trolled by one or more of the chlorinated hydrocarbons. In recent years large acreages of land formerly planted to cotton have been converted to soybeans. Technical Contribution No. 791, South Carolina Agricultural Experi- ment Station. Published with the approval of the Director. Agricultural Chemical Services Department, Clemson University, Clemson, S. C. 29631. Department of Entomology and Zoology, Clemson University, Clem- son, S. C. 29631. 42 Bruce et al. (2) grew soybeans, peanuts, oats, corn, and barley on soil that had been treated with varying rates of aldrin and heptachlor up to 20 lb/ acre. The concentration of pesticides in plants varied directly with the level of pesticides in the soil, but seed with a high oil content, such as soybeans and peanuts, con- tained the highest levels of pesticides. Morgan et al. (4) found that 2 lb/ acre each of aldrin, heptachlor, chlordane, and endrin applied as granules to separate plots resulted in relatively high residues of aldrin and heptachlor in both the shell and nuts of peanuts and of chlordane and endrin in the nuts. Bruce and Decker (/) reported that the concentration of chlorinated hydrocarbons in the soil was from 10 to 15 times larger than the residue concentration in soybean seed. However, the residue levels in both soil and soybeans decreased with time. In 1967, West Germany lowered the tolerances on chlorinated hydrocarbons allowed in soybeans. As soy- beans grown on land formerly used for cotton might contain excessive levels of chlorinated hydrocarbons, a survey was initiated to determine the residue levels in soybeans from four areas of South Carolina and to de- termine if there was a correlation between the levels of chlorinated hydrocarbons in the soil and the soybeans. Sampling Procedures Soybean and soil samples were collected from the Piedmont (Anderson, Greenville, and Spartanburg counties). Sandhills (Aiken, Lexington, and Richland counties), the Pee Dee section (Darlington, Marlboro, and Lee counties), and Savannah River Valley section Pesticides Monitoring JotmNAL (Allendale, Barnwell, and Hampton counties) of the Coastal Plain. The county agents contacted farmers that had been following the recommended insecticide program for cotton and selected fields planted to soy- beans that had a long history of cotton production. Soybeans were harvested, and soil samples (0-6 inch depth) were taken from a 12' x 12' section of each field. Each county agent collected soybean and soil samples from 10 fields (a total of 120 soybean and soil samples). In addition, 29 soybean samples were collected from 9 buying stations within the test areas. Analytical Procedures Each sample (soybeans and soil) was mixed thoroughly. The soybeans were ground in a Servall Omni-Mixer. Soil samples were air dried and screened through a 2- mm sieve. EXTRACTION Samples were extracted by a modification of the pro- cedure outlined by Van Middelem and Moye (i). Each sample — 50 g of soybeans or 25 g of soil — was shaken for 1 hour on a wrist-arm shaker with 100 ml of a 1 : 1 hexane-acetone mixture. The extracting solution was decanted through a filter into a 500-ml separatory funnel. The sample was shaken for another hour with an additional 100 ml of 1:1 hexane-acetone mixture, and the filtrate was added to the original extracting mixture. The acetone was removed from the mixture with three washings (100 ml each) of distilled water. Soils The hexane was evaporated to dryness and the residue transferred to a 100-ml bottle with 25 ml of nanograde hexane. Soybeans The pesticides were extracted from the hexane with four 25-ml portions of hexane-saturated acetonitrile. The acetonitrile was evaporated to dryness, and the residue was dissolved in hexane and transferred to a column of 20 g of activated Florisil (containing 5% HoO). The pesticides were eluted from the column and separated from the remaining oils with 80 ml of a 6% ether-94% hexane mixture. The extract was evap- orated to dryness and the residue transferred to a 100-ml bottle with 25 ml of nanograde hexane. Analysis Five microliters of each sample was injected into a MicroTek 220 gas chromatograph and concentration of DDE, o.p'-DDT, p,p'-DDT, lindane, heptachlor, aldrin, and chlordane determined with an electron capture de- tector with a Nigj source. Retention times were cross- VoL. 4, No. 2, September 1970 checked on 6' x '4" stainless-steel columns containing 5% SE-30, 5% DC-200, and 5% QF-1. Average recoveries of pesticides were as follows: Average Percent Recovery Insecticide Soil Soybeans Aldrin 50.5 21.4 DDE 38.0 40.4 o.p'-DDT 60.5 39.9 p.p'-DDT 81.0 60.4 Heptachlor 54.0 28.0 Lindane 53.5 58.0 Recovery values were obtained by adding pesticide standards to soil and soybeans at the following rates: 0.5 ng each of aldrin, heptachlor, and lindane; 1.25 ng of DDE: 1.14 ng of o.p'-DDT: and 3.86 ng of p.p'-BBT. Since heptachlor epoxide and dieldrin were not detected, some of the samples containing heptachlor and aldrin were analyzed by a GLC using a different column sys- tem (1.59'f OV-17/1.957f QF-1) from that reported. The peaks from the two column systems were identical, and no peaks for heptachlor epoxide and dieldrin were found on either system. Florisil that had been deacti- vated with 5% HoO did not retain heptachlor epoxide and dieldrin. Recovery factors were not applied to the reported values (Tables 1, 2, and 3). Results and Discussion The concentration of chlorinated hydrocarbons in soy- beans from the 120 fields varied from 0.007 /xg/g to 0.156 /Mg/g (Table 1). Even the sample with the high- est concentration of chlorinated hydrocarbons did not exceed established tolerance levels for DDT. The concentration of pesticides in both soybeans and soils (0-6 inch depth) varied with the soil region. The soybeans from the Piedmont contained the lowest aver- age concentration of chlorinated hydrocarbons while the soybeans from the two areas of the Coastal Plain region had the highest concentrations (Table 2). Conversely, the soils of the Piedmont contained the highest levels of pesticides while the lowest concentrations were in the soils of the Sandhill region. The surface soils contained from 8 (Savannah River Valley and Sandhills) to 26 (Piedmont) times more extractable chlorinated hydrocarbons than the soybeans. The difference between the concentration of pesticides in the soil and the uptake by soybeans appears closely related to soil texture and percent organic matter. Piedmont surface soils contain more clay and are higher 43 in aluminum and iron oxides than the soils of the other regions. Soils of the Pee Dee area contain more organic matter than the soils from the Savannah River Valley while the soils of the Sandhill region are coarse- textured sands. There was no significant correlation between the amounts of chlorinated hydrocarbons in the topsoil and the residue levels in the soybeans. Although difference in soil retention was a contributing factor, the poor corre- lation was probably due to cultural practices, such as deep plowing and subsoiling, which mechanically moved a portion of the pesticide residues into the root zone below the 6-inch depth. The soybean samples collected from nine buying stations contained a slightly lower average concentration of chlorinated hydrocarbons (Table 3) than the average of the soybeans from the survey fields in the same counties. The soybeans from the fields selected in the survey appear to be representative, in regard to pesticide residues, of the soybeans from the area that were moving into commercial channels. Sixty-three of the 149 soybean samples contained de- tectable levels of aldrin and/ or heptachlor, and I sample contained detectable levels of chlordane (0.001 ixg/g). These are chlorinated hydrocarbons for which a zero or no-tolerance has been established. Forty-six samples of soybeans contained aldrin (0.001 -0.0O8 fig/g, mean of 0.002 ;ug/g), and 21 soybean samples contained heptachlor (0.001-0.020 ^g/g, mean of 0.0035 /xg/g). However, only one sample from the Piedmont region (farms and buying stations) had detectable levels of these three chlorinated hydrocarbons. None of the soybean samples containing DDT exceeded established tolerance levels. This survey indicates that there is little danger of DDT concentrations in soybeans grown on soils formerly planted to cotton exceeding the present tolerances if current pesticide recommendations for soybeans are followed. However, soybeans grown on soils with a history of aldrin and heptachlor application probably will contain detectable levels of these chlorinated hy- drocarbons. TABLE I. ^Pesticide residue data for soils and soybeans from four regions of South Carolina (residue levels shown for indi- vidual compounds are those that were found in fields containing the highest and lowest total amount of chlorinated hydrocarbons) County Residues in fio/o Total Chlorinated Hydrocarbons Breakdown by Individual Compounds Region DDE o,p'-DDT p.p'-DDT Lindane Heptachlor Aldrin High Field Low Field High Field Low Field High Field Low Field High Field Low Field High Field Low Field High Field Low Field High Field Low Field Piedmont Anderson 2.392 .074 .504 .026 .356 .017 1.480 .024 .052 .007 Spartanburg 2.355 .019 .490 .019 .339 1.510 .016 _ Greenville 1.152 .032 .270 .008 .142 .007 .730 .017 .010 — — — — — SandhiUs Richland 1.428 .103 .386 .039 .150 .017 .856 .047 .036 _ _ _ Aiken 1.327 .052 .322 .027 .199 .010 .806 .014 — .001 — _ Lexington .614 — .290 — .054 — .270 — — — — — — — Pee Dee Darlington 1.241 .354 .170 .058 .215 .052 .856 .244 _ _ _ _ _ _ Marlboro 1.699 — .378 .325 .996 — Lee 1J43 .166 .270 .035 .139 .017 .810 .114 .024 — — — — — Savannah Hampton 3.582 .006 .540 .522 2.520 .006 _ River Allendale 1.239 .036 .300 .014 .143 .790 .022 — .006 Valley Barnwell 1.668 — .290 — ,298 — 1.080 — — — — — — — SOYBEANS Piedmont Anderson .068 .019 .065 .015 .003 .004 Spartanburg .068 .008 .046 .006 .011 .002 .011 Greenville .036 .007 .010 .007 .014 — .012 — — — — — — — Sandhills Richland .116 .030 .011 .010 .005 .003 .040 .006 .060 .011 Allien .067 .015 .031 .001 .017 .009 .016 .003 .005 Lexington .061 .022 .010 .012 .022 .001 .026 .003 .003 .006 — — — .001 Pee Dee Darlington .156 .026 .009 .010 .018 .007 .129 .008 .001 Marlboro .108 .021 .041 .019 .035 .028 .002 .002 .002 Lee .095 .015 .010 .010 .008 .001 .012 — .063 .002 — .002 .002 Savannah Hampton .143 .020 .024 .005 .061 .007 .056 .008 _ .002 River Allendale .146 .022 .062 .018 .033 .001 .044 .002 .001 .002 .002 .003 Valley Barnwell .083 .026 .008 .013 .021 .004 .034 .007 .020 .002 — — — — 44 Pesticides Monitoring Journal See Appendix for chemical names of paper. npounds mentioned in tliis LITERATURE CITED (7) Bruce, W. N. and G. C. Decker. 1966. Insecticide resi- dues in soybeans grown in soil containing various con- centrations of aldrin, dieldrin, heptaclilor. and hepta- chlor-epoxide. J. Agr. Food Chem. 14:395-398. (2) Bruce, W. N., G. C. Decker, and Jean G. Wilson. 1966. The relationship of the levels of insecticide contamina- tion of crop seeds to their fat content and soil concentra- tion of aldrin, heptachlor and their epoxides. J. Econ. Fntomol. 59:179-81. (3) Van Middelem, C. H. and H. A. Moyle. 1966. "Initial Results on Florida Preliminary Soils." Memo to S-58 cooperators. (4) Morgan. L. W.. D. B. Leuck. E. W. Beck, and D. W. Woodham. 1967. Residues of aldrin, chlordane, endrin, and heptachlor in peanuts grown in treated soil. J. Econ. Entomol. 60:1289-91. TABLE 2. — Average total residues of chlorinated hydrocarbons in surface soils and soybeans from four regions of South Carolina Residues in (ig/g ' o,p'-DDT p.p'-DDT Lindane Heptachlor Aldrin Chlordane Total Piedmont Anderson .418 .179 .682 .026 _ 1.305 Spartanburg .191 .106 .433 .008 — — — 0,737 Greenville .112 .029 .178 .007 — — — 0,326 Average .244 .105 .431 .014 - - - 0,789 Sandhills Richland .102 .061 .209 .003 — — .212 0.586 Aiken .118 .065 ,278 .002 — .001 — 0.463 Lexington .053 .018 .062 — — — — 0.133 Average .090 .048 .180 .002 - - .078 0.398 Pee Dee Darlington .147 .155 .472 _ — - — 0.774 Marlboro .139 .132 .379 .008 — — — 0.659 Lee .144 .074 .378 .002 — — — 0.598 Average .144 .122 .411 .004 - - - 0.681 Savannah Hampton .205 .110 .509 .009 — — — 0.833 River Allendale .108 .049 .214 .002 — — — 0.374 Valley Barnwell .066 .051 .234 .002 — — .033 0.386 Average .127 .069 .315 .004 — — .010 0.525 Piedmont Anderson .018 .004 .006 ,002 0.030 Spartanburg .016 .007 .010 ,001 — — — 0.034 Greenville .010 .006 .009 .002 — — — 0.027 Average .015 .006 .008 .002 - - - 0.031 Sandhills Richland .011 .010 .019 ,020 — — — 0.060 Aiken .016 .011 .006 .005 — — — 0.038 Lexington .011 .005 .015 .007 — .001 — 0.039 Average .013 .009 .014 .011 - - - 0.047 Pee Dee Darlington .011 .008 .027 .012 — .001 — 0.059 Marlboro .026 .020 .014 .004 .003 .001 — 0.068 Lee .011 .009 .019 .009 — .002 — 0.050 Average .016 .012 .020 .008 .001 .001 - 0.058 Savannah Hampton .017 .021 .018 _ _ .002 _ 0.058 River Allendale .025 .016 .016 .002 .002 — — 0.061 Valley Barnwell .015 .011 .016 .005 — — .001 0.048 Average .019 .016 .017 .002 .001 .001 — 0.056 1 County values are the means of 10 locations. Vol. 4, No, 2, September 1970 45 Soa Region TABLE 3. — Comparison of the average levels of chlorinated hydrocarbons in soybeans from survey fields to levels in soybeans collected from buying stations Residues in ^g/o ^ o.p'-DDT p,p'-DDT Lindane Heptachlor Aldrin Chlordane Total Piedmont Anderson .018 .004 .006 .002 _ .030 Sandhills Aiken Lexington .016 .011 .011 .005 .006 .015 .005 .007 - .001 - .038 .039 Pee Dee Darlington Marlboro Lee .011 .026 .011 .008 .020 .009 .027 .014 .019 .012 .004 .009 .003 .001 .001 .002 - .059 .068 .050 Savannah River Valley Hampton Allendale Barnwell .017 .025 .015 .021 .016 .011 .018 .016 .016 .002 .005 .002 .002 .001 .058 .061 .048 Average .017 .012 .015 .005 .001 .001 - .051 BUYING STATIONS Piedmont Anderson .019 .003 _ .003 _ .024 Sandhills Aiken Lexington .021 .010 .004 .003 .009 .009 .002 .003 z .001 - .036 .026 Pee Dee Darlington Marlboro Lee .011 .015 .013 .011 .032 .014 .027 .018 .019 .002 .007 .004 .005 .002 .001 .001 - .053 .078 .051 Savannah River VaUey Hampton Allendale Barnwell .016 .008 .011 .020 .004 .008 .024 .015 .017 .002 .001 .003 — .003 - .065 .028 .039 Average .014 .012 .017 .003 .001 .001 - .047 ^ County valu es are the means of 10 locatior s. 46 Pesticides Monitoring Journal PESTICIDES IN PEOPLE Serum Organochlorine Pesticide Levels in People in Southern Idaho^ Michael Watson, W. W. Benson, and Joe Gabica ABSTRACT In a study of 1,000 serum samples from people in southern Idaho, p,p'-DDE was found in 99.8% of all individuals, with a mean concentration of 22.0 parts per billion (ppb). Sam- ples were selected with no consideration of sex, race, age, or prior medical history at the time of collection. Pesticide levels within the group differed somewhat from those of similar demographic studies; this most likely may be at- tributed to regional differences in pesticide usage, the rela- tively large number of persons sampled, and possibly by the method of testing. Sampling Procedures Blood samples from 1,000 Idahoans were obtained at the Medical Center at Nampa, Idaho. For the sake of simplicity, first-time visitors to the Center for any reason at all who were willing to participate in this study were chosen. Selection was based wholly upon availability, and variables such as age, sex, race, and prior medical history were not considered. Although such a sampling method would by no means result in an unbiased representative cross-section of Idahoans, it was none- theless contended that useful preliminary data for the southern Idaho region could be obtained in this manner. Introduction Canyon County, Idaho, is one of 15 areas in the United States currently participating in a Community Studies Research Program to determine the effects of pesticides on human health. In-depth investigation of the total en- vironment is being conducted to determine the type and quantity of pesticide exposure. The study reported here was undertaken during the period 1967-1968 in order to establish a base line of serum pesticide levels in a sample group of Idahoans. Serum, rather than whole blood, was chosen for this study because of its known higher pesticide content (2,5). Although little is currently known concerning the relationship of serum pesticide levels to the more exten- sive "body burden" of adipose tissues, it appears that some proportionality may exist (4). ' From the Idaho Community Studies on Pesticides, Idaho Department of Health, Stalehouse, Boise, Idaho, 83707, Vol. 4, No. 2, September 1970 Laboratory Procedures EXTRACTION Whole bloods were centrifuged immediately to separate the serums. The serums were then frozen and taken to the Idaho State Health Laboratory in Boise for sub- sequent extraction and analysis for organochlorine pesticides. Serum extraction was carried out by a revised method of Dale and Cueto, as recommended by the Primate Re- search Branch Laboratory in Perrine, Fla. (2). Two ml of serum were placed in a 15-ml centrifuge tube; 6 ml of nanograde hexane was added; and the mixture was agitated for 3 minutes on a Vortex mixer. Following centrifugation at 2,000 rpm for 10 minutes, the hexane layer was transferred into a 50-ml concentrator tube by means of a disposable pipette. This extraction procedure was repeated three times. Emulsions that occasionally formed were broken by the addition of small amounts of acetonitrile. The three combined hexane fractions were then concentrated with a two-ball Snyder column on a 47 Temperatures: Column air flow: steam bath to a volume of 500 /aI. Five-microliter por- tions were then injected into a MicroTek 220 gas chromatograph equipped with a tritium-foil electron capture detector. ANALYSIS The operating parameters for gas chromatographic analysis were as follows: Columns: 1.5%, OV-17; 1.95%, QF-1 on Chromosorb W, DMCS, H.P., 100/ 120 mesh 4%, SE-30; 6%, QF-1 on Chro- mosorb W, DMCS, H.P., 100/ 120 mesh Column 200 C Injection chamber 225 C Detector 210 C SE-30/QF-1 100 ml/minute OV-17/ QF-1 70 ml/ minute All qualitative retention times were based on the reten- tion time of aldrin. Quantitation of pesticide residues was based on relative peak heights. Recovery, which ranged from 80%-100% was based on the addition of a known amount of aldrin to each sample prior to extrac- tion. (Although spiking the sample in this manner is not indicative of 100% recovery of the organically bound pesticides, it does provide an index of efficiency for the methodology used.) Results and Discussion Eight different pesticide residues were present at de- tectable levels in the serums of the study group. In decreasing order of detection frequency, these were: p,p'-DDE, p,p'-DDT, dieldrin, p,p'-DDD, ^-BHC, heptachlor epoxide, lindane, and a-BHC. The mean con- centrations, ranges, and percent occurrences of these residues are shown in Fig. 1. Of the total serums an- alyzed, 99.8% contained p,p'-DDE at a mean concen- tration of 22.0 ppb; p.p'-DDT occurred in 84% of the serums and averaged 4.7 ppb. Dieldrin, found in 33% of the samples, averaged 0.5 ppb. Only 7.1% of the serums contained p,p'-DDD, the average serum level being 0.24 ppb. Fewer than 2.6% of the individuals sampled contained any of the four remaining residues, the means of which were all less than 0.07 ppb (Fig. 1). In a similar study by Davies et al. (4) of pesticide levels in whole bood of 68 Florida Caucasians, /7,p'-DDE was found at a mean concentration of 8 ppb, and a range of 2-19 ppb was reported. These values are considerably lower than those found in the Idaho residents, but some of this variation is probably due to their use of whole blood rather than serum or plasma since the extraction methods in both studies were essentially the same. 48 FIGURE 1. — Mean levels of organocMorine pesticides in 1,000 serum samples 1 - 1 . ^„ - - "„ ,. .. .. 0, 0. Dale et al. (3) on the other hand, found plasma p,p'- DDE levels in a randomly chosen group of 20 persons of unspecified race in Georgia averaging 19.6 ppb and ranging from 3.9 to 41.6 ppb. They also reported an average of 17 ppb p,p'-DDT with a range from 2.4 to 49.0 ppb and dieldrin at a mean of 1.9 ppb and a range of 1.2 to 6.3 ppb. Our findings for mean levels of serum p,p'-DDE tend to approximate Dale's results, despite his use of a different method of extraction; however, the range for p,p'-DDE in the Idaho group was much greater than those re- ported by either Dale (2) or Davies et al. (4). Dale's mean levels for p,p"-DDT and dieldrin were greater than those found in Idaho residents, but the Idaho ranges were again much higher. Such variation seem? most likely due to the relatively small numbers of individuals sampled in the investigations cited, as well as differing geographical and ecological conditions such as the relative predominance of agriculture in Southern Idaho. Analysis of the three most commonly occurring residues as a function of sex of the person sampled showed several apparent differences (Fig. 2). Males had a higher Pesticides Monitoring Journal; FIGURE 2. — Sex differences in mean serum levels of p,p'-DDE, p,p' -DDT, and dieldrin FIGURE 3. — Age differences in mean serum levels of p,p'-DDE and p,p'-DDT 300 3!0 0-2! 0-15 mean serum level of p.p'-DDE (27.3 ppb) than females (18.8 ppb). Mean serum concentrations of p.p'-DDT, however, were similar in both sexes (4.9 ppb in males versus 4.6 ppb in females). Dieldrin, on the other hand, was more concentrated in women (1.3 ppb) than in men (0.3 ppb) as shown in Fig. 2. Dale reported a similar trend for p,p'-DDE, but males in his study had more p.p'-DDT and dieldrin than did females. Davies, using 23 men and 45 women, found p,p'-DDE levels in whole blood nearly equal for both sexes (8.3 and 7.9 ppb for men and women, respectively), but these mean values are both considerably less than those found in Idaho residents. Again, the use of so few individuals in the comparative studies makes meaningful correlation difficult. However, overall differences between the sexes in pesticide levels are probably influenced to a significant degree by the greater complexity of female hormonal mechanisms, differences in body fat deposition, etc. Serum pesticide levels also varied considerably among Idaho residents as a function of age (Fig. 3 and Table 1). The age of the individual was recorded for 782 serum samples, and the highest mean p,p'-DDE level (26 ppb) was found in persons 31-60 years of age. The 61- to 90- year-old individuals averaged a slightly lower level (24 ppb) while serums of persons in the 3- to 30-year age group averaged only 14.8 ppb. Lower p,p'-DDE levels in younger persons are even more apparent when the Vol. 4, No. 2, September 1970 3- to 30-year-old individuals are treated as a separate subgroup. Persons from 3 to 10 years of age averaged only 7.9 ppb, but this value increased steadily and dramatically to a high mean of 18.2 ppb in the 26- to 30-year-old subgroup. Mean concentrations, ranges, and percent occurrences of both p,p'-DDE and p,p'-DDT throughout individuals from 3 to 30 years of age are shown in Table 1. In comparing p,p'-DDE blood levels among persons of different age groups, Davies found that individuals 1-7 years of age had a mean of 8.4 ppb and a range of 2 to 17 ppb. Levels in those 18 years and older averaged 9.0 ppb and ranged from <1.0|to(55jppb. Our results for p.p'- DDE levels in very young persons (3 to 10 years) com- pare favorably with these findings, but persons older than age 10 showed higher means and ranges at all age levels, as well as a much more dramatic increase at each successive age group between age 1 1 and 30 (Fig. 3 and Table 1). Idahoans age 31-60 had considerably more p.p'-DDE than did those under age 30. Since the onset of widespread DDT usage was not until the early 1940's (/), many persons in the 3-30 age group lack comparable exposure time, thus supporting this finding. Idahoans in the 61- to 90-year category had slightly lower mean p.p'-DDE levels and ranges than those aged 31-60. While this difference is slight, it could possibly be attributed to dietary changes among older persons, as well as to their increased likelihood of undergoing chemotherapy, since persons currently taking drugs were not eliminated from this survey. The ability of 49 TABLE 1. — Percent occurrence, means, and ranges of p,p'-DDE and p,p'-DDT in serum of different age groups p,p'-DDE p,p'-DDT Age (YEARS) N Percent Mean Range Percent Mean Range OCCURRENCE (PPB) (PPB) occurrence (PPB) (PPB) 1 3-10 3 100.0 7.9 5.6- 10.9 66.6 2.1 0- 4.1 11-15 22 100.0 13.0 5.9- 28.8 81.8 3,0 0-20.3 16-20 96 99.0 14.9 0 - 47.1 76.0 3.0 0-15.5 21-25 94 100.0 17.1 1.3- 62.4 89.4 3.2 0-14.7 26-30 59 100.0 18.2 1.0- 89.8 83.1 4.7 0-29.8 Total (3-30) 274 99.6 14.8 0 - 89.8 82.5 3.5 0-29.8 31-60 230 100.0 26.0 10.0-102.0 89.6 4.5 0^2.0 61-90 278 100.0 24.0 1.0-133.0 85.3 4.8 0-29.0 certain drugs, e.g., phenobarbital, to induce hepatic microsomal enzyme systems responsible for the meta- bolism of various organochlorine pesticides (6) should certainly not be discounted as a possible contributing factor. Unlike p,p'-DDE, p,p'-DDT mean levels did not show such dramatic differences with respect to age. Several trends are nonetheless apparent. As with p,p'-DDE, /7,/?'-DDT residue averages were lowest (3.5 ppb) in the 3-to 30-year age group (Fig. 3 and Table 1). Highest mean p,p'-DUT levels (4.8 ppb) were found in persons 61-90 years of age, while individuals of 31-60 years had a slightly lower level (4.5 ppb) . Within the 3- to 30-year age group, those from 3-10 years of age had the lowest mean level of p,p'-DDT (2.1 ppb); the II- to 20-year- olds had an average level of 3.0 ppb; and average levels of p,p'-DDT in those 21-25 years of age reached 3.2 ppb. This mean then increased to 4.7 ppb in persons aged 26-30. Since p,p'-DDT is metabolized in the body to p,p'-DDE, such a comparative lack of clear-cut p.p'- DDT increase in older persons would be expected — the more accurate index of exposure being p,p'-DDE. Conclusions As was expected, the Idaho residents sampled in this study do not differ basically from persons in other areas in the accumulation of serum organochlorine pesticide residues. Certain pesticide level variations noted among Idahoans when compared to other study groups are most likely due to regional differences in pesticide usage and to the relatively large number of persons sampled. Sex differences noted in pesticide levels are probably influenced by hormonal mechanisms, as was concluded in similar studies. In persons under 30 years of age, mean levels of p,p'-DDE and, to a lesser degree, p.p'- DDT seem to be increasing. From this, one could as- sume that as age increases, pesticide retention would reach a plateau similar to the levels found in older persons. The Idaho Community Studies Pesticides Project is supported by Contract No. PH 21-2008 with the Division of Pesticide Community Studies, Office of Pesticides and Product Safety. Bureau of Foods. Pesticides, and Product Safely, Food and Drug Administration, Pub- lic Health Service, Department of Health, Education and Welfare, Chamblee, Ga. See Appendix for chemical names of compounds mentioned in this paper. LITERATURE CITED (7) Chichester, C. O., ed., 1965. Research in pesticides. Academic Press, New York. (2) Dale. W. E., A. Ciirley, and C. Cueto. 1966. Hexane ex- tractable chlorinated insecticides in human blood. Life Sci. 5:47-54. (3) Dale. W. £,, A. Curley, and W. J. Hayes. Jr. 1967. De- termination of chlorinated insecticides in human blood. Ind. Med. Surg. 36(4):275-280. (4) Davies, J. E., W. F. Edmundson, A. Maceo, A. Barquet, and J. Cassady. 1969. An epidemiologic application of the study of DDE levels in whole blood. Amer. J. Public Health 59(3):435-441. (5) Edmundson, W. F.. J. E. Davies, G. A. Nachman, and R. L. Roelh. 1969. P,p'-DDT and p,p'-DDE in blood samples of occupationally exposed workers. Public Health Rep. 84(l):53-58. (6) Williams, C. H. 1969. Drug-pesticide interactions. FDA Papers 3(7): 14- 17. 50 Pesticides Monitoring Journal RESIDUES IN FISH, WILDLIFE AND ESTUARIES Significance of DDT Residues From the Estuary Near Penascola, Fla.^ David J. Hansen and Alfred J. Wilson, Jr. ABSTRACT Pesticide residues in fishes from the estuary near Pensacola, Fla., monitored from April 1964 to November 1965. are compared with residues in fishes exposed to DDT in the labo- ratory. DDT in fish exposed to 0.1 ppb p.p'-DDT for 5 weeks failed to increase after the second weel<, when maximum concentrations reached 38,000 times that in the test water. Loss of DDT from these fish was slow, 78% -87% in 8 weeks. The amounts of DDD or DDE in fish did not increase either during or after exposure. Residues in fish from the estuary rarely exceeded 0.1 ppm except in those collected from the lower estuary in the sum- mer and fall when the amount of DDT and its metabolites reached 1.3 ppm. Fish from the lower estuary had more DDT and less DDD and DDE than fish from the upper estuary. The DDT content in lower estuarine fish increased in July, August, and September. One source of DDT was a county-sponsored spray program centered in the lower estu- ary in July and August. Introduction Both fish and shellfish accumulate pesticides from the water. Fish build up residues gradually and store them mainly in the fat. In shellfish, however, these chemicals are more generally distributed in the tissues and are concentrated continually in proportion to the level of environmental pollution. For example, oysters exposed to concentrations of DDT as low as 0.1 ppb in the sur- rounding water may concentrate up to 7 ppm in their tissues in about a month (1). This biological magnifi- cation is indicative of the extent to which trace pol- lutants in the environment may be concentrated and ' Contribution No. 99. Bureau of Commercial Fisheries Center for Estuarine and Menhaden Research, Pesticide Field Station, Gulf Breeze, Fla. 32561. Vol. 4, No. 2, September 1970 enter the food web within a relatively short time posing a threat to the reproduction, survival, or marketability of fish or shellfish. In this study we determined the concentrations of organochlorine pesticide residues in several species of fish collected throughout the year at different locations in the estuary near Pensacola, Fla., and the rates of storage and retention of DDT in fish in the laboratory. Methods and Procedures Fish were collected with a 5-meter otter trawl of 12-mm bar mesh at four locations within the estuary (Fig. 1) at about monthly intervals from April 1964 to Novem- ber 1965. Pinfish, Lagodon rhomboides. were collected in the lower estuary (locations A and B) and Atlantic croaker, Micropogon undulatus. were caught in the upper estuary (locations C and D). An additional non- pelagic species — either pigfish, Orthopristis chrysopterus; silver perch, Bairdiella chrysiira; or spot, Leiostomus xanthuriis — was also captured at each location. In order to assure a representative sample, when possible, a composite sample of at least 10 fish of each age group, species, and location was collected and frozen prior to pesticide analysis. This was deemed necessary because of the large variations between individuals. For example, residues of DDT and its metabolites in seven individual yearling pinfish collected from location A in August 1964 ranged from 0.5 to 13.7 ppm (mg/kg, wet weight) and from 0.4 to 1.1 ppm in nine fish caught in September 1965. Although dieldrin, endrin, or BHC was found in some samples at concentrations up to 0.02 ppm, DDT and its metabolites were the predominant pesticides. 51 FIGURE 1. — Collection sites in the estuary near Pensacola, Fla., 1964-1965 GULF OF MEXICO We investigated the rates at which fish remove DDT from water under controlled conditions in the labora- tory to aid in understanding results obtained in the field. Initially, 100 pinfish or 50 croakers, average standard length 25 and 50 mm, respectively, were used in each treatment and "control" group. Each species was exposed to p.p'-DDT at 1.0 ppb (/ig/ liter) for 2 weeks or 0.1 ppb for 5 weeks, according to the flowing water bioassay technique described by Lowe (2). Fish exposed to 0.1 ppb DDT were placed in pesticide-free water for 8 additional weeks to establish flushing rates. All test fish were fed daily on ground fish from this estuary with the naturally occurring DDT content pre- viously determined. At selected intervals, five individuals were removed from each group and frozen prior to pesticide analysis. Fish from the field and laboratory were analyzed for residues by the same procedures. Determination of pesticide residues was made on pooled samples of fish which had been thawed, homogenized, and mixed well prior to analysis. A 30-g aliquot was taken from the ground composite, mixed with anhydrous sodium sulfate in a blender, and extracted for 4 hours with petroleum ether in a Soxhlet apparatus. Extracts were concentrated and transferred to 250-ml separatory funnels. The extracts were diluted to 25 ml with f)etroleum ether and partitioned with two 50-ml portions of acetonitrile 52 previously saturated with petroleum ether. The acetoni- trile was evaporated just to dryness and the residue eluted from a Florisil column (5). The sample was then identified and quantitated by electron capture gas chro- matography. Three columns of different polarity (DC- 200, QF-1, and mixed DC-200 and QF-1) were used to confirm identification. Operating parameters on Var- ian Aerograph 610D gas chromatographs were as fol- lows: Columns: 5' x Vs" O.D., Pyrex glass, packed with 3% DC-200, 5% QF-1, and a 1:1 ratio of 3% DC-200 and QF-1, all on 80/100 mesh Gas Chrom Q Temperature: Detector 210 C Injector 210 C Oven 190 C Carrier Gas: Prepurified nitrogen at a flow rate of 40 ml/minute. A few samples were analyzed using thin-layer chroma- tography. The lower limit of detectability was 0.01 ppm. Laboratory tests conducted during the sampling period gave the following recovery rates: p,p'-DDE, 80%- 85%; p,p'-DDD (TDE), 92%-95%; p,p'-DDT, 91%- 95%. Data in this report do not include a correction factor for percent recovery. Polychlorinated biphenyl (PCB) compounds were not detected in these samples. Results FIELD STUDIES Fish from the lower estuary contained the highest pesticide residues in the late summer and fall, up to 1.3 ppm DDT and metabolites; whereas, residues in fish from the upper estuary varied monthly but were generally less than 0.1 ppm (Table 1). The ratio of DDT to its metabolites was greatest in fish from the lower estuary, particularly those from collection site A, in July, August, and September. In October and November the relative amounts of DDD and DDE in fish from the lower estuary increased. In fish from the upper estuary there was no pronounced seasonal shift in the total pesticide content nor in the amount of DDT, which was usually less than found in fish from the lower estuary. Butler (4) monitored pollution in this estuary and identified a similar increase in DDT residues in plankton, mussels, and oysters in the summer and fall in the lower estuary. The main source of the pesticide was, therefore, near the lower estuary rather than the rivers entering the upper estuary. The results of our investigations led to the discovery of a county- Pesticides Monitoring Journal sponsored program that used DDT to control the larvae of the dogfly, Stomoxys calcitrans, a biting insect. Dur- ing July and August, DDT spraying was evidently con- centrated near populated areas at the lower estuary. Fish from the lower estuary in their second year of life (age group I) contained more DDT and its metab- olites than fish in their first year of life (age group 0). Pesticide residues in the older fish were highest in 93% of the 41 samples, where both age groups were caught in the same month ()(' = 29.88, df:l). The average pesticide content (DDT + DDD + DDE) in ppm of fish of different age groups was as follows: Sampling location A Pinfish Pigfish Sampling location B Pinfish Pigfish Age Group 0 0.17 0.48 0.11 0.48 0.12 0.25 0.05 0.19 TABLE 1. — Pesticide content (mg/kg, wet weight — whole fish) in five species of fish from the estuary near Pensacola, Fla. 1964-1965 Month Collected Ace Group ■ DDT AND Met. No. OF Fish IN Composite Sample Pinfish May A O I .03 .03 .01 .06 .02 .03 .06 .12 15 6 B O I .03 .06 .03 .06 .02 .03 .08 .15 40 6 Atlantic croaker C o .06 .06 .09 .21 32 D o .01 .02 .01 .04 10 Spot o .01 .01 .02 .04 20 Pinfish June A o I .01 .16 .01 .19 .01 .05 .03 .40 10 10 Pigfish o I .01 .06 .01 .15 .01 .06 .03 .27 9 6 Silver perch o I .01 .05 <.01 .05 .01 .08 .02 .18 14 10 Spot o .01 .01 .01 .03 10 Pinfish B o I .01 .06 .01 .06 .01 .06 .03 .18 10 10 Atlantic croaker C o .01 .01 .01 .03 8 Spot o <.01 <.01 <.01 <.01 12 Atlantic croaker D o <.01 <.01 <.01 <.01 8 Spot o <.01 <.01 <.01 <.01 6 Pinfish July A o I .06 .62 .02 .37 .01 .12 .09 1.11 10 10 Pigfish o .13 .02 .03 .18 10 Silver perch o I .17 .63 .05 .02 .12 .13 .34 .78 10 10 Spot o .03 .01 .01 .05 10 Pinfish B o I .06 .12 .08 .09 .02 .06 .16 .27 10 5 Pigfish o I .03 .12 .03 .08 .06 .11 .12 .31 10 3 Silver perch o I .03 .02 .02 .02 .02 .02 .07 .06 10 10 Atlantic croaker C o .01 .06 .06 .13 10 Spot o .01 .01 .01 .03 10 Atlantic croaker D o .02 .02 .03 .07 10 Spot o .01 .01 .02 .04 10 Pinfish Aug. A o I .38 .54 .13 .27 .03 .08 .54 .89 10 10 Pigfish o I .11 .57 .02 .42 .03 .08 .16 1.07 10 10 Pinfish B o I .14 .10 .08 .10 .03 .04 .25 .24 10 5 Silver perch o I .05 .10 .03 .07 .04 .04 .12 .21 10 10 Atlantic croaker C o .02 .05 .03 .10 10 Spot o .02 .01 .02 .05 10 Atlantic croaker D o .02 .05 .05 .12 10 Spot o .01 .01 .01 .03 10 Vol. 4, No. 2, September 1970 53 TABLE 1.- —Pesticide content (mg/kg, wet weigh from the estuary near Pensacola, Fla. — whole fish 1964-1965— in five species of fish Continued SPEcms MONTH Collected Location Ace Group i DDT DDD DDE DDT AND Met. No. OF Fish IN Composite Sample 1964 — Continued Pinflsh Sept. A O I .25 .58 .18 .25 .03 .09 .47 .92 10 8 Pigfish O I .08 .26 .05 .43 .03 .09 .16 .78 10 2 Pinfish B o I .25 .31 .18 .25 .04 .09 .47 .65 10 4 SUver perch o .10 .05 .02 .17 10 AUantic croaker C o .03 .04 .04 .11 6 Spot o .02 .02 .01 .05 10 Atlantic croaker D o .02 .04 .03 .09 10 Spot o .01 .02 .02 .05 10 Pinfish Oct. A o .19 .16 .05 .40 91 Pigfish o I .14 .16 .14 .26 .06 .08 .34 .50 10 1 Pinfish B o .15 .09 .03 .27 10 Pigfish o .11 .08 .05 .24 10 Spot C o .03 <.01 .05 .08 10 Spot D o .02 .01 .04 .07 10 Pinfish Nov. A o .35 .47 .12 .94 10 Silver perch o .07 .09 .07 .23 10 Pinfish B o .40 .27 .11 .78 4 Silver perch o .14 .14 .10 .38 5 Atlantic croaker D o .02 .02 .03 .07 31 Spot o .01 .03 .07 .11 10 Atlantic croaker Jan. D O .02 .01 .03 .06 45 Spot Feb. A O .15 .12 .12 .39 10 Spot C O .04 <.01 .04 .08 20 Atlantic croaker D O .03 <.01 .04 .07 10 Spot March A O .05 .10 .02 .17 9 Spot B O .09 .09 .07 .25 10 Atlantic croaker C O .04 <.01 .04 .08 10 Spot O .02 <.01 .04 .06 10 Pinfish April A O I .06 .04 .06 .05 .03 .02 .15 .11 20 10 Spot o I .02 .03 .02 .05 .02 .02 .06 .10 10 10 Pinfish B o I .03 .07 .02 .11 .02 .04 .07 .22 20 9 Spot o I .01 .02 .01 .04 .01 .02 .03 .08 9 3 Atlantic croaker C o .01 .04 .02 .07 10 Spot o .01 .01 .02 .04 10 Atlantic croaker D o .02 .02 .02 .06 10 Spot o .03 .03 .02 .08 10 Pinflsh May A o I .02 .06 .01 .05 .01 .03 .04 .14 10 10 Pigfish o I .01 .11 .01 .22 .01 .07 .03 .40 10 10 Pinfish B o I .01 .07 .01 .07 .01 .04 .03 .18 10 10 Pigfish o 1 .01 .06 .01 .08 .01 .04 .03 .18 10 10 Atlantic croaker C o .01 .02 .02 .05 10 Spot o .02 .02 .03 .07 10 Atlantic croaker D o .02 .02 .01 .05 10 Spot o .01 .02 .01 .04 10 54 Pesticides Monitoring Journal TABLE 1. — Pesticide content (mg/kg, wet weight — whole fish) in five species of fish from the estuary near Pensacola, Fla. 1964-1965 — Continued Month Collected DDT AND Met. No. OF Fish IN Composite Sample 1 965 — Continued Pinfish June A O I .02 .07 .01 .09 .01 .03 .04 .19 10 10 Pigfish O I .03 .10 .01 .14 .01 .04 .05 .28 10 10 Pinfish B o I .01 .07 .01 .06 <.01 .03 .02 .16 10 10 Pigfish o I .01 .06 .01 .09 .01 .04 .03 .19 10 10 Atlantic croaker C o .01 .01 .01 .03 10 Spot o .01 .01 .01 .03 10 Atlantic croaker D o .01 .01 .01 .03 10 Spot o .01 .02 .01 .04 10 Pinfish July A o I .01 .09 .01 .06 <.01 .04 .02 .19 10 10 Pigfish o I .02 .10 .01 .14 .01 .05 .04 .29 10 10 Pinfish B o I .01 .07 <.01 .06 <.01 .03 .01 .16 10 10 Pigfish o I .01 .02 <.0I .04 <.01 .02 .01 .08 10 10 Atlantic croaker C o .01 .02 .01 .04 10 Spot o .02 .03 .02 .07 10 Atlantic croaker D o .01 .03 .01 .05 10 Spot o .02 .04 .02 .08 10 Pinfish Aug. A I .15 .13 .05 .33 10 Pigfish I .16 .18 .07 .41 10 Pinfish B I .09 .07 .03 .19 10 Pigfish I .10 .12 .05 .27 6 Atlantic croaker C o <.01 <.01 <.01 <.01 10 Spot o <.01 <.01 .02 .02 10 Atlantic croaker D o .01 .02 .01 .04 10 Spot o .01 .02 .02 .05 10 Pinfish Sept. A o I .18 .13 .05 .13 .02 .04 .25 .30 10 3 Pigfish o I .02 .06 .04 .12 .04 .04 .10 .22 9 5 Pinfish B o I .03 .14 .02 .14 .01 .05 .06 .33 10 10 Pigfish o I .02 .06 .01 .09 <.01 .03 .03 .18 10 2 Spot C o .02 .04 .02 .08 10 Pinfish Oct. A o I .07 .32 .08 .48 .03 .11 .18 .91 10 10 Pigfish o .16 .16 .08 .40 10 Pinfish B o .09 .04 .01 .14 10 Pigfish o I .02 .08 .05 .10 .02 .04 .09 .22 10 3 Pinfish Nov. A o .39 .33 .18 .90 10 Silver perch B o .43 .51 .32 1.26 7 ' O = Fish in their first year of life. I = Fish in their second year of life. Vol. 4, No. 2, September 1970 55 TABLE 2. — Levels of DDT, DDD, and DDE in ppm wet weight in fish food, control fish, and fish exposed to 0.1 and 1.0 ppb p,p'-DDT (each sample consisted of at least five fish) Exposure Time Concen- tration After flushing Week 3-5 SPECffiS IN Test Start Day 3 Week 1 Week 2 (Weekly g weeks Watek (PPB) AVERAGE) DDT DDD DDE DDT DDD DDE DDT DDD DDE DDT DDD DDE DDT DDD DDE DDT DDD DDE Food .08 .07 .02 .08 .07 .02 .08 .07 .02 .09 .08 .05 .10 .08 .06 .08 .09 .05 Pinfish Control .01 <.01 .01 .06 .04 .03 .06 .04 .03 .17 .13 .12 .20 .15 .14 .15 .16 .10 Atlantic croaker Control .01 .01 .01 .08 .06 .05 .09 .07 .07 _ _ _ .13 .10 .09 .23 .17 .17 Pinfish 0.1 .01 <.01 .01 .57 .05 .04 .85 .07 .06 3.8 .11 .09 3.3 .26 .17 .42 .15 .08 Atlantic croaker 0.1 .01 .01 .01 .33 .07 .07 .80 .08 .07 1.1 .08 .07 1.4 .11 .12 .31 .23 .16 Pinfish 1.0 .01 <.01 .01 2.8 .08 .06 6.9 .17 .07 10.6 .28 .14 — — — 16.3 .52 .25 Atlantic croaker 1.0 .01 .01 .01 2.3 .06 .05 6.9 .13 .12 12.0 .09 .08 210.0 .22 .21 - - - 1 Residue 2 Residue after 4 weeks of flushing, after 3 weeks of exposure. LABORATORY STUDIES Results of laboratory studies on uptake and retention of DDT in water by pinfish and Atlantic croakers showed that residues in fish exposed to 0.1 ppb DDT reached a maximum in 2 weeks and remained nearly constant until the exposure ended 3 weeks later (Table 2). Residues in these fish, depending on species and exposure, represent a maximum concentration of DDT from 10,000 to 38,000 times that in the test water. Contrary to what might be expected, there was no in- crease in the amounts of DDD or DDE in test fish as compared to control fish. We have no explanation for this phenomenon. The concentration of DDT and metabolites in control fish increased from 0.02 to 0.57 ppm in response to the DDT content of their food. DDT was lost slowly from fish previously exposed to 0.1 ppb (Table 2). After 8 weeks in pesticide-free water, the loss of DDT from pinfish was 87% and from Atlantic croakers 78%. The pesticide content at that time was similar to that in control fish. This loss was not accompanied by an increase in the concentration of DDD or DDE. After 4 weeks in pesticide-free water, pinfish, previously exposed to 1.0 ppb DDT for 2 weeks, had lost 41% of the accumulated DDT. These tests further the understanding of our field study. In laboratory tests, DDT was rapidly stored without increase in DDD or DDE while fish from the estuary usually had as much DDD and DDE as DDT. This indicates that fish from this estuary obtained the pesti- cide after it had been metabolized and passed through the food web. Discussion DDT residues in different species of fish can be compared and used to identify the source of a pollutant, providing the rate of uptake of DDT by each sp>ecies is known and the fish remain in one location long enough for their DDT content to reflect the magnitude of contam- 56 ination in that particular area. Pinfish and Atlantic croakers contained similar residues when eating the same food or after exposure to 1 ppb DDT; however, when these two species were exposed to 0.1 ppb DDT, pinfish stored 2.4 times as much DDT as croakers. This difference in storage traits must be considered when comparing pesticide contamination in different areas represented by different species. Benthic fishes, like the five species used in this study, usually remain in one location while f>elagic fishes do not. Pesticide residues in benthic fishes would, therefore, be better indicators of contamination in a particular area than the residues in pelagic fishes. The migrations of both pelagic and benthic fishes trans- port DDT in and out of the estuary. Atlantic croakers and pinfish migrate to the Gulf of Mexico in the late fall and rarely return to this estuary (5). Butler (6) stated that this removed each year about Vz lb of DDT from this estuary. See Appendix for chemical names of compounds mentioned in this paper. LITERATURE CITED (7) Butler, Philip A. 1966. The problem of pesticides in estuaries. Trans. Amer. Fish. See. Spec. Publ. No. 3;1 10- 115. (2) Lowe, J. I. 1964. Chronic exposure of spot, Leiostomiis .xanlhurus, to sublethal concentrations of toxaphene in seawater. Trans. Amer. Fish. Soc. 93(4):396-399. (.?) Mills, Paul A., J. H. Onley. and R. A. Gaither. 1963. Rapid method for chlorinated pesticide residues in non- fatty foods. J. Ass. Offic, Agr. Chem. 46(2):186-191. (4) Butler. Philip A. 1968. Fixation of DDT in estuaries. Trans. 31st N. Amer. Wildlife and Natur. Resources Conf. p. 184-189. (5) Hansen, David J. 1970. Food, growth, migration, repro- duction, and abundance of pinfish, Lagodon rhomboides, and Atlantic croaker. Micropogon undulalus, near Pen- sacola, Florida. 1963-1965. U. S. Fish Wildlife Serv., Fish. Bull. 68(1):135-146. (6) Butler, Philip A. 1968. Pesticides in the estuary. Proc. Marsh and Estuarj' Manage. Symp. p. 120-124. Pesticides Monitoring Journ.\l Chlorinated Hydrocarbon Pesticides in Representative Fishes of Southern Arizona^ Donald W. Johnson" and Sam Lew' ABSTRACT Chlorinated hydrocarbon residues are reported for represen- tative fishes of the lower Colorado River Basin. While most residues were in the ppb range, DDT and metabolites to 187.5 ppm were found. Toxaphene was a common contami- nant at levels as great as 172.9 ppm. Dieldrin was also present in significant concentrations. DDE was the principal residue found. These findings should prove of value in ap- praising the effect of the 1969 ban on the use of DDT in Arizona. They appear sufficient to warrant concern for fish and fish-consuming populations. Introduction A 1965 estimate of cotton land in southern Arizona was 345,000 acres. Pesticide application for each acre averaged 7.59 lb of toxaphene, 1.35-4.00 lb of DDT, 1.20 lb of endrin, 0.58-1.02 lb of endosulfan and un- determinable quantities of aldrin, dieldrin, BHC, hepta- chlor, and chlordane. DDT residue levels in starlings from the Phoenix area, reflecting DDT contamination in southern Arizona, are the Nation's highest — 20 ppm (/); the Gila River appears to be the most DDT- burdened stream of 20 sampled in the Western United States (2). Fish kills nationwide for 1967 were up 21% from 1966, and insecticides were involved in 16% of the kills for which reports were submitted to the Federal Water Quality Administration. No cause was designated for Arizona kills, and eight States including the cotton- growing State of Mississippi submitted no report (3). DDT use in Arizona for 1967 increased by a factor 1 Work completed at Zoology Department, Arizona State tjniversity, Tempe, Ariz. 85281. ' Present adress. University of California (Berkeley), Bodega Marine Laboratories. Bodega Bay, Calif. 94923. ' Pesticide Laboratory, Arizona State Health Department, Phoenix, Ariz. 85007. Vol. 4, No. 2, September 1970 of five from 1965. In 1967, three times as much toxa- phene was applied as in 1965. From 1967 to 1968, a period marked by legislative opposition to DDT, use of less-studied chlorinated hydrocarbons increased (Strobane and Telone 8 and 7 times, respectively); the use of the organic phosphate parathion increased 1 1 times (4). Although this may provide some relief for ectotherms from acute toxicity, organic phosphates are generally more toxic to endotherms (5). A fluctuating pattern of pesticide use producing continuous change in residue levels further compounds the difficult task of determining the significance of sublethal residues in fish tissues to both fish and human populations (5). The residues reported here represent reference data from a preliminary survey of some chlorinated pesti- cides found in representative fishes of the lower Colo- rado River Basin. The carp (Cyprinus carpio) is probably the most abundant and widely distributed fish in southern Arizona. Channel catfish {Ictahiriis punc- tatits) and green sunfish (Lepoinis cyanellus) are also abundant and occur throughout the Lower Colorado River Basin. Threadfin shad (Dorosoma petenense) is a common forage species heavily utilized by carnivorous species including the channel catfish and green sunfish. Tilapia (Tilapia mossamhica) is an herbivore that has been established in the lower Colorado and the irriga- tion canals of the Buckeye and Yuma area where it contributes to the sport fishery. The Sonoran {Catos- tomus insignis) and Gila (Pantosteus clarki) suckers are bottom feeders and abundant in Phoenix area canals. The last fish for which data are presented is the striped mullet (Mugil cephalus). An amphidromous bottom feeder characterized by an exceptionally high fat con- tent, it is found in the Gulf of California and the waters of the Colorado River and canals below Imperial Dam. Collection sites and dates are shown in Fig, 1. 57 FIGURE 1. — Southern Arizona showing origin of pesticide residues presented — collection sites and dates are listed Californi 1. Picacho Reservoir (8-28-66) 2. Mesa CanaK 11-14-66) 3. Tempe CanaK 1-2 1-67) 4. Buckeye Canal (12-10-65) 5. Martinez Lake (7-24-66) 6. Mitry Lake (8-10-66) 7. Imperial Dam (12-6-68) 8. Yuma Canal (3-16-66) 9. Summerton Canal (9-25-66) 10. El Golfo de Santa Clara (6-7-69) Materials and Methods Fishes were collected by gill net or seine. All tissues were frozen prior to extraction for analysis by gas chromatography. Extraction and chromatographic an- alysis were completed at the Arizona State Health Laboratory in Phoenix by standard methods (6). Five- gram samples were used for muscle, liver, and low-fat organs. Half-gram samples of fatty tissue were ex- tracted. Elution for endrin and dieldrin extraction was not completed for all tissues analyzed. The Aerograph Pestilyzer Model 680 gas chromatograph with Vs" x 5' glass columns was used. The nonpolar column was packed with 2% Dow 11 on Chromosorb G, the polar column with 2% QF-1 on Chromosorb G. Nitrogen at 20 psi was the carrier gas at an oven temperature of 180 C. Toxaphene was identified by the location of the "y" peak and quantified through its height. Standard peak height curves of v, w, x, and y were obtained by injecting a series of toxaphene standards from 0.5-30 ng. The v, s. and x peak heights for an un- known are extrapolated from the standard graphs based on "y" peak height. Differences from the extrapolated V, w, and x values indicate the quantity of DDE, TDE, and DDT, respectively. The most abundant of the DDT compounds was p,p'-DDE, and there was no significant interference in determinations by polychlorinated bi- phenyls (PCB's). Risebrough, Reiche, and Olcott have recently published a similar conclusion (7). Confusion of toxaphene and PCB's is considered unlikely. During the period of the study use of PCB's in Arizona was unreported, while DDT and toxaphene represented 41% and 40%, respectively, of all agricultural chlorinated 58 hydrocarbons applied. High-residue specimens in agri- cultural areas reflected the relative presence of DDT compounds and toxaphene as environmental contam- inants. Toxaphene chromatograms produced recogniz- able characteristic peaks on a mountain-like profile. PCB chromatograms do not exhibit this profile and when compared to toxaphene were distinctive. Sensitivity levels were based on check samples which were spiked with DDT, TDE, and DDE at two levels, 0.2 ppm and 0.02 ppm. Recoveries averaged 88%, 85%, and 91% for the higher level and 88%, 86%, and 84% for the lower level. Values reported here were not corrected after recoveries were calculated. Recoveries for dieldrin and toxaphene were not determined. The only confirma- tion technique used was comparing the retention time of the unknown/ known ratios from polar (QF-1) and nonpolar (QV-1) columns. Residts and Discussion Residue data are shown by species and tissue in Table 1. The following conclusions can be made from these data: 1. In 1966, it was not uncommon for fishes in southern Arizona to far exceed the interim guide- line for DDT and its metabolites of 5 ppm estab- lished for fish in 1969 by the FDA (8). 2. The 5-ppm level of DDT contamination was exceeded in the edible flesh by a factor as great as 23 in carp, 8 in channel catfish, 4 in tilapia, 2 in suckers, and 37 in the striped mullet. 3. In addition to DDT, toxaphene and dieldrin were present in some fish in extremely high con- centrations together with trace amounts of endrin. 4. Most DDT either enters the tissue as DDE or is rapidly metabolized to that form. 5. DDT and its metabolites are found concentrated in liver tissue. 6. Concentration may be closely correlated to fat content of the tissues. Feeding habits and age may explain the variation in residue levels between species within a body of water. Local agricultural practices can explain seasonal and geographic variation. The limitation of small sample sizes and the absence of seasonal collections and residue analyses from all species present at each site make interpretation highly speculative. The data, however, have provided a degree of support for the above con- clusions. In addition, they provide reference values for comparison with those which will be reported subsequent to the current Arizona ban on the application of DDT. A primary goal of the National Pesticide Monitoring Program is the location of possible problem areas where PESTicroES Monitoring Journal concentration levels justify concern. The only sur- veillance station in the Southwest is located above Imperial Dam on the Colorado River (8). Agricultural waste water from Arizona's cropland enters the system below this station. The 1966 data on pesticide residues in fish resulting from agricultural contamination of southern Arizona waters are sufficient to warrant concern for both fish and fish-consuming populations. A com- prehensive surveillance program should not overlook this source of environmental contamination (5). A cknowledgments The staff of the Arizona State Health Laboratory pro- vided analytical support and full cooperation in carrying out this study. Arizona Game and Fish personnel cooperated in the collection of fish as did E. McClendon and S. Silver. See Appendix for chemical paper. of compounds mentioned in this UTERATURE CITED (7) Martin, W. E. 1969. Organochlorine insecticide residues in starlings. Pesticides Monit. J. 3(2): 102-114. t2) Manigold, D. B. and J. A. Schulze. J 969. Pesticides in selected western streams — a progress report. Pesticides Monit. J. 3(2):124-135. (3) Federal Water Pollution Control Administration. 1968. Pollution caused fish kills— 1967. Fed. Water PoUut. Contr. Admin. Eighth Annu. Rep. CWA-7, p. 16. {4) Roan, C. C, D P. Morgan, C. H. Kreader, and L. Moore. 1969. Pesticide use in Arizona as shown by sales. Progr. Agr. Ariz. 21:14-15. (5) Johnson, D. W. 1968. Pesticides and fishes — a review of selected literature. Trans. Amer. Fish. Soc. 97:398- 424. (6) Food and Drug Administration. 1968. Pesticide analyt- ical manual, U. S. Dep. Health, Educ. and Welfare. Vol. 1, Sect. 211.13A, 211.15, 302.1, 302.13b, 302.14, 302.15, 302.4a, 302.4e, and 31 1.01-311.01J. (7) Risebrough. R. W.. P. Reiche, and H. S. Olcott. 1969. Current progress in the determination of the polychlori- nated biphenyls. Bull. Environ. Contamination Toxicol. 4:192-201. (8) Henderson. C. W. L. Johnson, and A. Inglis. 1969. Organochlorine insecticide residues in fish (National Pesticide Monitoring Program). Pesticides Monit. J. 3(3):145-171. TABLE 1. — Pesticide residues in fish, by species and tissue Standard Residues in PPM (Wet Weight) i Speomen Tissue Length (CM) Collection DDE DDD DDT DiELDRIN TOXAPHENE SrrE Cyprinus carpio M-2 2 "muscle" 90.0 16.9 8.5 2.1 8 M-2 red muscle 0.15 0.06 0.03 P.4 red muscle 16 0.30 0.08 0.16 ND P-5 red muscle 16 0.22 0.10 0.15 M-2 white muscle 0.01 0.01 0.01 P-5 white muscle 16 0.05 0.02 0.05 P-4 white muscle 16 0.02 0.01 0.02 ND M-1 scraped skin 0.22 0.07 0.03 M-0 scraped skin 0.13 0.09 0.03 P-5 scraped skin 16 0.33 0.15 0.22 P-+ scraped skin 16 0.28 0.07 0.21 0.01 2 fat 48.0 15.8 13.0 1.14 50.0 2 fat 153.0 24.8 7.2 0.5 M-0 fat 1.15 0.70 0.18 M-3 blood 0.03 0.01 0.01 0.002 M-7 blood ND 0.01 0.01 0.002 P-1 to P-6 blood 15-21 0.03 0.01 0.02 0.002 P-5 gills 16 0.17 0.08 0.07 0.45 P-4 giUs 16 0.22 0.09 0.12 0.42 P-5 kidney 16 0.69 0.28 0.48 P-4 kidney 16 0.48 0.12 0.54 M-0 liver 0.20 0.11 0.08 P-5 liver 16 1.25 0.83 0.38 P-4 liver 16 1.22 0.48 0.18 M-0 ovaries 0.02 0.01 0.01 M-1 ovaries 0.02 0.0 1 0.01 P-S ovaries 16 0.07 0.05 0.06 P-4 ovaries 16 0.11 0.04 0.05 ND M-2 intestine contents ND ND 0.002 0.008 Vol. 4, No. 2, September 1970 59 TABLE 1. — Pesticide residues in fish, by species and tissue — Continued Specimen Tissue Standard Length (CM) Residues in ppm (Wet Weight) 1 Collection Sfte DDE DDD DDT Dieldrin TOXAPHENE Channel catfish Ictaturus punctatus _2 "muscle" 18.4 6.8 13.7 0.06 6.8 8 P-U red muscle 23 1.87 0.60 0.66 0.55 1 P-11 white muscle 23 0.04 0.02 0.03 ND 1 M-8 white muscle 0.17 0.05 0.04 5 M-9 white muscle 0.27 0.05 0.06 5 S-33 skin and muscle 14.5 1.54 0.03 0.13 9 p-n scraped skin 23 1.00 0.28 0.33 1.32 1 M-9 scraped skin 0.51 0.13 0.23 5 2 fat 38.0 14.0 25.0 0.6 8.2 8 P-U fat 23 34.78 10.42 11.16 11.38 1 M-9 fat 7.23 0.96 1.84 5 P-11,I2 blood 26,28 0.10 0.05 0.04 0.003 P-Il gills 23 0.48 0.17 0.23 0.35 1 P-11 kidney 24 0.81 0.32 0.32 1 . M-8 kidney 0.30 0.12 0.13 5 M-9 kidney 0.19 0.04 0.06 ND 5 2 liver 19.8 8.4 9.0 1.2 8 P-U liver 24 0.24 0.11 0.05 1 M-9 liver 0.09 0.03 0.02 5 M-8 intestine 0.06 0.02 0.02 ND 5 P-U intestine contents 24 0.07 0.01 0.02 ND 1 S-33 viscera 14.5 0.63 0.02 0.05 8 Green sunfish Lepomis cyanellus P-34 white muscle 13 0.08 0.01 0,04 P-35 white muscle 14 0.06 0.01 0.03 P-34 scraped skin 13 0.45 0.02 0.10 P-35 scraped skin 14 0,15 0.02 0.06 P-31 gills 13 0.26 0.03 0.07 P-35 gills 14 O.U 0.02 0.10 P-34 liver 13 3.86 0.68 0.15 P-35 liver 14 3.07 1.03 0.39 P-35 ovaries 14 6.59 1.41 2.46 P-34 testes 13 1.56 0.12 0.44 P-34 caecae 13 1.20 0.13 0.09 P-35 caecae 14 1.10 0.34 0.17 Threadfin shad Dorosoma petenensis P-29 whole fish 12 0.58 0.30 0.95 1.05 M-13 whole fish 0.04 0.02 0.03 M-14 whole fish 0.03 0.02 0.04 P-33 skin and muscle U 0.82 0.66 1.67 1.73 P-31 gills 11 1.4 0.58 2.5 4,75 P-31 ovary U 0.12 0.07 0.23 0.70 P-31 viscera (w/o giUs or ovary) U 0.86 0.61 0.91 1.71 Tilapia Tilapia mossambica _2 "muscle" 8.8 6.6 4.5 1.0 8 S-13 red muscle 15 0.01 ND ND 9 _3 white muscle 15 ND ND ND 9 S-13 scraped skin 15 0.01 ND 0.01 9 S-2 3 skin and muscle 15 0.01 0.03 0.03 9 S-18 gills 15 0.45 0.05 0.06 9 2 liver 10.3 9.7 3.2 8 S-18 viscera (w/o gills) 15 0.02 ND ND 9 S-2S viscera (w/o gills) 15 0.07 0.02 0.03 9 60 Pesticides Monitoring Journal TABLE 1. — Pesticide residues in fish, by species and tissue — Continued Standard Residues in PPM (Wet Weight) i Specimen TiSSlIE Length (CM) Collection DDE DDD DDT DIELDRIN TOXAPHENE SrrE Sonoran sucker Catastomus inslgnis MC-1 skin and muscle 16 0.42 ND 0.21 ND ND 2 MC-2 skin and muscle 16 0.33 0.05 0.18 ND 0.25 2 TC-7 skin and muscle 10 1.89 0.30 0.98 ND 2.19 3 TC-8 skin and muscle 10 5.89 0.69 2.89 ND 5.78 3 TC-9 skin and muscle 13 1.28 0.13 0.68 1.60 3 TC-1 skin and muscle 13 2.10 0.23 1.18 ND 2.63 3 TC-2 skin and muscle 12 1.70 0.23 1.18 ND 4.00 3 TC-3 skin and muscle 11 0.80 0.15 0.40 ND 1.10 3 TC-1-3 kidneys 11-13 1.99 0.30 1.65 ND 4.23 3 MC-1 viscera 16 4.75 1.25 1.25 trace 2.75 2 MC-2 viscera 16 6.00 2.00 2.75 ND 4.50 2 TC-7 viscera 10 9.65 1.02 5.59 0.007 13.71 3 TC-8 viscera 10 15.22 2.17 11.24 trace 15.95 3 TC-9 viscera 13 11.33 1.33 4.89 9.11 3 TC-1 viscera 13 12.00 1.38 7.25 ND 13.75 3 TC-2 viscera 12 22.50 3.25 17.75 trace 27.50 3 TC-3 viscera 11 30.42 0.84 15.00 trace 172.92 3 Gila sucker Pantosteus clarki TC-6 whole fish 6.5 7.25 1.50 6.75 trace 25.00 3 TC-4 skin and muscle 10.5 0.71 0.09 0.49 ND 0.71 3 TC-5 skin and muscle 9.5 1.75 0.36 1.49 ND 4.91 3 TC-4 viscera 10.5 26.15 2.75 10.55 0.01 25.24 3 TC-5 viscera 9.5 16.15 3.13 16.93 trace 42.94 3 Striped mullet Mugil cephalus "muscle ■ 4 18 0.005 0.005 0.105 10 _ "muscle 18 0.005 0.005 0.005 10 _ "muscle 18 0.005 0.005 0.005 10 _ "muscle 18 0.005 ND ND 10 _ "muscle 18 0.005 ND ND 10 _ "muscle 18 0.005 ND ND 10 _ "muscle 33 0.140 0.095 0.200 7 _ "muscle 35 0.195 0.145 0.440 7 _ "muscle 39 0.235 0.130 0.075 7 _ "muscle 38 0.265 0.100 0.200 7 _ "muscle 36 0.135 0.100 0.145 7 _ "muscle 36 0.185 0.100 0.240 7 _8 red muscle 42 45.00 92.50 50.00 6 8 white muscle 42 23.00 50.00 28.50 6 _3 skin and muscle 42 3.50 6.75 3.75 6 — 8 scraped skin 42 0.10 0.18 0.11 6 3 adipose eyelid 42 0.04 0.08 0.05 6 1 ND = not detected; blank = data not obtainable. 2 Collected by Arizona Game and Fish personnel. 8 Collected by E. McClendon. * Fillet including both red and white muscle. Vol. 4, No. 2, September 1970 61 Pesticide Residues in Channel Catfish From Nebraska^ N. P. Stucky ABSTRACT Channel catfish (Ictalurus punctatus) were collected in all of the major watersheds in Nebraska during the summer of 1964. Individual fat samples and composite blood samples obtained from these fish were analyzed to determine the con- centrations of residues of DDT and its metabolites (^o,p'- DDT, p,p'-DDT, p,p'-DDD, and DDE) and dicldrin. A total of 178 fish, collected from 18 sites, were analyzed. As ex- pected, the fat samples contained higher concentrations of the pesticides than did the blood samples. DDT residues were found in all fat samples, and average levels from 10 fish sampled at each site ranged from a low of 2.2 ppm in the sandhills region to a high of 92.2 ppm in Salt Creek below Lincoln, Nebr. Average dieldrin residues in fat samples ranged from 0.1 to 6.7 ppm. All values are expressed as ppm (fig/g) of residue detected in each sample of fat. The com- posite blood samples were found to contain DDT residue concentrations ranging from a low of less than 0.01 ppm to a high of 0.16 ppm. Dieldrin residue concentrations ranged from a low of less than O.OI ppm to a high of 0.07 ppm. Introduction In 1964, the Research Division of the Nebraska Game and Parks Commission initiated a statewide exploratory study to determine the extent of environmental con- tamination by pesticides. The primary objective of this investigation was to determine the concentrations of DDT (including its isomers and metabolites) and dieldrin residues in Nebraska watersheds using the chan- nel catfish, Ictalurus punctatus, as an indicator species. In 1968, Lyman et a! (7) used fish to demonstrate the presence of DDT in an aquatic environment. Anderson From the Research Division, Nebraska Game and Parks Commission. Lincoln, Nebr. 68509. 62 and Everhard (/) conducted similar studies relating to DDT in fish, and Weiss (//) discussed the use of fish as indicator organisms to determine the extent of en- vironmental contamination by pesticides. The channel catfish was chosen as the species to be analyzed primarily because of its ubiquitous occurrence, omnivorous food habits, and value as a sport and food fish. During the summer of 1964, fat samples and composite blood samples were obtained from channel catfish in all major watersheds throughout the State. Analyses for DDT and dieldrin were performed in our laboratory. Data were evaluated to determine the relative con- centrations of pesticide residues in channel catfish with- in each watershed. Due to the mobility of channel catfish as reported by Welker (12) ana Muncy (9), the levels expressed in this study should be interpreted as quantitative values repre- senting the amount of DDT and dieldrin contamination of fish at the collection site but not throughout entire watersheds. However, results of these analyses can in- dicate areas in the State where pesticide concentrations exceed the maximum allowable level established by the Food and Drug Administration and therefore warrant additional study. Sampling Procedures Eighteen collection sites were selected for study (Fig. 1), representing all of the major drainage systems in Ne- braska. Topography and land use in the watersheds were the primary considerations in selection of the collection sites. With respect to these physical factors, homogeneity of the watersheds above the collection sites was sought as much as possible. Pesticides Monitoring Journal Samples were collected during a 2-week period in the summer of 1964. Ten channel catfish were collected at each site by means of either a back-pack shocker or rotenone. An exception to this was the site on the Middle Loup River from which only eight fish were collected after several days of sampling. To eliminate the possibility of introducing additional variables, an efi'ort was made to obtain fish ranging from 25 cm to 35 cm in total length. At several sites however, it was necessary to deviate from this range in order to obtain a sample of 10 fish. In the field a sample of visceral fat was obtained from each fish, and one or more composite blood samples were obtained for each collection site. Blood samples were obtained by removing the caudal fin with scissors and allowing blood to drain into a vial. Blood and fat samples were then placed in a cooler where they were held until freezing. Each fat sample was carefully weighed to within 0.0001 g, attempting to maintain a range within 0.1-0.3 g. The sample was then placed in 5 ml of 70% DMF and subjected to 10 minutes of ultrasonic vibration. Five ml of isooctane was added, and this mixture was shaken vigorously for 2 minutes. Equilibration between the two phases was accomplished by either centrifugation or allowing the mixture to stand for a period of 2 to 24 hours. Aliquots (1-6 |U,1) of the upper phase were then qualitatively and quantitatively analyzed for pesticides by injection into the columns described above. Each sample was analyzed by this method, and a separate in- jection for each of the five standards (pesticide samples of a known concentration) followed every sample. The minimum level to be recorded quantitatively was set at 0.01 ppm for dieldrin, DDE, o,p'-DDT, p,p'- DDT, and p.p'-DDD. Recovery effectiveness ranged from 76% to 99%. The results expressed in this study were corrected accordingly. Analytical Procedures Both the quantitative and qualitative analyses were carried out using an Aerograph Model 204, dual column, electron capture gas chromatograph. TTie following in- strument parameters were used: Columns: For dieldrin and DDE — Metal, Vs" x 5' containing 41/2' of 4% SE-30/6% QF-1 and 6" of 3% OV-17 on 60/80 Chromosorb W, regular solid support For o.p'-DDT. p.p'-DDT, and p,p'-DDD— Glass. Vs" X 5' containing 11% 0V-17/QF-1 on 80/100 Gas Chrom Q, DMCS treated solid support Temperatures: Column 1 85 C Detector 195 C Injector 230 C Carrier Gas Flow Rate (Nitrogen) : For dieldrin and DDE column — 75 and 35 ml/ minute for fat and blood, respectively For o.p'-DDT. p.p'-DDT and p.p'-DDD Column— 42 and 35 ml/minute for fat and blood, respectively The method employed in the extraction of pesticides was rapid and convenient for an exploratory study of this nature; however, it is not recommended for a study where extreme accuracy is of paramount importance. The single distribution method was first applied to the extraction of pesticides in 1965 by Beroza and Bowman (2). The solvent system used was as follows: Lower phase: 70% DMF (dimethylformamide) Upper phase: isooctane Vol. 4, No. 2, September 1970 Results and Discussion The exploratory nature of this study warranted expres- sion of the results only as average concentrations present at each collection site. Data were not subjected to statistical treatment such as the calculation of standard error because the mobility of channel catfish makes it unreasonable to assume that all fish at a given location should have the same residue concentration. As pointed out by Buhler et al. (4). residue concentrations may also be a function of size of fish. To demonstrate extremes in concentrations found at each collection site, ranges are included in the data presented for fat samples. Be- cause blood samples were comprised of up to five fish, ranges are not given. FAT The residue concentrations of DDT and its isomers and metabolites found in the fat of channel catfish in Ne- braska watersheds are presented in Table 1 and Fig. 1. The values represent averages for the 10 fish collected from each sampling site. While samples were analyzed for the residual concentration of each specific isomer and metabolite of DDT and are expressed as such, the most significant value is the "total DDT" (Table 1) be- cause, as pointed out by Spencer (10), the ratio of DDT to its metabolites changes in accordance with the length of storage time prior to analysis. The samples collected in this study were analyzed over a 1-year period. Residues ranged from a low of 2.16 ppm (Niobrara River) in the sandhills region, to a high of 92.16 ppm in Salt Creek below Lincoln, Nebr. 63 A followup study to locate the main source of DDT pollution is presently being done on Salt Creek where the concentration was found to be 92.2 ppm. Laboratory analyses showed these fish to be comprised of approxi- mately 9% fat. Therefore, by extrapolation, these fish contained approximately 10.3 ppm DDT, on a whole- fish basis, well above the maximum allowable level of 5.0 ppm established by the Food and Drug Administra- tion. The results of this investigation are supported by a study by Henderson et al. (5). Three samples comprised of a total of 15 channel catfish, collected from the Missouri River at Nebraska City, were analyzed for residues of DDT and its metabolites. Concentrations ranged from 0.21 to 2.03 ppm on a whole-fish basis. This compares to the range of 0.15 to 2.20 (1.67 - 24.46 ppm on a fat basis) found in fish from the Missouri River in this study. Dieldrin Dieldrin residues were found in fat samples from channel catfish collected from all 18 Nebraska water- sheds. Residual concentrations found in the various watersheds are presented in Table 2 and Fig. 2. Values shown are averages of the 10 fish collected at each site. Residues ranged from a low of 0.08 ppm in the Middle Loup River to a high of 6.71 ppm in the Missouri River. FIGURE I. — Residues of DDT (including ils isomers and metabolites) in fat of channel catfish from Nebraska water- sheds (values expressed to nearest 0.1 ppm) Scale: 1 inch = approximately 80 miles * — Collection sites — Watershed boundary River or creek FIGURE 2. — Residues of dieldrin in fat of channel catfish from Nebraska watersheds (values expressed to nearest 0.1 ppm) Scale: 1 inch = approximately 80 miles * — Collection sites — Watershed boundary River or creek TABLE 1. — Concentration of DDT in fat samples from 10 channel catfish collected at each site Average Residiie Levels nPPM Watershed DDE o.p'-DDT p.p'-DDT P.p'-DDD Total DDT DDT AND metabolites (10 FISH) (PPM) Missouri River 3.54 0.13 3.17 2.41 9.25 1.67- 24.46 Niobrara River 1.01 0.03 0.80 0.32 2.16 0.72- 2.99 White River 0.92 0.03 3.81 0.75 5.51 2.19- 8.62 Platte River Lower 2.84 0.86 8.57 5.28 17.55 8.58- 35.86 Upper 17.89 0.24 5.06 4.35 27.54 5.86- 87.17 Salt Creek 10.57 4.50 39.85 37.24 92.16 13.61-258.68 Loup River 3.81 0.62 3.54 3.59 11.56 0.28- 26.59 Middle Loup River 4.52 0.01 2.66 1.25 8.44 2.80- 28.66 Elkhorn River Lower 5.13 3.11 5.23 5.24 18 71 2.71- 38.79 Upper 10.29 2.24 4.29 4.03 20.85 4.75- 56.86 Logan Creek 4.40 0.05 2.65 8.04 15.14 9.60- 23.22 Little Nemaha River 2.44 0.41 5.25 1.74 9.84 3.74- 31.24 Big Nemaha River 4.38 0.06 2.42 1.14 8.00 3.64- 35.50 Big Blue River 1.56 0.12 1.43 0.63 3.74 1.73- 10.44 West Fork Big Blue River 2.33 0.03 1.54 1.64 5.54 2.87- 10.83 North Fork Big Blue River 1.68 0.02 1.28 0.51 3.49 1.28- 6.91 Little Blue River 1.60 0.14 4.04 0.87 6.65 1.93- 21.02 Republican River 1.39 0.14 2.63 1.17 5.33 1.17- 16.91 64 Pesticides Monitoring Journal The dieldrin residue concentration found by Henderson et al. (5) in channel catfish from the Missouri River ranged from 0.04 to 0.18 ppm on a whole-fish basis. The concentrations found in this study ranged from 0.32 to 1.43 ppm (2.88-12.86 on a fat basis). BLOOD As would be expected, the chlorinated hydrocarbon pesticide residue concentrations were considerably lower in blood samples than in fat. Results of the analyses of composite blood samples from each watershed (with the exception of the Middle Loup River and the Elkhorn River, upper site, from which no samples were obtained) are presented in Table 3. The composite blood samples were found to contain average DDT residue concentra- tions ranging from a trace (<0.01 ppm) to a high of 0.16 ppm. Dieldrin residue concentrations ranged from a trace to a high of 0.07 ppm. TABLE 2. — Concentration of dieldrin in fat samples from 10 channel catfish collected at each site Watershed DrELDRIN RESIDireS IN PPM Average Range Missouri River 6.71 2.88-12.86 Niobrara River 0.16 <0.01- 0.72 White River OM 0.08- 0.53 Platte River Lower 2.88 1.36- 4.79 Upper Salt Creek 1.78 452 0.69- 5.58 2.58- 7.35 Loup River Middle Loup River Elkhorn River Lower S.68 0.08 3.98 1.87- 9.55 <0.01- 0.27 0.74-11.25 Upper Logan Creek Little Nemaha River 2.79 2.62 3.08 0.52- 5.76 1.95- 3.10 2.08- 4.13 Big Nemaha River 0.99 0.31- 1.82 Big Blue River 2.55 1.32- 3.97 West Fork Big Blue River North Fork Big Blue River Little Blue River 1.42 4.72 0.63 0.29- 2.19 0.69- 7.64 0.34- 1.17 Republican River 0.98 0.33- 2.03 TABLE 3. — Concentration of pesticides in blood samples from channel catfish [T = Trace, <0.01 ppm] Watershed Residues IN PPM Total DDT Dieldrin Missouri River 0.15 0.07 Niobrara River T T White River 0.01 T Platte River Lower 0.05 0.04 Upper Salt Creek 0.13 0.12 0.01 0.02 Loup River Middle Loup River Elkhorn River 0.05 0.03 Lower 0.06 0.03 Upper Logan Creek Little Nemaha River 0.16 0.07 0.07 0.03 Big Nemaha River 0.06 0.02 Big Blue River 0.01 0.02 West Fork Big Blue River 0.12 0.01 North Fork Big Blue River 0.03 0.02 Little Blue River 0.12 0.01 Republican River 0.02 0.01 NOTE: Samples were comprised of blood from 1-5 fish. Vol. 4, No. 2, September 1970 65 A review of literature indicated that blood samples are not generally used in pesticides monitoring work. Bridges et al. (5) found that blood samples from black bullheads, Ictahirus melas, contained a high of 2.7 ppm DDT and metabolites. Samples were obtained 13 months after the farm pond in which the fish were held had been treated with 0.02 ppm DDT. Witt et al. (13) found that a good correlation existed between DDT in blood and the amount in adipose tissue. Studies of the Mississippi River fish kills by Mount et al. (8) indicate that acute toxicity, resulting from endrin, can be di- agnosed from blood concentrations, which are inde- pendent of time of exposure and water concentration. Johnson (6) suggests that for monitoring work, analyses of blood samples may be more practical than other techniques more commonly employed. Acknowledgments The author gratefully acknowledges the analytical work of chemists William J. Ihm, Glen E. Dappen, and Larry A. Witt. This is a contribution of Federal Aid in Fish Restoration, Project F-4-R, Nebraska. See Appendix for chemical names of compounds mentioned in this paper. LITERATURE CITED (/) Anderson. R. B. and W. H. Everhart. 1966. Concen- trations of DDT in landlocked salmon (Saltno salar) at Sebago Lake, Maine. Trans. Amer. Fish. Sec. 95(2): 160-164. (2) Beroza, M. and M. C. Bowman. 1965. Identification of pesticides at nanogram level by extraction p-values. Anal. Chem. 37(2):291-292. (3) Bridges, W. R., J. B. Kallman, and K. A. Andrews. 1963. Persistence of DDT and its metabolites in a farm pond. Trans. Amer. Fish. See. 92(4):42 1-427. (4) Buhler. D. R., E. M. Rasmusson, and E. W. Shanks. 1969. Chronic oral DDT toxicity in juvenile coho and Chinook salmon. Toxicol. Appl. Pharamacol. 14(3):535- 555. (5) Hender.wn, Croswell, L. Wendell Johnson, and Anthony IngUs. 1969. Organochlorine insecticide residues in fish (National Pesticide Monitoring Program). Pesticides Monit. J. 3(3):145-172. (6) Johnson, Donald W. 1968. Pesticides and fishes — ^A review of selected literature. Trans. Amer. Fish. Soc. 97(4):398-424. (7) Lyman, Lee D., William A. Tompkins, and James A. McCann. 1968. Massachusetts pesticide monitoring study. Pesticides Monit. J. 2(3):109-122. (S) Mount, D. I., L. W. Vigor, and M. L. Schafer. 1966. Endrin: use of concentration in blood to diagnose acute toxicity to fish. Science 152:1388-1390. (9) Muncy, R. J. 1958. Movements of channel catfish in Des Moines River. Boone County, Iowa. Iowa State Coll. J. Sci. 23:563-571. {10) Spencer, Donald A. 1967. Problems in monitoring DDT and its metabolites in the environment. Pesticides Monit. J. l(2):54-57. (11) Weiss, Charles M. 1965. Use of fish to detect organic insecticides in water. J. Water Pollut. Contr. Fed. 37: 647-658. (12) Welker, Bill. 1967. Movements of marked channel cat- fish in the Little Sioux River, Iowa. Trans. Amer. Fish. Soc. 96(3):351-353. (13) Witt, J. M., W. H. Brown, G. I. Shaw, L. S. Maynard, L. M. Sullivan. F. M. Whiting, and J. W. Stull. 1966. Rate of transfer of DDT from the blood compartment. Bull. Environ. Contamination Toxicol. 1:187-197. 66 Pesticides Monitoring JotmNAL PESTICIDES IN SOIL 'Apparent" Organochlorine Insecticide Contents of Soils Sampled in 1910^ B. E. Frazier, G. Chesters, and G. B. Lee ABSTRACT A total of 34 soil samples taken in Wisconsin between 1909 and 1911 were extracted for organochlorine insecticides and analyzed by gas chromatography on 3 columns — a mixed QF-l/IV-17 column; a mixed QF-l/DC-200 column; and a diethylene glycol succinate column. The samples had been stored continuously since collection in tightly sealed glass jars and presumably were free of insecticide contamination. Of the 34 samples 32 showed some "apparent" insecticide residues on at least one of the columns. Because peaks cor- responding to particular organochlorine insecticides on chro- matograms from one column did not recur on other columns, it was concluded that the peaks arose from co-extracted indigenous soil components. Peaks corresponding to hepta- chlor epoxide on the QF-1 /OV-17 column and to aldrin on the QF-l/DC-200 column provided greatest interference in chromatographic determination. Introduction to be interfering compounds (3). Extraction and an- alytical procedures suitable for organochlorine insecti- cide determinations in soils must be able to disprove "apparent" residues when used on soils known to be free of insecticide residues. In the routine analysis of soil samples for insecticides it is advantageous if methods can be designed to elimi- nate cleanup. Partitioning and Florisil cleanup steps are used commonly, but contamination from indigenous soil components may still be present (2), and a con- firmatory analysis is necessary. This investigation on insecticide-free samples was de- signed to determine the extent of interference with in- secticide determination arising from indigenous soil components. Insecticide-free soils were obtained from a collection of 34 soil samples collected between 1909 and 1911 and stored in tightly stoppered glass jars since that time. Multicolumn gas chromatography has been used suc- cessfully to determine qualitatively and quantitatively the content of organochlorine insecticides in soils; how- ever, the method has not been tested extensively using soil samples known to be free of insecticides. The chromatogram of one uncontaminated soil revealed small interfering peaks, suggesting that concentrated soil ex- tracts may show "apparent" insecticides where none exist (/). In another investigation, five uncontaminated soil samples gave gas chromatographic responses to •y-BHC, aldrin, and endrin (2). One response which was apparently caused by y-BHC was in fact caused by sulfur. By use of multicolumn gas chromatography, aldrin-like compounds found in plant materials proved ' From the Department of Soil Science, University of Wisconsin, Madi- son, Wis. 53706. Vol. 4, No. 2, September 1970 Methods and Materials The soils (Table 1) consisted of samples with a wide range of textural class; their organic matter content ranged from 1.0% to 7.9%. An unpublished report on the soil samples indicates that organic matter was de- termined by a chromic acid wet oxidation method, but details of the procedure are not available. The textural class was obtained by observation. Although the descrip- tion of the methodology is inadequate, it is believed that these data are as reliable as data which might be obtained presently on 60-year-old samples. The organochlorine insecticides — j^-BHC, heptachlor and its epoxide, aldrin, dieldrin, endrin, p,p'-DDD, p,p'- DDT, and p,p'-methoxychlor — used as standards were those described earlier {4). 67 TABLE 1. — Properties of soil samples used Sample No. Soil Series Texture Percent Organic Matter 1 Ontonagon clay 3.4 2 Miami sUt loam 2.2 3 Fayette silt loam 1.4 4 Miami silt loam 2.4 5 Dubuque silt loam 1.8 6 Plainfleld sand 1.1 7 Tama silt loam 4.1 8 Tama silt loam 7.9 9 Morley silt loam 2.7 10 Piano silt loam 5.2 11 Oshkosh silt loam 2.6 12 Piano silt loam 5.7 13 Ontonagon clay 3.0 14 Oshkosh clay loam 1.7 IS Miami silt loam 2.3 16 Goodman silt loam 3.3 17 Fox silt loam 3.7 18 Fox silt loam 1.5 19 Warsaw silt loam 5.3 20 Casco sandy loam 2.3 21 Plainfleld sand 1.8 22 Piano silt loam 6.2 23 Oshkosh silt loam 2.1 24 Warsaw sandy loam 2.9 25 Huntsville silt loam 7.6 26 Miami silt loam 1.5 27 Plainfleld sand 1.3 28 Tama silt loam 4.3 29 Oshkosh clay loam 3.9 30 Fayette silt loam 4.4 31 Plainfleld sand 1.7 32 Hixton fine sandy loam 1.5 33 Lomira silt loam 2.3 34 Hochheim gravelly loam 2.1 A Packard Model 7620 gas-liquid chromatograph was used for analysis of organochlorine insecticides. Gas chromatographic conditions were: carrier gas, No with flow rate of 125 ml/ minute; ^H-foil electron-capture detector at 210 C, 50 volts; column temperature, 190 C; inlet temperature, 235 C; outlet temperature, 225 C. The three types of columns were: (1)2 parts 10% QF-1, 1 part 3% OV-17 on 60/80 mesh Gas Chrom Q (2 meters x 4 mm I.D.); (2) 1 part 17% QF-1, 1 part 11% DC-200 on 60/80 mesh Gas Chrom Q (2 meters X 4 mm I.D.); and (3) 10% dietheylene glycol suc- cinate (DGS) on 60/80 mesh Gas Chrom Q (1 meter X 4 mm I.D.). The instrument incorporates the use of glass columns and on-column injection to avoid sample degradation resulting from contact of the sample with metal surfaces. Air-dried soil from the surface 20 cm was extracted in quantities of 100 g or 50 g (when a limited amount was available) with 200 ml of a 41:59 Skelly B: acetone azeotropic mixture using a Soxhlet technique in all glass apparatus (5). Concentration of the extract to 25 ml was achieved by forced air evaporation at 40 C. For samples 1-9 inclusive a fivefold dilution of the 25 ml concentrated extract was required for quantitative gas chromatography. The concentrated extracts were not subjected to any method of cleanup. 68 Results and Discussion Using three-column gas chromatography most of the soil extracts gave peaks corresponding to organochlo- rine insecticides. On the QF-l/OV-17 column, 32 of 34 samples apparently showed measurable quantities of "heptachlor epoxide," and 24 samples contained "hepta- chlor"; peaks comparable to y-BHC, aldrin, and dieldrin were found in a few samples (Table 2). On the QF-1/ DC-200 column 20 of 34 samples apparently contained measurable quantities of "aldrin" with slight interfer- ence from "y-BHC" and "dieldrin." A low degree of confusion between organochlorine insecticides and in- digenous soil components was found on the DGS column; small amounts of "y-BHC" and "heptachlor epoxide" would have been reported if this column had been used exclusively. No indigenous soil components which would interfere with determination of endrin, p.p'-DDD, p,p'-DDT, or p,p'-methoxychlor were found on any of the columns. Major interferences (>100 ppb) were found for "heptachlor epoxide" in soil samples 1-8 inclusive on the QF-l/OV-17 column and for "aldrin" in samples 1, 2, 5, and 6 on the QF-1 /DC- 200 column. Chromatograms of the Skelly B: acetone extract of sample 9 on each of the three columns are shown in Fig. 1, 2, and 3. This sample was chosen because it was qualitatively similar to the other soil samples while showing a moderate amount of interference with in- Pesticides Monitoring Journal TABLE 2. — "Apparent" organochlorine insecticide contents of air-dried soils sampled in 1910 "Apparent" Insecticide Contents in PPB on Columns Sample QF-l/OV-17 QF-l/DC-200 DCS No. Heptachlor Epoxide Heptachlor 7-BHC Aldrin Dieldrin T-BHC Aldrin Dieldrin i-BHC Heptachlor Epoxide 1 823 3 3 0 0 0 141 0 0 0 2 807 3 2 0 0 0 294 0 4 0 3 742 1 0 0 0 0 13 0 0 0 4 696 5 2 0 0 0 63 0 9 0 5 564 20 9 5 0 0 164 0 9 0 6 378 5 6 0 0 0 lOO 0 3 0 7 188 1 1 0 0 0 60 0 0 0 8 156 0 2 0 0 0 26 0 0 0 9 51 2 0 0 0 0 43 0 2 0 10 49 3 2 0 0 0 8 0 0 2 11 43 5 2 1 0 0 0 0 0 0 12 40 5 2 0 0 0 0 0 0 3 13 33 8 0 2 0 0 12 0 2 2 14 28 6 0 0 0 0 8 0 3 0 15 25 8 2 2 0 0 0 0 3 0 16 20 0 0 0 0 0 41 0 0 0 17 19 4 0 0 0 0 4 0 0 0 18 14 3 0 0 0 0 4 0 2 0 19 9 1 1 0 0 0 20 0 0 0 20 8 2 0 0 0 0 0 0 0 0 21 7 8 0 9 0 0 0 40 4 6 22 7 2 0 0 0 0 0 0 0 0 23 7 1 0 0 0 0 6 0 0 4 24 4 0 2 0 0 6 5 0 0 0 25 4 1 0 0 0 0 0 0 0 0 26 4 4 0 0 3 0 0 0 2 1 27 3 2 0 0 0 0 0 0 0 0 28 3 0 0 0 0 0 0 0 0 0 29 2 0 0 0 0 0 0 0 0 0 30 2 0 0 0 0 0 0 0 0 0 31 2 0 0 0 0 0 3 0 0 0 32 1 0 2 0 0 0 3 0 0 0 33 0 0 0 0 0 0 0 0 3 0 34 0 0 0 0 0 0 0 0 2 0 secticide determination (Table 2). Extensive contam- ination by "heptachlor epoxide" and slight contamina- tion by "y-BHC" and "heptachlor" are indicated on the QF-l/OV-17 column (Fig. 1). "Aldrin" is the apparent contaminant found on the QF-l/DC-200 col- umn (Fig. 2) while a small amount of "y-BHC" was found on the DGS column (Fig. 3). Interfering peaks were classed as those which had a retention time (R,) equal to R, ±30 seconds of that of the standard in- secticide. With a larger discrepancy in R,-value, it is believed that the insecticide would be resolved satisfac- torily from the indigenous soil contaminant. From examination of Table 2, it can be seen that the "apparent" insecticide residues in the soil arise from indigenous compounds which display chromatographic characteristics similar to a particular organochlorine insecticide. However, the characteristic is unique to one of the columns and not displayed on either of the other two columns. The use of a combination of any two of the three columns described would reveal satisfactorily any chromatographic discrepancies arising from co-extraction of naturally occurring soil com- ponents. Six of the samples indicated small amounts of "heptachlor epoxide" on the DGS column which might be considered confirmatory for the "heptachlor epoxide" found on the QF-l/OV-17 column. However, no peaks Vol. 4, No. 2, September 1970 corresponding to heptachlor epoxide were found on the QF-l/DC-200 column, and it is not believed that this discrepancy would lead to confusion if only two columns were used since the quantities of "heptachlor epoxide" on DGS are extremely small; the highest amount was 6 ppb for sample 21. In Table 1 the samples are arranged in decreasing order of "apparent" insecticide contamination based on the "heptachlor epoxide" peak shown on the QF-l/OV-17 column. No relationship was found between "apparent" insecticidal contamination (Table 2) and the soil prop- erties described in Table 1 . On each of the three columns a large peak was found which displayed an Revalue greater than that for p,p'- methoxychlor on the QF-l/OV-17 and QF-l/DC-200 columns which are relatively nonpolar. However, on the relatively polar DGS column, the peak had a short R, -value (comparable to that of dieldrin and endrin on the DGS column) which would interfere with insecti- cide determination (Fig. 3). This peak was found to result from the forced air evaporation of the soil extracts using Tygon tubing to blow air over the extracts. Small amounts of Tygon must have been dissolved by the Skelly B: acetone solvent during this procedure since Skelly B was capable of extracting the material from Tygon in less than 10 seconds (Fig. 3). 69 Approved by the Director, Research Division, College of Agricultural and Life Sciences, University of Wisconsin, Madison, Wis. This investigation was supported in part by the U.S. Department of Agri- culture, ARS Contract No. 12-14-100-8154(14) and the U.S. De- partment of the Interior, Office of Water Resources Research Proj- ect No. B-016-WIS. FIGURE 2. — Gas chromatograms of organochlorine insecti- cide standards and the Shelly B: acetone extract of soil sample 9 on a QF-l/DC-200 See Appendix for chemical names of compounds mentioned in this paper. LITERATURE CITED (/) Bowman, M. C, H. C. Young, and W. F. Bartliel. 1965. Minimal concentrations of aldrin, dieldrin, and hepta- chlor in soil for control of White-Fringed Beetles as de- termined by parallel gas chromatographic and biological assays. J. Econ. Entomol. 58:896-902. (2) De Vries, D. M., R. D. Collins, and R. T. Rossi. 1968. What is a residue — or analytical artifacts? Symposium on the science and technology of residual insecticides in food production with special reference to aldrin and dieldrin. Shell Oil Company. (3) Goodwin, E. S., R. Goulden, and J. G. Reynolds. 1961. Rapid identification and determination of residues of chlorinated pesticides in crops by gas-liquid chromatog- raphy. Analyst 86:697-709. (-0 Pionke, H. B., J. G. Konrad, G. Chesters, and D. E. Armstrong. 1968. Extraction of organochlorine and organophosphate insecticides from lake waters. Analyst 93:363-367. (5) Pionke, H. B., G. Cliesters, and D. E. Armstrong. 1968. Extraction of chlorinated hydrocarbon insecticides from soQs. Agron. J. 60:289-292. FIGURE 1. — Gas cluoniaiograms of organochlorine insecti- cide standards and the Skelty B: acetone extract of soil sample 9 on a QF-1 /OV-17 column FIGURE 3. — Gas chromatograms of organochlorine insecti- cide standards and the Skelly B: acetone extract of soil sample 9 and Tygon tubing on a DGS column 70 Pesticides Monitoring Journal PESTICIDES IN WATER Pesticides in Surface Waters of the United States — a 5-Year Summary, 1964-68^ James J. Lichtenberg, James W. Eichelberger, Ronald C. Dressman, and James E. Longbottom ABSTRACT This report summarizes the results of five annual synoptic surveys (1964-68) for chlorinated hydrocarbon pesticides in surface waters of the United States. The results showed widespread occurrence of these compounds. The number of occurrences reached a peak in 1966 and then declined sharply in 1967 and 1968. Dieldrin and DDT and its congeners DDE and DDD were the compounds most frequently de- tected throughout the 5-year period. The maximum concen- trations found have not exceeded permissible limits as they relate to human intake directly from a domestic water sup- ply. However, they have often exceeded the environmental limit of 0.050 fig/liter recommended by the Federal Commit- tee on Water Quality Criteria. Introduction Since September 1964, the Federal Water Quality Ad- ministration has conducted annual synoptic surveys for chlorinated hydrocarbon pesticides in surface waters (1,2,3). In September 1967 the fourth such survey was conducted, and in June 1968, the first spring survey was made. This surveillance activity has been a part of a continuing program for determining refractory organic substances in surface waters. The purpose is to provide information on present levels and trends of pesticides in waters to permit pollution control authorities to assess the degree of hazard and, if necessary, to provide the required control. Through 1967 the surveys were conducted in September when streamflows are minimal. The 1968 survey was conducted in June in an effort to obtain comparative data during runoff periods after pesticide application. ' From the Analytical Quality Control Laboratory, Federal Water Quality Administration, U.S. Department of the Interior, 1014 Broad- way, Cincinnati, Ohio 45202. Vol. 4, No. 2, September 1970 Previous reports (2,3) have compared synoptic grab sample data with data obtained by the carbon adsorption method (CAM). Generally good agreement was noted between the two types of samples, and no further com- parisons are reported here. Samples were collected through the cooperative efforts of Federal, State, local, and private agencies at ap- pro.ximately 100 sampling stations. These stations are located mainly on interstate and international boundary waters at locations ranging from water treatment plant intakes to sites near mouths of rivers as they discharge to tidal waters. This report summarizes the data obtained throughout the 5 surveys with emphasis on the 1967 and 1968 surveys. The number of samples analyzed for these surveys was 110 and 114, respectively. A total of 529 samples were analyzed for the 5 surveys. Methods The basic procedures for determination of chlorinated hydrocarbon pesticides are detailed in U.S. Department of the Interior Publication WP-22 (4) and in the "FWPC.'V Method for Chlorinated Hydrocarbon Pesti- cides in Water and Wastewater" (5). The samples were collected in I -quart glass bottles (two per sample) equipped with screw caps fitted with Teflon liners. The samples were subjected to liquid-liquid extraction with 15% ethyl ether in hexane, dried over anhydrous sodium sulfate and concentrated to approximately 5 ml in a Kuderna-Danish evaporator. The extracts were carefully evaporated to 0.5 ml in a warm water bath, and up to 10 fA was injected into an electron capture gas chromatograph. If no response was obtained, the extracts were further concentrated to a maximum of 71 0.2 ml and again injected into the chromatograph. Those samples producing a response were then subjected to thin-layer chromatographic separation. The eluates from the thin layer were concentrated as before and again injected into the chromatograph. In addition, the use of the flame photometric detector provided specificity for many organophosphorus pesti- cides. For the 1967 and 1968 surveys, samples were also analyzed for methyl parathion, parathion, fenthion, ethion, malathion, and carbophenothion. The results were confirmed by multiple gas chromato- graphic analyses of the thin-layer eluates using the following conditions: (1) A Perkin-Elmer Model 880 equipped with a parallel plate electron capture detector, an aluminum column 6' x V4" O.D. packed with Gas-Chrom Q (60/80 mesh) coated with 5% QF-1 and 3% Dow-200, and a nitrogen carrier flow of 100 ml/ minute. Temperatures were: injection port — 250 C, column oven — 185 C, and detector— 205 C. (2) A MicroTek Model 179 equipped with a Ni^^ detector, an aluminum column 6' x V4" O.D. packed with Gas Chrom Q (60/80 mesh) coated with 5% OV-17, and a nitrogen carrier flow of 100 ml/ minute. Temperatures were: injection port — 250 C, column oven — 205 C, and detector — 360 C. (3) A MicroTek Model 179 equipped with a flame photometric detector, a glass column 4' x 4 mm I.D. packed with Gas Chrom Q (60/80 mesh) coated with 2% Reoplex-400, and a carrier flow of 75 ml/ minute. Temperatures were: injection port — 185 C, column oven — 185 C, and detector— 160 C. When the quantity of pesticide in the sample permitted, further confirmation was obtained using a microcoulo- metric gas chromatograph under conditions similar to (2) above; in which case, a lack of response for thio- phosphate p>esticides was considered as supporting ev- idence for their presence. Similarly, lack of response to the flame photometric detector supported the identi- fication of the organochlorine pesticides. Recovery data were obtained by dosing distilled water samples with the pesticides and carrying them through the entire analytical procedure. Recovery of organ- ochlorine pesticides ranged from 65% -97% when samples were dosed at levels of 25-100 ng/liter. Re- coveries of thiophosphate pesticides ranged from 40%- 75% at dosages of 50-500 ng/liter. Every tenth sample was run in duplicate. Where results differed, the higher value was reported. The methods are specific for dieldrin, endrin, DDT, DDE, DDD, aldrin, heptachlor, heptachlor epoxide, lindane, BHC, y-chlordane, and technical chlordane. 72 The practical lower limit of detectability for the chlo- rinated pesticides is 0.001 to 0.002 ;u,g/liter, except for technical chlordane which has a limit of 0.005 ^g/ liter. Toxaphene can be detected, if it is present, at levels of the order of 1 /tg/ liter. The detection limits for the phosphorus compounds are 0.010 to 0.025 jug/liter. All results are reported without correction for recovery efficiencies. Thus, the reported concentrations represent minimum values, the actual value being equal to or greater than the reported value. Results and Discussion The results of the 1967 and 1968 surveys are listed in Tables 1 and 2. Table 3 lists the total number of samples and positive pesticide occurrences for each of the five surveys. The data show that the total occurrences peaked in 1966 and fell off significantly in 1967 and 1968. Fig. 1 summarizes the percent occurrences of 10 pesticides for the 5 surveys. It shows that the occur- rences decreased sharply after 1966 for all pesticides except BHC, which showed only a slight decline. It also shows that the 1966 peak in total occurrences (Table 3) is largely due to the increase in DDD occur- rences. The spring survey showed a slight increase in dieldrin and DDT. Fig. 2 shows the geographical occurrence of dieldrin, the DDT group, and BHC for 1967 and 1968. Table 4 summarizes the occurrences (1964-68) by FWQA region. In 1966, the number of occurrences peaked in the South Central Region and in all regions east of the Mississippi. The Missouri Basin Region showed a gradual decline from 1964-1966, then a very sharp drop in 1967 and 1968. In the Southwest and Northwest Regions the occurrences fluctuated from 1964-1966 and then fell off to virtually nothing in 1967 and 1968. Throughout the 5 surveys dieldrin dominated the pest- icide occurrences in all regions and in total occurrences with 199 positive results. DDT was second in overall occurrences with 86. DDT and its congeners DDE and DDD as a group accounted for 183 occurrences, Aldrin and chlordane were low with just two and five occur- rences, respectively. Consistent geographical relation- ships among the various pesticides are difficult to iden- tify; however, the overall occurrences show that dieldrin was slightly predominant in all regions east of the Mississippi and the DDT group, considered as one, was predominant in regions west of the Mississippi. PESTicroES Monitoring Journal Since 1966, BHC has been detected in 10 of 12 samples from the main stem of the Ohio River. This consistent occurrence was verified by the results of the analyses of monthly CAM samples performed in this laboratory. The synoptic surveys and additional investigations by this laboratory produced only one positive result for BHC in eight major tributaries to the Ohio. That one was at Pittsburgh on the Allegheny River in September 1966. Twenty-three other BHC occurrences were widely scattered throughout the Country. Endrin was found in over 30% of the samples in 1964; the reduction of endrin occurrences to zero in 1968 is particularly significant in light of its association with major fish kills in the Lower Mississippi prior to 1964. Since pesticides are so common in surface waters, it is of interest to note those locations at which they are absent or occur infrequently. Table 8 lists the Stations that fall in this category. Locations in the west and northwest dominate this group. Spring runofif after pesticide application was expected to cause an increase in the number of occurrences and in concentration levels in agricultural areas. Such an in- crease was not evident from the data obtained. This may be, in part, due to the wet spring experienced in much of the Country in 1968 which delayed planting and subsequent pesticide application in many areas. As a result, our collection period may have been too early to catch an increased pesticide load. Heptachlor was found in 14% of the samples in 1965 and in less than 1 % thereafter. Heptachlor epoxide was found in approximately 14% of the samples in 1965 and 1966 and dropped to zero thereafter. The 10 locations at which the highest levels of each pesticide were observed for each survey are listed in Table 5. Individual locations varied considerably. How- ever, 2 stations on the Savannah River, at North Augusta, S. C. and Port Wentworth, Ga., were in the top 10 locations for dieldrin occurrences for all 5 sur- veys. Other rivers and locations that were consistently in the top 10 are the Merrimack, Schuylkill, Connecticut, Delaware, Potomac, Lower Ohio, Lower Mississippi, Missouri (at Kansas City), Rio Grande, and Red River (North). The highest level of each pesticide found is listed in Table 6 along with water quality criteria for public water supplies and farmstead uses (6) and suggested maximum reasonable stream allowance (7). While the maximum concentrations have not exceeded permissible limits as they relate to human intake directly from a domestic water supply, they have in some cases exceeded or come quite close to the maximum reasonable allow- ance suggested by Ettinger and Mount (7). Because of the biological concentration factor, these levels are con- sidered hazardous in waters from which fish are har- vested for human consumption. In addition, because of their toxicity to fish, the Federal Committee on Water Quality Criteria recommends that environmental levels of these substances not be permitted to rise above 0.050 jUg/ liter ((5). Of the 84 stations where samples were collected in all 5 surveys, 12 had at least 1 positive occurrence in each survey. These are listed in Table 7. All but one of these are east of the Mississippi River. In addition, 16 widely spread locations had at least 1 positive occurrence in 4 of the 5 surveys. Vol. 4, No. 2, September 1970 FIGURE 1. — Percent occurrence of ten chlorinated hydro- carbon pesticides, 1964-68 DIELDRIN ALDRIN HEPTACHLOR I HEPTACHLOR EPOXIDE NO DAT* LINDANE i BHC CHLORDANE KO NO DATA DATA 1964 1965 1966 1967 1968 73 FIGURE 2. — Occurrence of chlorinated hydrocarbon pesticides in surface waters, synoptic surveys of 1967 and 1968. (• — present; o — not detected) y^, r° gg-\V-l DIELDRIN y *-=>»! r~7=rv^o june isee '^c^l 01 SEPT. 1967 JUNE 1968 74 Pesticides Monitoring Journal TABLE 1. — Results of synoptic survey for pesticides in surface waters, September 1967 Location CONCENTKATION IN ^G/LITER 1 DiELDRIN Endrin DDT DDE DDD Lindane BHC NORTHEAST REGION Connecticut River Enfield Dam, Conn. Northfleld, Mass. Wilder, Vt. .005 .017 - - - .002 - Schuylkill River Philadelphia, Pa. Hudson River Poughkeepsie, N.Y. Narrows, N.Y. .044 — — — — - - Merrimack River Lowell, Mass. .066 __ _ _ _ _ _ Delaware River Trenton, N.J. Martins Creek, Pa. .010 ,013 - .017 - .036 .002 - Raritan River Perth Amboy, N.J. Delaware Bay a - - - - - - - b — — — — — — — MIDDLE ATLANTIC REGION Potomac River Great Falls, Md. Washington, D.C. .025 - - - - - - Shenandoah River Berryville, Va. _ _ _ .002 Susquehanna River Conowingo, Md. Sayre, Pa. - - - - - - - Roanoke, River John H. Kerr Dam, Va. _ _ _ _ Neuse River Raleigh, N.C. _ _ _ _ _ — _ SOUTHEAST REGION Apalachicola River Chattahoochee, Fla. .015 — — — .053 .003 — Beauclair River Lake Apopka, Fla. — — .316 .050 .231 — — Escambia River Century, Fla. — — — — — — — Oklahawa River Orlando, Fla. — — — — — — — W. Pahn Beach Canal W. Pahn Beach, Fla. — — — — — .003 — Chattahoochee River Lanett, Ala. — — — — — P — Savannah River Port Wentworth, Ga. .039 . „ — North Augusta, S.C. .087 — — — — — — Clinch River Kingston, Tenn. .004 — — — .032 — — Tennessee River Bridgeport, Ala. — — — Lenoir City, Tenn. — — — — — — — Tombigbee River Columbus, Miss. — — — — P — — OHIO BASIN REGION Allegheny River Pittsburgh, Pa. _ _ _ _ _ Kanawha River Winfield Dam, W. Va. _ _ P _ Monongahela River Pittsburgh, Pa. _ _ _ _ Ohio River Cairo, 111. Evansville, Ind. Cincinnati, Ohio Above Addison, Ohio .020 - - - - - .008 .013 006 Wabash River Lafayette, Ind. New Harmony, Ind. .009 - - - - - - Vol. 4, No. 2, September 1970 75 TABLE 1. — Results of synoptic survey for pesticides in surface waters, September 1967 — Continued Location Concentration in /ig/liter i DIELDRIN Endrin DDT DDE DDD Lindane BHC GREAT LAKES REGION St. Lawrence River Massena, N.Y. P Lake Erie Buffalo, N.Y. P _ _ Detroit River Detroit, Micli. .014 _ _ _ _ .002 St. Clair River Port Huron, Mich. _ _ _ St. Mary's River Sault Ste. Marie, Mich. .004 .003 _ Saginaw River Bay City, Mich. P _ _ _ _ _ .007 Lake Superior Duluth, Minn. _ _ _ _ Lake Michigan Milwaukee, Wis. _ _ _ _ _ Maumee River Toledo, Ohio _ .086 _ _ .270 niinois River Peoria, 111. _ _ _ _ _ _ Mississippi River Cape Girardeau, Mo. E. St. Louis, 111. Burlington, Iowa Dubuque, Iowa St. Paul, Minn. - - - - - - - Fox River Green Bay, Wis. - - - - - - - MISSOURI BASIN REGION Missouri River St. Louis, Mo. Kansas City, Kans. Omaha, Nebr. Yankton, S.Dak. Bismarck, N.Dak. .012 - .066 - - .010 E North Platte River Henry, Nebr. _ _ _ _ Platte River Plattsmouth, Nebr. _ _ _ _ _ _ South Platte River Julesburg, Colo. .024 _ Yellowstone River Sidney, Mont. _ Rainy River Baudette, Minn. _ _ _ _ Red River (North) Grand Forks. N.Dak. Emerson, Manitoba .087 P - .054 - - - - Kansas River Lawrence, Kans. _ .133 .840 _ Big Horn River Hardin, Mont. - - - - - - SOUTH CENTRAL REGION Atchafalaya River Morgan City, La. _ _ _ _ _ _ Arkansas River Pendleton Ferry, Ark. Fort Smith, Ark. Ponca City, Okla. Coolidge, Kans. - - — — — P - Brazos River Areola, Tex. .024 - - - P - - 76 Pesticides Monitoring Journal TABLE 1. — Results of synoptic survey for pesticides in surface waters. September 1967 — Continued Location Concentration in /iO/LFTER i DiELDRIN Endrin DDT DDE DDD Lindane BHC SOUTH CENTRAL REGION— Continued Mississippi River New Orleans, La. Vicksburg, Miss. Delta, La. West Memphis, Ark. New Roads, La. .008 - .019 - .015 .024 - Red River (South) Alexandria, La. Denison, Tex. z - - - - _ Rio Grande River Brownsville, Tex. EI Paso, Tex. Alamosa, Colo. .002 - .018 .022 - - - Verdigris River Nowata, Okla. — — — — — .009 - Trinity River Houston, Tex. - - - - - - - SOUTHWEST REGION Bear River Preston, Idaho _ _ _ _ — — — Colorado River Yuma, Ariz. Parker Dam, Calif. Boulder City, Nev. Page, Ariz. - - - - - - — Green River Dutch John, Utah _ _ _ _ _ — — Klamath River Keno, Oreg. _ _ _ _ _ — — Sacramento River Greens Landing, Calif. _ _ _ _ — — — San Joaquin River Vemalis, Calif. _ _ _ _ _ _ — San Juan River Shiprock, N. Max. _ _ _ _ — — Truckee River Farad, Calif. - - - - - - - NORTHWEST REGION Qearwater River Lewiston, Idaho _ _ _ _ Columbia River Clatskanie, Oreg. Bonneville Dam, Oreg. McNary Dam, Oreg. Pasco, Wash. .018 - - - - - - Pend Oreille River Albeni Falls, Idaho _ _ _ Snake River Wawawai, Wash. American Falls, Idaho - - - - - - - Spokane River Post Falls, Idaho _ _ _ _ _ Willamette River PorUand, Oreg. _ _ _ _ _ _ Yakima River Richland, Wash. - - - - - - - iThe Lanett, Ala. sample contained .036 Ag/liter of chlordane (tech). The Nowata, Okla. sample contained .002 ;ig/liter of aldrin and .003 Ag/liter of heptachlor. The Wawawai, Wash, sample contained .050 /ig/liter of parathion and .380 (ig/hter of ethion. All other samples gave negative results for aldrin, heptachlor, heptachlor epoxide, parathion, methyl parathion, fenthion, ethion, malathion, and carbophenothion. NOTE: — = not detected. P = presumptive. Data are reported as presumptive in instances where the results of chromatography were highly indicative but did not meet all requirements for positive identification and quantification. Vol. 4, No. 2, September 1970 77 TABLE 2. — Results of synoptic survey for pesticides in surface waters, June 1968 I Concentration in ao/liter i Endrin DDT DDE DDD NORTHEAST REGION Connecticut River Enfield Dam, Conn. Northfield, Mass. WUder, Vt. .022 - - - - - Schuylkill River Philadelphia, Pa. .027 _ _ _ Hudson River Poughkeepsie, N.Y. Narrows, N.Y. .013 .004 - .030 - - - _ 1 Merrimack River Lowell, Mass. .012 Delaware River Trenton, N.J. Martins Creek, Pa. .007 .007 - .015 - - - Raritan River Perth Amboy, N,J. _ _ _ _ Delaware Bay — — — - - - - ' MIDDLE ATLANTIC REGION Potomac River Great Falls, Md. Washington, D.C. .007 - .033 - - - - Shenandoah River Berryville, Va. _ _ Susquehanna River Conowingo, Md. Sayre, Pa. .007 - - - - - .009 Roanoke, River John H. Kerr Dam, Va. .010 Neuse River Raleigh, N.C. - - - - - - - SOUTHEAST REGION Apalachicola River Chattahoochee, Fla. .027 Beauclair River Lake Apopka, Fla. _ .220 .041 .156 Escambia River Century, Fla. .006 Oklahawa River Orlando, Fla. .004 .005 .015 W. Palm Beach Canal W. Palm Beach, Fla. Chattahoochee River Lanett, Ala. .025 Savannah River Port Wentworth, Ga. North Augusta, S.C. .039 .059 - - - - - - Tennessee River Bridgeport, Ala. Lenoir City, Tenn. Oak Ridge, Tenn, - - - - - - - Tombigbee River Columbus, Miss. .407 - - - - - - OHIO BASIN REGION Allegheny River Pittsburgh, Pa. Kanawha River Winfield, W.Va. .154 Monongahela River Pittsburgh, Pa. .051 Ohio River Cairo, lU. Evansville, Ind. Cincinnati. Ohio Above Addison, Ohio .005 .014 - - - - .020 .055 .028 .112 Wabash River Lafayette, Ind. .005 - - - - - 78 Pesticides Monitoring Journal TABLE 2. — Results of synoptic survey for pesticides in surface waters, June 1968 — Continued Location Concentration in /io/lites i DiELDRiN Endrin DDT DDE DDD Lindane BHC GREAT LAKES REGION St. Lawrence River Massena, N.Y. — — — — — — — Lake Erie Buffalo, N.Y. — — — — — — — Detroit River Detroit, Mich. — — — — — — — Grand River Grand Haven, Mich. — — — — — — — St. Clair River Port Huron, Mich. — — — — — — — St.'Mary's River Sault Ste. Marie, Mich. — — — — — — — Saginaw River Bay City, Mich, — — — — — — — Lake Superior Duluth, Minn. — — — — — — — Lake Michigan Milwaukee, Wis. — — — — — — — Maumee River Toledo, Ohio — — — — — — — Illinois River Peoria, III. — — — — — — — Vlississippi River Cape Girardeau, Mo. .014 — — — — — — E. St. Louis, 111. .011 — — — — . — — Burlington, Iowa .010 — — — — — — Dubuque, Iowa — — — — _ — — St. Paul, Minn. .011 — — — — — — Fox River Green Bay, Wis. — - — — — — — MISSOURI BASIN REGION Missouri River St. Louis, Mo. .010 — _ Kansas City, Kans. .009 — — — — — — Omaha, Nebr. — — — — — Yankton, S. Dak. ^ .053 Bismarck, N. Dak. — _ — St. Joseph, Mo. — — — — — — — North Platte River Henry, Nebr. — — — — — — — Platte River Plattsmouth, Nebr. .005 — — — South Platte River Julesburg, Colo. — — -^ Yellowstone River Sidney, Mont. — — — — — Rainy River Baudette, Minn. — — .037 — — — — Red River (North) Grand Forks, N.Dak. .027 Emerson, Manitoba — Kansas River Lawrence, Kans. — __ .008 .003 Big Horn River HardHi, Mont. — — — — — — — SOUTH CENTRAL REGION Atchafalaya River Morgan City, La. .005 — — — — Arkansas River Pendleton Ferry, Ark. .005 .037 Fort Smith, Ark. Ponca City, Okla. _ .013 Coolidge, Kans. .009 — .025 Brazos River Areola, Tex. — Mississippi River New Orleans, La. Vicksburg, Miss. -_ .109 .004 West Memphis, Ark. — .005 St. Francisville, La. — — — — — — Vol. 4, No. 2, September 1 1970 79 TABLE 2. — Results of synoptic survey for pesticides in surface waters, June 1968 — Continued Concentration in /ig/liter i SOUTH CENTRAL REGION— Continued Red River (South) Alexandria, La. — — — — — — — Denison, Tex. — — — — — — — Rio Grande River Brownsville, Tex. — — — — — — — El Paso, Tex. — — — — — — — Alamosa, Colo. — — .029 — — — — Verdigris River Nowata, Okla. — — — — — — — Trinity River Houston, Tex. — — — — — — — SOUTHWEST REGION Bear River Preston, Idaho _ _ _ _ Colorado River Yuma, Ariz. Parker Dam, Calif. Boulder City, Nev. Page, Ariz. Loma, Colo. - — - - — — ' Green River Dutch John, Utah _ _ _ _ _ Klamath River Keno, Oreg. _ _ _ _ Sacramento River Green's Landing, Calif. _ _ _ _ _ San Joaquin River Vernalis, Calif. .030 _ _ San Juan River Shiprock, N. Mex. _ _ _ _ _ _ _ Truckee River Farad, Calif. _ _ _ _ _ Kiikii Stream Oahu, Hawaii _ _ _ _ _ Waikele Stream Oahu, Hawaii — — — — - - - NORTHWEST REGION Clearwater River Lewiston, Idaho _ _ _ _ _ _ Columbia River Clatskanie, Oreg. Bonneville Dam. Oreg. McNary Dam, Oreg. Pasco, Wash. - - - - — - - Pend OreiUe River Albeni Falls, Idaho _ _ _ _ _ _ Snake River Wawawai, Wash. Payette, Idaho American Falls, Idaho .004 - .015 - - - - Spokane River Post Falls, Idaho _ _ _ _ _ _ Willamette River Portland, Oreg. _ _ _ _ _ Yakima River Richland, Wash. .006 — .017 — — — — 1 The Lanett, Ala. sample contained .169 /tg/liter of chlorane (tech). All samples parathion, methyl parathion, fenthion, ethion, malathion, and carbophenothion. gave negative r ■suits for aldrin, heptachlor, heptachlor epoxide. NOTE: — = not detected. 80 Pesticides Monitoring Journal ' TABLE 3. — Total number of chlorinated pesticide occurrences Yeak Number of Samples Collected NUMBEK OF Samples With Positive occurkences Total Number OF Positive Occurrences 1964 1965 1966 1967 196S 97 99 109 110 114 73 56 80 34 48 130 120 177 56 63 Total 529 291 546 TABLE 4.— Pesticide occurrences by FWQA Region, 1964-68 Middle Atlantic Ohio Basin Great Lakes Basin Missouri Basin South Central >ieldrin indrin DDT DDE DDD Mdrin Heptachlor Heptachlor epoxide Lindane BHC Chlordane Total Mo. of Samples Vol. 4, No. 2, September 1970 81 1 ' i z o P i 5 1 < z o 1 1 z 3 i o. t- D m § s s s S S 0 0 0 o o o o o o o 0 0 0 ll it 2s ill 1 2I2S •is =3 0 If iJ II 1^ II l' rt O 3 2 Ills iu iz 2-§ •pa So I& .2U us w Kl « < u 0 S X t~ ^ ^ ^ o S s s o g S s s 0 o o O o o 0 0 0 0 0 w Q O S a 2u « a = E? U > a 2 = 1^- zl !l 2-t 2 1 > H go 2l P > "2 2 2 SI S ^ ^1 "SO OS s ft, w 2^ 2« 0 f u u o s o S 0 o o o o o d 0" 0 0 0 O § u > > ^ 2| U !l ra 3 2 g aj •a" III .> 5 i i 1 c 1 "^ 2| ■2 "rt > 2 2° £^ ?Z 'i'^ "U X. gz gw J= ft. ?J i^ 2 « « Q u 0 w u i4 s i o s g 00 00 0 g § s g o o o o o 0 0 0 0 0 > 0 WJ U ^ , c SB fi ¥ c/5 s E -5 fe^ >l 'S. ^ a o" ,2 0 w 2 1 i< >z Il5| uU S S ^ '5 2 S 0 c »;2 b: c S w 5 2 H 0 s s w V, i o o 0 0 o O o o o 0 0 0 0 0 0 U 1 s 0. „■ ■S u.g „• j:B. 0 W I- 3 *- XJ fill > » 2 c > 0. > 2 2 M II IS ^ 2.| a 0 .B-o: -z WJ s o:"^m-|Q >Z giS^ gZ.sQ Q S*Z |.||2£ gH| H 0; (J M . .£ >, Z 2 g- °* 5 .£ z 2 t^ ■a a s 2 -SE osa g o a^ s-g .^s= hS -Sh |u .s« o< gs: i,:j S<0hu2Sh - = o H 2 s ■53 o 2^,«=„-zl24°fl 2>? go oW mK 82 ID lis Pesticides Monitoring Journal 0 0 d C .sj 15 si.>S .« uclair Riv ake Apop sissippi R icksburg, souri Riv( ankton, S M J .•£> .2>< m S S « >J5 ^ 000 i «• "■ Q River popka, F River City, Ka r (North Forks, N ■S<-=ss s-a Beaucl Lake Missou Kans Red R Gran S « 3 000 H U Q 4; I- « ver Tex. e Riv Tex. 1 Rive rg, M 5 M c 0 Si, g§o^|| 2 < 0 w .2 > CO 2 S 000 River Colo. ver N. Me ver Grande lamosa, Juan R liprock, orado R age, Ari 0 <: c in -go. 2 K u r- ts vc 000 i 0 ver hio Nort rks, Riv Calif 2°r£i.» aumee Toledo d Rive Grand n Joaq Vernal S 2 ^ 1 ■3- a. z o P > i z 1 1 z o H . \ z o 0 l-J i 1 z i .J 1 - 1- 1^ m O o o> o o o O O o y o o o o © o o ,u ^ [) Monongahela Rive Pittsburgh, Pa. Rainy River Baudette, Minn. £ Q >^ >a hi S5 5 g < o c 1 5Z > . ■fz K o £ c> to ^ „ o o o u^ b.u Rive Riv le, T ■T Z ssippi wOrle Grande ownsvi " O -H § s O ^ ^ s s o o S g o o o o o o o o r/i ^ ansas River ort Smith, Ar 3mac River reat Falls, M > l"3 > 2 Is o-S ? 1 < 0. .icu .li ^ -H ■^^ «< s il Q ^ ^ m m o o o o O a o o o o e> O o o o •s t." "O « 1- Z B P''^ go u I-. ^ Platte River Plattsmouth, Spokane River Post Falls Da I. _: 2^ 'E. J 02 DC 0^ 5 o 0 u E-8 11 2 Tf „ o> s s O o O a o o O o o O o - . ^^ uIt; a River City, L River g. Miss o ■a _> O I::- 20 2| •s " °--° ,b' M I- -o Si T3 ■^? "a ^ ",g a: s F ■^ 2i1 'it ll.i> S£ u ■?:^ .3<5: lu < s GQ U m St < «>-i .on n 2 ■£S 5>^ ts^u gzj; c. SZ 3 J .> ■ ° ■ 2 £ asi»MU^ c S fe o ^ u is'^-s i&2i2| v3 O >' cu 0^ — ^ o -. — o £ ^ 5 .0 -S .i .5 2 (3 S £ O - ^ a -» -v u O' > a oj « ij ~ g 2 1 I ^. s 2 .o i2g. .> S20 2ao§ 2^. Sc 2=^o■.^<|| |g22 i| E,g||n| sS-gg «l S CO < Q u _;d=£S S.D Si ".2 ■=?=Sac«<2uu'^ ifoo oc Itig is "^11226 S .$ g z ^ f "s "S p5 3" s"l*2 on -jjH .= a. .2t^ .S is bS Q ^ 2 « E d d 2> ■a > 5 Vol. 4, No. 2, September 1970 83 i a. z o p i z " o P i-i i I z - o p o 1-1 H z o P < o I-I H =i i z o P •< u o ~| ^3 SH 1 1 1 1 S § S H n jii £ ^ S & s >; Kl !£■? BS5 Is «£ si |5n ^ PS S (£ ^ w. ^ in .r, ^ r» n 0 S g a g s 0 0 8 0 0 0 0 0 0 0 0 0 0 i 0 c 0 0 •0 £ S2 2 j,| Ie S,o > 2 u C mi II r£ P ■^s JT.R C ni 2| 2= S9 .c ^ rt 3 2S S 3. !S ™ Xi u r< ■fiw nU ri-J r rU vo r 0 0 0 a U 0 0 VO ^ 0 8 S s g 8 Bh 0 0 0 0 0 0 0 u i! 0 •0 c I'l « g 0 > 2 a c ^ mm 0 > 0 21 •P5 1^' c p Jm 2.:: „ 0 ^.1 0 0 w 0 Q w v> ■* ^ a 0 S S S 0 0 0 0 8 X S 0 0 0 0 0 0 0 0 0 0 0 0 0 i > § 0 ^ 0 > , 2^ c £ ll a ,0 0 t- • > 0 J < 0 > 2 S oB 2 Is If H > 0 2a H 2-g, |l < > c 2s '5.— D. 3 la. S 'E. 3 > 2 ° a 5' 3 II s i a< Cl. 0. 0 0 4 0 P 2z I.U S 1^2 > 0 2 S 2£ r£ 2 . 3 "o 1^ as 22 is c ■§ 1^0 rU S'^; ^w oa. PS 0 > u S ^ a. 0 V ^' £ ^ teu SI 3 So « E £.1 |S 1 2" ^ 84 Pesticides Monitoring JoimNA| TABLE 6. — Maximum pesticide concentration found vs. permissible water supply criteria and reasonable stream allowance [blanks = criteria not given] ;iC/LITER Maximum Permis- Desir- Reasonable Pesticide sible able Stream Criteria i Criteria i Allow- ance 1 Found 17 absent 0.25 0.407 1 do 0.1 0.133 DDT 42 do 0.5 0.316 DDE 0.050 DDD 0.840 Heptachlor 18 absent 1.0 0.048 Heptachlor epoxide 18 do 1.0 0.067 Aldrin 17 do 0.25 Lindane (BHC) 56 do 5.0 0.112 Chlordane 3 do 0.25 0.169 Methoxychlor 35 do 20.0 13) Toxaphene 5 do 2.5 (4) Organophosphates plus Carbamates 100 do 0.380 Herbicides: 100 do 131 2,4-D plus 2,4,5-T and 2,4,5-TP Phenols 1 do (3) 1 From the "Report of the Committee on Water Quality Criteria" 2 Suggested by Ettinger and Mount (7) 3 Not determined 4 Not detected TABLE 7. — Locations with high frequency of pesticide occurrence {at least one pesticide found in each survey) River Location Merrimack Lowell. Mass. Delaware Trenton. N. J. Delaware Martins Creek. Pa. Schuylkill Philadelphia, Pa. Potomac Great Falls. Md. Apalachicola Chattahoochee, Fla. Chattahoochee Lanett. Ala. Savannah Port Wenrworth, Ga. Savannah North Augusta, S. C. Ohio Evansville, Ind. Ohio Cincinnati, Ohio Kansas Lawrence, Kans. TABLE 8. — Locations with low frequency of pesticide occurrence Rjver Location Surveys Occurrences Connecticut Wilder, Vt. 5 1 Raritan Perth Amboy, N. J. 3 1 Lake Erie Buffalo, N. Y. 5 1 St. Clair Port Huron, Mich. 4 0 Rainy International Falls, Minn. 3 0 Colorado Parker Dam, Ariz.-Calif. 5 0 Colorado Boulder City, Nev. 5 1 Truckee Farad, Calif.-Nev. 5 0 Green Dutch John, Utah 5 0 Snake American Falls, Utah 3 1 Pend Oreille Albeni Falls, Idaho 5 0 Klamath Keno, Oreg. 5 1 Columbia McNary Dam, Oreg. 5 0 Columbia Pasco, Wash. 5 1 Columbia Bonneville, Oreg. 3 1 Summary and Conclusions The occurrences of chlorinated hydrocarbon pesticides continue to be widespread. However, after reaching a peak in 1966, the total number of occurrences through- out the Country dropped sharply in 1967 and 1968. This trend is consistent with production and usage re- ports of the U.S. Department of Agriculture (8) and the U.S. Department of the Interior (9) which show a trend toward decreased use of the persistent chlorinated hy- drocarbon compounds and an increase in the use of organophosphorus and carbamate compounds. The ab- sence of a corresponding increase in the occurrences of organophosphates may be due to their relatively rapid hydrolysis rate in water and the method of analysis which was not designed specifically for this class of compounds. The data reported here and the grab sample and CAM sample data reported earlier (.1,2,3) represent pesticide levels and trends in the major interstate waterways sampled. They do not, necessarily, reflect the conditions existing in all sub-basins or areas of heavy pesticide use, such as irrigation districts. For example, in extensive surveillance operations conducted by FWQA in the Lower Colorado River area during the summers of 1967 and 1968, the occurrences were frequent and the levels generally higher for both chlorinated and organophos- phorus pesticides impithUshed data. Dieldrin continued to dominate the pesticide occur- rences, although the total number of occurrences had dropped significantly. BHC has been found consistently in the main stem of the Ohio River since 1966. The source or sources of this material have not yet been determined. The pesticide concentrations found were 1/10 to 1/500 of the permissible levels for water supplies given in Water Quality Criteria (6). However, in some instances the concentrations found have exceeded the suggested maximum reasonable stream allowance (7), as well as the environmental limit recommended by the Committee on Water Quality Criteria [6). Future surveys should be conducted to determine if the decreasing trend of chlorinated hydrocarbon pesticides occurrences is continuing. The methods of analysis should include procedures specifically designed to de- termine organophosphorus compounds. A greatly ex- panded sampling program would be necessary to deter- mine seasonal variations in pesticide occurrences. This could best be done on a regional basis. See Appendix for chemical na paper. Vol. 4, No. 2, September 1970 of compounds mentioned in this 85 Acknowledgments (4) Breidenbach, A. W.. J. J. Lichlenberg, C. F. Henkey, D. J. Smith, J. W. Eichelberger, and H. Slierli. 7966. _, , » r 11 1 1 J »i. • . c The identification and measurement of chlorinated The authors eratefully acknowledge the assistance of , , , .-,4.4; . it c r-. r auiiiv/13 gici*. u J £, hydrocarbon pesticides in surface waters. U. S. Dep. of William Middleton and James W. O'Dell in extracting the Interior. Publ. WP-22. and preparing the samples for analysis. (J) U. S. Department of the Interior, Federal Water Pollu- tion Control Administration. 1969. FWPCA method for chlorinated hydrocarbon pesticides in water and waste- water. LITERATURE CITED ^^^ ^j ^ Department of the Interior, Federal Water Pollu- tion Control Administration. 1968. Water quality cri- (7) Weaver, L., C. G. Gunnerson, A. W. Breidenbach. and teria — report of the National Technical Advisory Com- J. J. Lichtenberg. 1965. Chlorinated hydrocarbon pesti- mittee to the Secretary of the Interior, p. 20, 37, and cides in major U. S. river basins, a synoptic view. Pub- 116. lie Health Rep. 80:481-493. (7) Ettinger, M. B. and D. 1. Mount. 1967. A wild fish (2) Breidenbach, A. W., C. G. Gunnerson, F. K. Kawahara, should be safe to eat. Environ. Sci. Techno!. 1:203-205. J. ]. Lichtenberg, and R. S. Green. 1967. Chlorinated (8) U. S. Department of Agriculture, Agricultural Stabiliza- hydrocarbon pesticides in major river basins 1957-65. tion and Conservation Service. 1968. The pesticide re- Public Health Rep. 82:139-156. view— 1968. ASCS-155. p. 38-43. (3) Green, R. S., C. G. Gunnerson, and J. J. Lichtenberg. (9) U. S. Department of the Interior, Federal Water Pollu- 1967. Pesticides in our national waters. In agriculture tion Control Administration. Central Pacific Basins and the quality of our environment, Amer. Ass. for the Project. 1967. Effect of San Joaquin master drain on Advance of Sci. Publ. 85, p. 137-156. San Francisco Bay and Delta, p. 40-41. 85 Pesticides Monitoring Journal APPENDIX Chemical Names of Compounds Mentioned in This Issue ALDRIN ARAMITE® BHC CARBOPHENOTHION CHLORDANE DDD (TDE) DDE DDT (including its isomers and dehydroclilorination products) DIELDRIN ENDOSULFAN ENDRIN ETHION FENTHION HEPTACHLOR HEPTACHLOR EPOXIDE LINDANE MALATHION METHOXYCHLOR METHYL PARATHION PARATHION rOXAPHENE !,4-D !,4,5-T !,4,5-TP Not less than 95% of l,2,3,4,10,]0-hexachloro-l,4,4a,5,8,8a-hexahydro-l,4-<'«do-«xo-5,8-dimethanonaphthaIene 2-(p-ferr-butylphenoxy)-l-methylethel 2-chloroethyl sulfite 1,2,3,4,5,6-hexachlorocyclohexane, mixed isomers S-t(p-chlorophenylthio)metliyll 0,0-diethyl phosphorodithioate l,2,4,5,6,7,8,8-octachloro-3a,4,7.7a-tetrahydro-4.7-methanoindane l,l-dichIoro-2,2-bis(p-chlorophenyl)ethane: technical DDD contains some o,p'-isomer also. l,l-dichloro-2,2-bis(p-chlorophenyl)ethylene l,l,l-trichloro-2.2-bis(p-chlorophenyl)ethane; technical DDT consists of a mixture of the p.p'-isomer and the o.p'-isomer (in a ratio of about 3 or 4 to 1) Not less than 85% of 1.2,3,4,10,10-hexachloro-6,7-epoxy-l,4,4a.5,6.7,8,8a-octahydro-l,4-fndo-«o-5,8-dimethano= naphthalene 6,7,8,9,10,10-hexachloro-l,5,5a,6,9,9a-hexahydro-6.9-methano-2.4,3-benzodioxathiepin 3-oxide 1,2,3,4, 10, 10-hexachloro-6,7-epoxy-l,4,4a,5,6,7,8,8a-octahydro-l,4-endo-fndo-5,8-dimethanonaphthalene 0,0,0' ,(?'-tetraethyl 5,5'-methylene bisphosphorodithioate 0,0-dimethyl 0-[4-(methylthio)-m-tolyl] phosphorothioate l,4,5,6,7,8.8-heptachloro-3a,4,7,7a-tetrahydro-4,7-methanoindene l,4,5,6,7,8,8-heptachloro-2,3-epoxy-3a,4,7.7a-tetrahydro-4,7-methanoindan 1,2,3,4,5,6-hexachlorocyclohexane, 99% or more gamma isomer diethyl mercaptosuccinate, 5-ester with 0,0-dimethyl phosphorodithioate l,l,l-trichloro-2,2-bis(p-methoxyphenyl) ethane 0,0-dimethyl 0-p-nitrophenyl phosphorothioate 0,0-diethyl 0-p-nitrophenyl phosphorothioate chlorinated camphene containing 67% to 69% chlorine 2,4-dichlorophenoxyacetic acid 2,4,5-trichlorophenoxyacetic acid 2-(2,4,5-trichlorophenoxy) propionic acid Vol. 4, No. 2, September 1970 87 Information for Contributors The Pesticide Monitoring Journal welcomes from all sources qualified data and interpretive information which contribute to the understanding and evaluation of pesticides and their residues in relation to man and .his environment. The publication is distributed principally to scientists and technicians associated with pesticide monitoring, research, and other programs concerned with the fate of pesticides following their application. Additional circulation is maintained for persons with related in- terests, notably those in the agricultural, chemical manu- facturing, and food processing industries: medical and public health workers; and conservationists. Authors are responsible for the accuracy and validity of their data and interpretations, including tables, charts, and refer- ences. Accuracy, reliability, and limitations of the sampling and analytical methods employed must be clearly demonstrated through the use of appropriate procedures, such as recovery experiments at appropriate levels, confirmatory tests, internal standards, and inter- laboratory checks. The procedure employed should he referenced or outlined in brief form, and crucial points or modifications should be noted. Check or control samples should be employed where possible, and the sensitivity of the method should be given, particularly when very low levels of pesticides are being reported. Specific note should be made regarding correction of data for percent recoveries. Preparation of manuscripts should be in con- formance to the Srv'LE Manual for Biological Journals, American Institute of Biological Sciences, Washington, D. C, and/or the Style Manual of the United States Government Print- ing Office. An abstract (not to exceed 200 words) should accompany each manuscript submitted. All material should be submitted in duplicate (original and one carbon) and sent by first-class mail in flat form — not folded or rolled. Manuscripts should be typed on S'-i x 11 inch paper with generous margins on all sides, and each page should end with a completed para- graph. All copy, including tables and references, should be double spaced, and all pages should be num- bered. The first page of the manuscript must contain authors' full names listed under the title, with affiliations, and addresses footnoted below. Charts, illustrations, and tables, properly titled. should be appended at the end of the article with a notation in text to show where they should be inserted. Charts should be drawn so the numbers and texts will be legible when considerably reduced for publication. All drawings should be done in black ink on plain white paper. Photographs should be made on glossy paper. Details should be clear, but size is not important. The "number system" should be used for litera ture citations in the text. List references alpha betically, giving name of author/ s/, year, full title of article, exact name of periodical, volume, and inclusive pages. Pesticides ordinarily should be identified by common or generic names approved by national scientific so- cieties. The first reference to a particular pesticide should be followed by the chemical or scientific name in parentheses — assigned in accordance with Chemical Abstracts nomenclature. Structural chemical formulas should be used when appropriate. Published data and information require prior approval by the Editorial Advisory Board; however, endorsement of published in- formation by any specific Federal agency is not intended or to be implied. Authors of accepted manuscripts will receive edited typescripts for approval before type is set. After publication, senior authors will be provided with 100 reprints. Manuscripts are received and reviewed with the under- standing that they previously have not been accepted for technical publication elsewhere. If a paper has been given or is intended for presentation at a meeting, or if a significant portion of its contents has been published or submitted for publication elsewhere, notation of such should be provided. Correspondence on editorial and circulation matters should be addressed to: Mrs. Sylvia P. O'Rear. Editorial Manager. Pesticide Monitoring Journal, Pesticides Program, Food and Drug Administration. 4770 Buford Highway. BIdg. 29, Chamblee, Georgia 30341. Pesticldes Monitoring Journal The Pesticides Monitoring Journal is published quarterly under the auspices of the WORKING GROUP, Subcommittee on Pesticides, President's Cabinet Committee on the Environment, and its Panel on Pesticide Monitoring as a source of information on pesticide levels relative to man and his environment. The WORKING GROUP is comprised of representatives of the U. S. Department of Agricul- ture; Defense; the Interior; Health, Education, and Welfare; State; and Transportation. The Pesticide Monitoring Panel consists of representatives of the Agricultural Research Service, Consumer and Marketing Service, Federal Extension Service, Forest Service, Department of Defense, Fish and Wildlife Service, Geological Survey, Federal Water Quality Administration, Food and Drug Administration, Environmental Health Service, Department of Defense, National Science Foundation, and Tennessee Valley Authority. Publication of the Pesticides Monitoring Journal is carried out by the Division of Pesticide Community Studies of the Food and Drug Administration. Pesticide monitoring activities of the Federal Government, particularly in those agencies repre- sented on the Pesticide Monitoring Panel which participate in operation of the national pesticides monitoring network, are expected to be principal sources of data and interpretive articles. How- ever, pertinent data ii\ summarized form, together with interpretive discussions, are invited from both Federal and non-Federal sources, including those associated with State and community monitoring programs, universities, hospitals, and nongovernmental research institutions, both domestic and foreign. Results of studies in which monitoring data play a major or minor role or serve as support for research investigation also are welcome; however, the Journal is not intended as a primary medium for the publication of basic research. Manuscripts received for publication are reviewed by an Editorial Advisory Board established by the Monitoring Panel. Authors are given the benefit of review comments prior to publication. Editorial Advisory Board members are: Reo E. Duggan, Food and Drug Administration. Chairman Anne R. Yobs, Food and Drug Administration Andrew W. Briedenbach, Environmental Health Service Thomas W. Duke, Fish and Wildlife Service William F. Stickel, Fish and Wildlife Service Milton S. Schechter, Agricultural Research Service Paul F. Sand, Agricultural Research Service Mention of trade names or commercial sources in the Pesticides Monitoring Journal is for identification only and does not represent endorsement by any Federal agency. Address correspondence to: Mrs. Sylvia P. O'Rear Editorial Manager PESTICIDES MONITORING JOURNAL Food and Drug Administration 4770 Buford Highway, Bldg. 29 Chamblee, Georgia 30341 CONTENTS Volume 4 December 1970 Number 3 RESIDUES IN FOOD AND FEED Pesticide residues in total diet samples (V)_ P. E. Corneliussen Page 89 Monitoring DDT residues on forage plants following a forest insect control program Gerald S. Strickler and Paul J. Edgerton .106 Residues in sorghum treated with the isooctyl ester of 2,4-D- M. L. Ketchersid, O. H. Fletchall, P. W. Santelmann, and M. G. Merkle 111 RESIDUES IN FISH, WILDLIFE, AND ESTUARIES Organochlorine pesticides in nursing fur seal pups Raymond E. Anas and Alfred J. Wilson 114 International cooperative study of organochlorine pesticide residues in terrestrial and aquatic wildlife, 1967/1968— A. V. Holden 117 Organochlorine and heavy metal residues in bald eagle eggs W. C. Krantz. B. M. Mulhern, G. E. Bagley, A. Sprunt, IV, F. J. Ligas, and W. B. Robertson, Jr. Organochlorine residues and autopsy data from bald eagles 1966-68 PESTICIDES IN SOIL Monitoring pesticides in soils from areas of regular, limited, and no pesticide use Lynn J. Stevens, Carroll W. Collier, and Donald W. Woodham 136 Bernard M. Mulhern, William L. Reichel, Louis N. Locke, Thair G. Lamont, Andre Belisle, Eugene Cromartie, George E. Bagley, and Richard M. Prouty 141 .145 APPENDIX Chemical names of compounds mentioned in this issue_ .167 RESIDUES IN FOOD AND FEED p. E. Corneliussen' Pesticide Residues in Total Diet Samples (V) ABSTRACT Pesticide residue levels detected in ready-to-cat foods re- •nained at relatively low levels during the fifti) year of tlic Total Diet Study in its present form. Samples were collected 'rom 30 markets in 24 different cities. Population of cities ■■anged from less than 50.000 to 1.000,000 or more. Averages md ranges of pesticides commonly found are reported for 'he period June 196S- April 1969 by region and food class. Pesticides found infrequently also are reported for this period hy region and food r/a.M. Data showing loss of residues iirough cooking and processing of food are presented. Results of recovery studies with various classes of pesticides ire also presented. The study of pesticide residues in ready-to-eat foods. ;onducted by the Food and Drug Administration from fune 1964 through April 1968 has been described in ;arlier reports (2.3,4.9). This report covers the period fune 1968 through April 1969. Tabular data are in- ;luded comparable to that reported for the previous /ears. Vo changes were made in the sampling and compositing procedures described in the "Food and Feed Section" of he initial issue of the Pesticides Monitoring; Journal (6). Earlier reports (2. 3. -4.9) discuss data collected Torn June 1964 through April 1965, June 1965 through \pril 1966, June 1966 through April 1967. and June Field Scieniific Coordinaiion Staff in tlie OfTice of tlie Assistant Com- missioner for Field Coordination, Food and Drug Administration. U. S. Department of Health. Education and Welfare. Washington, D. C. 20204. 1967 through April 1968, respectively. On the basis of these data, average daily pesticide intake from the diet was calculated and reported elsewhere (5,7). Dietary intake for this reporting period will be discussed in a future publication. During the 4-year period, June 1964 through April 1968, the residues of most pesticide chemicals present in a high consumption well-balanced diet have been below and in most cases substantially below the limits established for acceptable daily intakes by the World Health Organization and the United Na- tions Committees (H) and in no case above the safe levels anticipated when legal tolerances were established for food. Samples were collected from 30 markets in 24 different cities. Population of cities ranged from less than 50,000 to 1 ,000,000 or more, with average sampling from the 250,000-500,000 bracket. The samples were analyzed for the presence of chlorinated hydrocarbons, organic phosphates, chlorophenoxy acids, bromides, arsenic, ami- trole, carbaryl (Sevin®), cadmium, and dithiocarbamate residues. Quantitative values reported for both chlorinated and organic phosphorus compounds were obtained by either electron capture or thermionic gas-liquid chromatog- raphy. Confirmation was made by thin layer chroma- tography and /or microcoulometric gas-liquid chroma- tography. This procedure determines chlorinated compKJunds at a sensitivity of 0.003 ppm (heptachlor epoxide) and organic phosphorus compounds at 0.05 ppm (parathion). The analytical sensitivity for both of Vol, 4, No. 3, December 1970 89 these classes of compounds varies with the individual compound being measured. Each composite was also tested for chlorophenoxy acids and esters at a sensitivity of 0.02 ppm; for amitrole at a sensitivity of 0.05 ppm; for dithiocarbamates, calculated as zineb, at a sensitivity of 0.2 ppm; for carbaryl at a sensitivity of 0.2 ppm; for bromides at a sensitivity of 0.5 ppm; and for arsenic as AsoO.T at a sensitivity of 0.1 ppm. Methods used in these studies are described in the FDA Pesticide Analytical Manual. Vol. I and II (/). No cor- rection was made for recovery studies which were per- formed continuously. Some finite residues are reported which are below the stated sensitivity levels. For ex- ample, values as low as 0.00 1 ppm for chlorinated pesticides are reported although the quantitative sen- sitivity level is 0.003 ppm. Such values are not quantita- tive but are more useful than a "trace" reporting for estimating dietary intake. Each composite was examined for cadmium during this period. Although cadmium residues do not result from pesticide usage, increased awareness of this potential hazard warranted inclusion in this study. All reported results were obtained through the use of either polarog- raphy or atomic absorption. The procedure is sensitive to 0.01 ppm cadmium (Gajan. R. J.. FDA DHEW personal communication. 1967). The vegetable and fruit composites from five market baskets in each of the geographical areas were examined for residues of chlorinated hydrocarbon, organophos- phorus, and chlorophenoxy acid pesticides both before and after preparation of the food items by dieticians. Table 3 presents comparative data for residues which were found six or more times in a given food group through the current year. The data indicate a loss in residue when food is prepared for eating through peel- ing, stripping outer leaves, cooking, etc. In most cases, the water used to cook the fruit or vegetables was discarded. This study did not determine the relative significance of any of the individual processing opera- tions in reducing the residue content of table-ready food. This report, except for Table 3, presents data only on prepared composites in accordance with past reports. Results The total of 30 market baskets examined (360 food class composites) was unchanged from previous report- ing periods. A total of 1336 residues were detected during the current reporting period. While this appears to be a marked increase over the 1088 reported for the previous period, the increase in the total number of residues was due to an increase in the number of cad- mium residues reported. Discounting cadmium report- ings for both periods, 1092 residues were detected dur- 90 ing the current period, and 1045 were detected in the previous period. The increase in cadmium reportings this period is duej to the use of either atomic absorption or polarography for analysis without the requirement of the previous period for confirmation by both techniques. Collabora- tive work and continuous recovery studies have not shown significant differences between these techniques down to the 0.01 ppm level for cadmium. There has been no significant change in the levels, frequency, or types of residues from those in the past. Thirty-four different residues were found in the sample; during the current period. The frequency of the residues is given in Table I. The most common residues, maxi- mum levels of those residues, and residues reported less frequently are discussed below for each class. DAIRY PRODUCTS: Ten chlorinated organic pesticides were found in varying combinations in 26 of the 3C composites. The most common, and their maximum values on a fat basis were: DDE (0.280 ppm); DDT (0.129 ppm); dieldrin (0.073 ppm); heptachloi epoxide (0.045 ppm); TDE (0.088 ppm); and BHC (0.03 ppm). Also present were PCP, diazinon, lindane MCP, and arsenic (As,0.,). Bromides were found (O.f ppm to 13 ppm) in 21 of the 30 composites. Cadmiurr was detected in 10 composites (0.09 ppm maximum). MEAT, FISH, AND POULTRY: Thirteen chlorinatec pesticides were found in 29 of the 30 composites in vary ing combinations. Most common were DDT, DDE TDE, dieldrin, BHC, heptachlor epoxide, and lindane with maximum values of 0.73 ppm, 0.470 ppm. 0.2! ppm, 0.170 ppm, 0.130 ppm, 0.082 ppm, and 0.03 ppm respectively, all on a fat basis. Also present wen heptachlor, PCP, aldrin, diazinon, endrin, toxaphene MCP, and malathion. Arsenic (As-.O.j) was detected If times (0.1 ppm to I.O ppm). Bromides were detected ir 18 of the 30 composites (1.0 to 28 ppm). Cadmium was detected in 21 composites (maximum of 0.06 ppm). GRAIN AND CEREAL PRODUCTS: A total of 1; chlorinated organic pesticides were found in 26 of tht 30 composites in varying combinations. Most commor were DDT, lindane, dieldrin, DDE, and TDE, with re- spective maximum values of 0.024 ppm, 0.009 ppm 0.069 ppm, 0.004 ppm, and 0.015 ppm. Thirteen com- posites contained malathion, with a maximum value ol 0.073 ppm. Also present were aldrin, diazinon, hepta- chlor epoxide, Perthanet, PCP, methyl parathion. TCNB, heptachlor, methoxychlor, ronnel, and BHC Arsenic (As^.O-,) was found in seven composites (0.1 ppm to 0.2 ppm). Bromides were detected in 27 of the 30 composites (1.0 ppm to 47 ppm). Cadmium was de- tected in 27 composites (maximum 0.08 ppm). Pesticides Monitoring Journai POTATOES: Ten chlorinated organic pesticides were found in 20 of the 30 composites. The most common pesticides found were dieldrin. DDT, DDE, and endrin, with maximum respective values of 0.006 ppm, 0.02 ppm, 0.004 ppm. and 0.009 ppm. Also detected were heptachlor epoxide, chlordane, lindane, TCNB, endosul- fan sulfate, diazinon, and TDE. Arsenic (As^.O^j) was detected in three composites, each at 0.1 ppm. Bromides were found in 21 of 30 composites (0.5 ppm to 33 ppm). Cadmium was found in 26 composites (0.02 ppm to 0.13 ppm). LEAFY VEGETABLES: A total of 14 chlorinated organic pesticides were found in 25 of 30 composites. Most commonly found were DDT, DDE, TDE, and endosulfan with maximum respective levels of 0.044 ppm, 0.032 ppm, 0.012 ppm, and 0.042 ppm. Five com- posites contained paralhion with a maximum value of 0.09 ppm. Also detected were dieldrin, lindane, BHC, Dacthal®, heptachlor epoxide, Perthane®, diazinon. methyl parathion, endrin, chlorbenside, toxaphene, disul- foton, and 2,4-D. Arsenic (As^G-^) was detected four times each at 0.1 ppm. Bromides were found 21 times (1.0 ppm to 15 ppm). Cadmium was found 27 times (0.01 ppm to 0.23 ppm). LEGUME VEGETABLES: A total of eight chlorinated organic pesticides were found in 15 composites. DDT, DDE, and TDE were found most frequently, with maximum values of 0.086 ppm, 0.006 ppm, and 0.022 ppm, respectively, one or more of these three was found in 15 of the 30 composites. Also detected were dieldrin, lindane, parathion, carbaryl, toxaphene, Dacthal®, and TCNB. Bromides were detected in 17 composites (1.0 to 32 ppm). Arsenic (AsjO^,) was detected three times, each at 0.1 ppm. Cadmium was detected 16 times (0.01 ppm to 0.03 ppm). ROOT VEGETABLES: Seven chlorinated organic pesti- cides were found in 16 of 30 composites. DDE was found 13 times at a maximum of 0.027 ppm, and DDT was found 12 times at a maximum of 0.026 ppm. Also found were TDE, dieldrin, aldrin, heptachlor, and toxaphene. Bromides were detected in 1 8 of 30 com- posites (0.5 ppm to 12 ppm). Arsenic (As^.O-j) was detected three times, each at O.I ppm, and cadmium 24 times (0.01 ppm to 0.08 ppm). GARDEN FRUITS: A total of 1 1 chlorinated organic pesticides were detected in 26 of 30 composites. Most common were DDT, DDE, TDE, dieldrin, lindane, and toxaphene at respective maximum levels of 0.140 ppm, 0.006 ppm, 0.100 ppm. 0.028 ppm, 0.004 ppm, and 0.230 ppm. Five composites contained parathion at a /OL. 4, No, 3, December 1970 maximum level of 0.033 ppm. Also found were endo- sulfan, heptachlor epoxide, PCP, malathion, endrin, diazinon, and Dacthal®. Bromides were found in 17 of 30 composites (1.0 ppm to 44 ppm). Arsenic (AS2O3) was found four times, each at 0.1 ppm and cadmium 25 times (0.01 ppm to 0.07 ppm). FRUITS: Nine chlorinated organic pesticides were found in 25 of 30 composites. DDT, DDE, TDE, and dicofol were found most frequently at maximum re- spective levels of 0.072 ppm, 0.005 ppm, 0.033 ppm, and 0.19 ppm. Ethion was found in 6 of 30 composites at a maximum 0.265 ppm level. Also found were heptachlor epoxide, endosulfan, lindane, dieldrin, car- baryl, ovex, and malathion. Bromides were found in 17 of 30 composites (0.5 to 84 ppm). Arsenic (AsjO-j) was found 5 times, each at 0.1 ppm and cadmium 15 times (0.01 ppm to 0.38 ppm). OILS, FATS, AND SHORTENING: A total of 8 chlorinated organic pesticides were found in 14 of 30 composites. Most common were DDT, DDE, TDE, BHC, and dieldrin at maximum respective levels of 0.018 ppm, 0.031 ppm. 0.022 ppm, 0.041 ppm, and 0.025 ppm. Also found were lindane, heptachlor epoxide, diazinon, PCP, and parathion. Five composites con- tained malathion; one of these levels was unusually high, at 2.99 ppm, but the laboratory could not determine the source of the residue because of an inadequate supply of the individual food components. Bromides were found in 22 of 30 composites (1.0 ppm to 70 ppm). Arsenic (AsjOj) was found twice, each at 0.1 ppm. Cadmium was found in 27 composites (0.01 ppm to 0.13 ppm). SUGARS AND ADJUNCTS: A total of 7 chlorinated organic pesticides were found in 10 of 30 composites. DDT was found in 8 of 30 composites (maximum 0.141 ppm), DDE in 6 composites (maximum 0.002 ppm), and lindane in 5 composites (maximum 0.011 ppm). Also detected were TDE, dieldrin, PCP, and MCP. Bromides were found in 19 of 30 composites (1.0 ppm to 58 ppm). Arsenic (As.,0,-j) was found in five composites, each at 0.1 ppm. Cadmium was found in 18 of 30 composites (0.01 ppm to 0.07 ppm). BEVERAGES: DDT was found once at a trace level. Bromides were found in 14 of 30 composites (1.0 ppm to 8.0 ppm). Arsenic (AsoO,,) was found in three com- posites, each at 0.1 ppm, and cadmium was found in eight composites (0.01 ppm to 0.04 ppm). Bromide reportings include naturally occurring bromides as well as residues from pesticide treatment. Of the 360 composites, 232 contained bromides above the sensitivity level of 0.5 ppm. This incidence is 64.4% as compared with 76.9%, 83.1%, and 76.8% for 1967-1968, 1966- 1967, and 1965-1966, respectively. A total of 8.6% of 91 the residues exceeded 25 ppm compared with 5.8%, 4.2% and 3.8% for the three respective earlier periods. The data obtained for the fifth year of the study are reported in detail in Table 2a, where findings are ar- ranged by food class and region. Similar information is given in Table 2b for pesticides found infrequently (less than five detections per commodity class). The data are reported in the same format used for earlier periods (2,4,9) for comparison. Trace amounts, <0.001 ppm, are not included in the averages. Where no average value is given, the results on the individual composites are shown. In these tabulations, as in the earlier reports, the bromide and arsenic values are reported on an "as is" basis for three food classes: Dairy Products (I); Meat, Fish, and Poultry (III); and Oils, Fats, and Shortening (X), even though the earlier tabulations (4) indicated a "fat basis." Cadmium results are also reported on an "as is" basis in all cases. Discussion The presence of chlorinated organic residues was con- firmed in 233 of the 360 composites examined (64.7%) for this class of chemicals. Corresponding percentages for previous years were 65.6% for 1967-1968, 62.3% for 1966-1967, and 53.8% for 1965-1966. Organic phosphorus compounds were found in 59 composites. The three previous reporting periods showed 26, 25. and 27 detections of organic phosphorus residues, respec- tively. Chlorophenoxy acids and PCP were found 14 times during the current year, with 10 of the 14 being PCP; 7 chlorophenoxy acid residues were found in 1967-1968, 8 were found in 1966-1967, and 13 were found in 1965- 1966. Carbaryl was detected in three composites. No carbaryl was found in the previous period while there were four occurrences in the 1966-1967 period. No dithiocarbamates or amitrole was detected during the current period. Unprepared fruits and vegetables were examined for dithiocarbamates before compositing in order to prevent decomposition by hydrolysis. Previ- ous report periods showed zero to four dithio-carbamate findings (calculated as zineb). Amitrole has never been found in any total diet composites. Recovery studies were conducted through the entire year with all classes of pesticides in various food groups. Table 4 gives recovery data for seven of the more com- monly occurring organochlorine pesticides, as well as data for representative pesticides in the other residue categories. It should be pointed out that each recovery experiment consisted of a single determination for the blank and a single determination for the spiked sample. Generally, these determinations were made simultaneously and often the spiking level was much less than the blank level. In other cases, not enough recovery experiments were performed to be statistically significant. Such re- covery data are not reported. Many recovery experi- ments were performed for each class of compounds, but Table 4 presents only those which fit these criteria. Based on the recovery data it is apparent that residue reportings may vary considerably from the "true" value; however, results thus far are useful in appraising the national residue picture. At low fortification levels, re- coveries from zero to 200% may be encountered. Ac- curacy is expected to increase with higher fortification levels. See Appendix for chemical names of compounds not included in Table 1. A cknowledgments The author gratefully acknowledges the analytical work from the FDA laboratories in Baltimore, Md., Boston, Mass., Kansas City, Mo., Los Angeles, Calif., Minne- apolis, Minn., and Washington, D. C. LITERATURE CITED (/) Barry. H. C, J. G. Hundley. L. Y. Johnson. 1963. (Re- vised 1964. 1965. 1968). Pesticide Analytical Manual, Vol. I and II, Food and Drug Admin., U. S. Dep. of Health, Educ, and Welfare. t2) Corneliiissen. P. E. 1969. Pesticide residues in total diet samples (IV) Pesticides Monit. J. 2(4):140-152. I3l Diiggan. R. E., H. C. Barry, and L. Y. Johnson. 1966. Pesticide residues in total diet samples. Science 151: 101-104. (4) Duggan. R. £.. H. C. Barry, and L. Y. Johnson. 1967. Pesticide residues in total diet samples (II). Pesticides Monit. J. 1(2):2-12. (5) Duggan. R. E. and G. Q. Lipscomb. 1969. Dietary in- take of pesticide chemicals in the United States (II), lune 1966-ApriI 1968. Pesticides Monit. J. 2(4):153-162. (6) Duggan. R. E. and F. J. McFarland. 1967. Assessments include raw food and feed commodities, market basket items prepared for consumption, meat samples taken at slaughter. Pesticides Monit. J. l(l):l-5. (7) Duggan. R. E. and J. R. Weatherwax. 1967. Dietary in- take of pesticide chemicals. Science 157:1006-1010. (8) Food and Agriculture Organization and World Health Organization. 1967. Evaluation of some pesticide resi- dues in food. Report of a joint meeting of the FAO Working Party and the WHO Expert Committee on Pesticide Residues. 1966. PL:CP/15; WHO/Food Add./ 67.32. (9) Martin, R. J. and R. E. Duggan. 1968. Pesticide residues in total diet samples (III). Pesticides Monit. I. l(4):ll-20. 92 PESTicroEs Monitoring Journai TABLE 1. — Number of composites where pesticide residues were found and ranges in the amounts (June 1968-Aprit 1969) No. OF PosmvE Ranges At No. OF Composites with AND Above Pesticide Composites Residltes below Sensitivity WITH Residues Sensitivity Level ' Level (PPM) CADMIUM 244 0 0.01-0.38 BROMIDES 232 0 0.5-84 DDT l.l,l-trichloro-2,2-bis(p-chlorophenyl) ethane 176 13 0.003-0.73 DDE 1 ,l-dichloro-2,2-bis(p-chlorophenyl) ethylene 142 49 0.003-0.47 TDE l,l-dichloro-2,2-bis(p-chlorophenyl) ethane 101 24 0.003-0.25 DIELDRIN not less than 85% of l,2,3.4,I0,10-hexachIoro-6,7-epoxy-l,4,4a, 5,6,7,8,8a- octahydro-l,4-e«do-fxo-5,8-dimelhanonaphthaIene 91 24 0.003-0.17 ARSENIC (AssOs) 57 0 0.1-1.0 LINDANE 1,2,3,4,5,6-hexachlorocycIohexane, 99% or more gamma isomer 48 30 0.003-0.03 HEPTACHLOR EPOXIDE l,4,5,6,7,8,8-heptachloro-2,3-epoxy-3a.4,7,7a-tetrahydro^,7-methanoindan 44 8 0.003-0.082 BHC 1,2,3,4,5,6-hexachlorocyclohexane, mixed isomers 38 7 0.003-0.13 MALATHION diethyl mercaptosuccinate. 5-ester with 0,0-dimethyl phosphorodithioate 21 13 0.054-2.99 ENDOSULFAN 6,7,8,9, 10, 10-hexachloro-l, 5, 5a,6,9.9a-hexahydro-6.9-methano- 2,4,3-ben2odioxathiepin 3-oxide 19 1 0.003-0.033 DIAZINON 0,0-diethyl 0-(2-isopropyl-6-methyI-4-pyrimidinyl ) phosphorothioate 14 14 DICOFOL (KELTHANE®) 4,4'-dichloro-a-(trichloromethyl) benzhydrol 13 0 0.004-0.19 rOXAPHENE chlorinated camphene containing 67% to 69% chlorine 13 1 0.022-0.33 ENDRIN l,2,3,4,10,I0-hexachloro-6,7-epoxy-l,4,4a,5,6,7,8,8a-octahydro- l,4-endo-frirfo-5,8-dimethanonaphthalene 12 5 0.003-0.019 PARATHION 0.0-diethyl O-p-nitrophenyl phosphorothioate 12 12 PCP pentachlorophenol 10 5 0.02-0.04 ETHION 0,0,0',0'-tetraethyl 5,5'-methylene bisphosphorodithioate 6 5 0.265 DACTHAL® 2,3,5,6-tetrachloroterephthalic acid dimethyl ester 6 5 0.032 HEPTACHLOR l,4,5,6,7,8,8-heptachloro-3a,4,7,7a-tetrahydro-4,7-methanoindene 6 2 0.003-0.014 ALDRIN not less than 95% of I,2,3,4.10,10-hexachloro-I,4,4a,5,8.8a-hexahydro- l,4-€nrfo-exo-5,8-dimethanonaphthalene 5 4 0.004 METHYL PARATHION 0,0-dimethyl O-p-nitrophenyl phosphorothioate 5 5 PERTHANE® l,l-dichloro-2,2-bis(p-ethylphenyl) ethane 4 0 0.01-0.528 TCNB l,2,4,5-tetrachIoro-3-nitrobenzene 3 2 0.007 MCP 4-chloro-2-methyl-phenoxyacetic acid 3 1 0.04-0.047 CARBARYL 1-naphthyl methylcarbamate 3 2 0.3 CHLORDANE 1,2,4,5,6,7,8, 8-octachloro-3a,4,7,7a-tetrahydro-4,7-methanoindane 2 0 0.026-0.043 CHLORBENSIDE p-chlorobenzyl p-chlorophenyl sulfide 1 0 0.029 METHOXYCHLOR l,l,l-trichloro-2,2-bis(p-methoxyphenyl) ethane 1 1 RONNEL 0,0-dimethyl 0-2.4,5-trichlorophenyl phosphorothioate 1 1 2,4-D 2,4-dichlorophenoxyacetic acid 1 1 DISULFOTON 0,0-dimethyl 5-2-(ethylthio) ethyl phosphorodithioate 1 1 OVEX p-chlorophenyl p-chlorobenzenesulfonaie 1 0 0.003 ' Pesticide chemicals capable of being detected by the specified analytical methodology may be confirmed qualitatively but are not quantifiable when they are present at concentrations below the sensitivity level. Vol. 4, No. 3, December 1970 93 TABLE 2a. — Levels of pesticide residues commonly found — by food class and region (June 1968-April 1969} [T = Trace <0.001 PPM] Los Angeles I. Dairy Products (8-13% Fat) ' Residues In Parts Per MiUion — Fat Basis DDT Average Positive Composites Number Range 0.027 3 0.014-0.129 0.036 4 0.015-0.113 0.030 4 0.03-0.074 0.004 4 0.004-0.007 0.019 5 0.014-0.034 DDE Average Positive Composites Number Range 0.014 3 0.021-0.037 0.018 5 0.015-0.034 0.190 5 0.14-0.280 0.005 4 0.005-0.010 0.015 6 0.007-0.020 TDE Average Positive Composites Number Range 0.020 3 0.01-0.088 I T •0.019 4 0.01-0.042 0.004 4 0.004-0.007 0.006 4 0.008-0.01.2 DIELDRIN Average Positive Composites Number Range 0 0.052 6 0.028-0.073 0.021 4 0.019-0.059 1 0.004 0.020 6 0.010-0.039 HEPTACHLOR EPOXIDE Average Positive Composites Number Range 0.01 1 3 0.015-0.030 0.034 6 0.025-0.045 0 0.002 2 0.004-0.005 0.013 6 0.007-0.024 BHC Average Positive Composites Number Range 1 0.017 0.010 4 0.008-0.020 0.014 4 0.01-0.03 0 0.007 5 0.005-0.013 TOTAL BROMIDES Average Positive Composites Number Range 1.0 5 1.0-2.0 3.0 3 1.0-13 2.0 4 2.0-3.0 1.5 5 0.5-5,0 3.5 4 3.0-7.0 CADMIUM Average Positive Composites Number Range 0.02 3 0.01-0.09 0 0.01 3 0.01-0.02 <0.01 2 0.01 <0.01 2 0.01-0.02 II. Meat, Fish, and Poultry (17-23% Fat) ' Residues In Parts Per Million— Fat Basis DDT Average Positive Composites Number 0.287 5 0.055 6 0.092 6 0.020 4 0.052 6 Range 0.145-0.73 0.029-0.083 0.016-0.147 0.014-0.055 0.030-0.076 DDE Average Positive Composites Number 0.171 5 0.039 6 0.240 6 0.019 4 0.033 6 Range 0.099-0.337 0.016-0.088 0.020-0.470 0.014-0.049 0.018-0.049 TDE Average Positive Composites Number 0.076 3 0.045 4 0.043 6 0.015 0.034 6 Range 0.076-0.22 T-0.25 0.011-0.079 0.034-0.056 0.010-0.076 DIELDRIN Average Positive Composites Number 0.030 2 0.026 5 0.030 6 1 0.015 6 Range 0.129-0.170 0.011-0.078 0.006-0.077 0.005 0.009-0.029 94 Pesticides Monitoring Journal : TABLE 2a. — Levels of pesticide residues commonly found — by food class and region (June 1968-April 1969) — Continued Kansas City Los Angeles Minneapolis II. Meat, Fish, and Poultry (17-23% Fat) ' — Continued Residues In Parts Per Million — Fat Basis HEPTACHLOR EPOXIDE Average Positive Composites Number Range 0.024 3 0.027-0.082 0.017 6 0.008-0.036 0 0 0.013 6 0.005-0.021 BHC Average Positive Composites Number Range 1 0.053 0.028 6 T-0.130 0.008 4 T-0.02 0 0.003 5 T-0.009 ^INDANE Average Positive Composites Number Range 0.003 2 T-0.016 0 0.007 0.014-0.03 0 0.004 4 0.003-0.014 kRSENIC (As=03) Average Positive Composites Number Range 0.2 4 0.2-0.4 0 0.1 4 0.1-0.4 0.4 4 0.2-1.0 <0.1 3 0.1 OTAL BROMIDES Average Positive Composites Number Range 6.0 5 2.0-17 5.5 3 1.0-28 2.0 2 4.0-7.0 3.5 3 4.0-9.0 5.5 5 2.0-14 ;admium Average Positive Composites Number Range 0.02 4 0.01-0.06 0.0 1 4 0.01-0,02 0.01 4 0.01-0.03 0.02 5 0.02-0.04 0.01 4 0.01 III. Grain and Cereal > Residues In Parts Per Million )DT Average Positive Composites Number Range 0.009 5 0.007-0.024 0.004 5 T-O.OlO 0.006 4 0.003-0.016 0.002 3 0.003-0.005 0,005 5 0.002-0.011 )DE Average Positive Composites Number Range 0.001 3 T-0.004 <0.001 4 T-0.002 1 0.001 0 0.001 5 T-0.003 TDE Average Positive Composites Number Range 0.003 2 0.003-0.015 1 T <0.001 2 0.001-0.002 0 0.001 3 0.001-0.005 DIELDRIN Average Positive Composites Number Range 0.012 3 T-0.069 T 0.003 3 0.005-0.006 1 0.02 0.004 5 0.003-0.006 -INDANE Average Positive Composites Number Range 0.003 3 T-0.009 0.001 4 T-0.002 0.001 3 0.002 O.OOI 2 0.002-0.003 0.002 5 0.001-0.007 vlALATHION Average Positive Composites Number Range 0.013 2 0.036-0.041 0.041 4 0.043-0.073 0.006 3 0.005-0.02 0 0.026 4 0.014-0.060 Vol. 4, No. 3, December 1 1 970 95 TABLE 2a. — Levels of pesticide residues commonly found — by food class and region (June 1968-April 1969) — Continued Kansas City Los Angeles III. Grain and Cereal ' — Continued Residues In Parts Per Million ARSENIC (As=03) Average 0.1 <0.1 Positive Composites Number 4 0 1 0 2 Range 0.1 0.1 0.1-0.2 TOTAL BROMIDES Average 12 30 6.0 25 14 Positive Composites Number 6 6 4 6 5 Range 7.0-17 13-47 4,0-12 15-36 1.0-43 CADMIUM Average 0.04 0.03 0.04 0.03 0.02 Positive Composites Number 5 6 5 6 5 Range 0.02-0.08 0.02-0.04 0.03-0.06 0.02-0.05 0.02-0.04 IV. Potatoes ^ Residues In Parts Per Million DDT Average Positive Composites Number Range 0.002 5 T-0.006 1 T 1 0.001 0.006 2 0.019-0.02 1 T DDE Average Positive Composites Number Range <0.001 2 T-0.003 T <0.001 2 T-O.OOl I 0.004 0 DIELDRIN Average Positive Composites Number Range 0.001 2 0.001-0.004 0.002 3 0.004-0.006 0.002 5 0.001-0.005 0 1 T ENDRIN Average Positive Composites Number Range T 0 0.004 5 0.002-0.009 0 0.004 5 0.002-0.009 TOTAL BROMIDES Average Positive Composites Number Range 3.5 5 0.5-8 8.5 4 3.0-33 3.5 4 3.0-9.0 10 4 4.0-32 3.5 4 3.0-9.0 CADMIUM Average Positive Composites Number Range 0.04 6 0.02-0.09 0.04 5 0.03-0.07 0.04 5 0.03-0.06 0.05 5 0.02-0.13 0.04 5 0.03-0.06 V. Leafy Vegetables ^ Residues In Parts Per Million DDT Average Positive Composites Number 0.008 5 0.005 3 0.022 5 0.003 3 0.014 5 Range T-0.022 0.003-0.023 0.009-0.044 0.003-0.010 0.007-0.031 DDE Average Positive Composites Number 1 1 0.005 3 1 0.008 3 Range T 0.007 0.004-0.020 0.020 0.007-0.032 96 Pesticides Monitoring Journal : 1 TABLE 2a. — Levels of pesticide residues commonly found — by food class and region (June 1968-April 1969) — Continued Pesticide Boston Kansas City Los Angeles Baltimore Minneapolis V. Leafy Vegetables ' — Continued Residues In Parts Per Million NDOSULFAN (TOTAL) Aicr.iee 0.010 0.016 Positive Composites Number 2 1 0 1 4 Range 0.018-0.042 0.003 T 0.014-0.033 DE Average 0.002 0.001 Positive Composites Number 0 0 3 0 2 Range T-0.012 0.003-0.004 ARATHION Average 0.002 Positive Composites Number 0 1 2 1 1 Range 0.007 0.003-0.007 0.09 0.002 OTAL BROMIDES Average 3.0 5.0 1.5 6.0 2.5 Pi^Mtive Composites Number 5 3 4 6 3 Range 1.0-9.0 2.0-15 2.0-3.0 1.0-10 2.0-7.0 ADMIUM Average 0.06 0.08 0.04 0.04 0.04 Positive Composites Number 6 6 6 5 4 lI Range 0.02-0.10 0.01-0.2.1 0.02-0.06 0.02-0.12 0.03-0.10 VI. Legume Vegetables > 1 Residues In Parts Per Million )DT Average 0.001 T 0.024 0.015 Positive Composites Number 2 2 3 0 2 Range 0.002-0.004 T 0.020-0.071 0.005-0.086 )DE Average T 0.002 0.001 Positive Composites Number 3 I 2 0 3 Range T T 0.005-0.006 T-0.005 DE Average 0.003 0.003 Positive Composites Number 1 0 2 0 4 Range 0.022 0.008-0.010 T-O.OII OTAL BROMIDES Average 2.5 6.0 I.O 2.0 5.5 Positive Composites Number 5 2 2 3 5 Range 1.0-7.0 5.0-32 1.0-6.0 3.0-5.0 2.0-10 ADMIUM Average 0.01 <0.01 0.01 0.01 Positive Composites Number 5 2 4 4 1 Range 0.01-0.03 0.01-0.02 0.01 0.01-0.02 0.0 1 VII. Root Vegetables ■ Residues In Parts Per Million >DT Average 0.001 0.004 0.002 0.007 Positive Composites Number 2 3 3 1 3 Range T-0.005 0.003-0.014 T-0.009 0.003 0.004-0.026 /OL. 4, No. 3, December 1 970 97 TABLE 2a. — Levels of pesticide residues commonly found — by food class and region (June 1 968- April /969j— Continued Pesticide Boston Kansas City Los Angeles Baltimore Minneapolis VII. Root Vegetables ' — Continued Residues In Parts Per Million DDE Average Positive Composites Number Range 0.001 1 0.008 0.001 3 T-0.008 0.006 5 T-0.015 1 O.COl 0.007 3 0.001-0.027 TOTAL BROMIDES Average Positive Composites Number Range 3.0 4 1.0-10 2.0 3 0.5-12 1.5 3 2.0-5.0 4,0 4 3.0-12 3.0 4 1.0-7.0 CADMIUM Average Positive Composites Number Range 0.03 5 0.02-0.05 0.02 4 0.02-0.04 0.03 5 0.01-0.08 0.02 5 0.01-0.06 0.01 5 0.01-0.03 VIII. Garden Fruits 1 Residues In Parts Per Million DDT Average Positive Composites Number Range 0.003 2 0.004-0.015 0.032 6 0.008-0.066 0.048 6 0.007-0.140 0.020 2 0.021-0.102 0.037 4 0.013-0.100 DDE Average Positive Composites Number Range 1 T 0.002 3 0.002-0.005 0.002 5 0.001-0.006 0 0.002 4 0.001-0.005 TDE Average Positive Composites Number Range 1 0.010 0.026 4 0.016-0.100 0.004 3 0.003-0.017 0.006 3 0.004-0.025 0.012 6 0.005-0.016 DIELDRIN Average Positive Composites Number Range 0.006 2 0.006-0.028 0.001 3 T-0.005 0.002 4 0.001-0.003 1 0.002 1 0.006 LINDANE Average Positive Composites Number Range 1 T 1 0.003 0.001 2 0.002-0.004 0 T 2 T TOXAPHENE Average Positive Composites Number Range 0 0 0.056 4 0.03-0.230 0 0.038 2 0.077-0.150 PARATHION Average Positive Composites Number Range 0 0 <0.001 2 T-0.002 0 0.010 3 0.007-0.033 TOTAL BROMIDES Average Positive Composites Number Range 2.0 4 1.0-6.0 8.5 3 2.0-44 1.0 3 1.0^.0 6.0 3 1.0-19 2.5 4 2.0-5.0 CADMIUM Average Positive Composites Number Range 0.03 6 0.01-0.07 0.01 3 0.02-0.03 0.02 5 0.02-0.04 0.02 6 0.01-0.04 ! 0.01 5 0.01-0.02 98 Pesticides Monitoring Journal TABLE 2a. — Levels of pesticide residues commonly found — by food class and region (June 1968- April 7969)— Continued Pesticide Boston Kansas City Los Angeles Baltimore Minneapolis IX. Fruits 1 Residues In Parts Per Million DDT Average Positive Composites Number Range 0.010 5 0.006-0.021 0.010 4 0.003-0.032 0.014 4 0.003-0.072 I 0.004 0.003 4 0.003-0.005 DDE Average Positive Composites Number Range 0.001 4 T-0.004 0.001 2 0.003 0.002 4 0.001-0.005 1 0.003 <0.001 3 0.001 roE Average Positive Composites Number Range 0.003 2 0.007-0.011 0.007 2 0.009-0.033 0.007 2 0.009-0.033 0.004 2 0.001-0.025 0.002 3 0.001-0.005 3ICOFOL Average Positive Composites Number Range 1 0.058 0.052 3 0 019-0.148 0.083 6 0.007-0.19 0 0.005 3 0.004-0.017 =THION Average Positive Composites Number Range 1 0.006 0.007 2 0.010-0.033 1 0.008 0 0.047 2 0.017-0.265 TOTAL BROMIDES Average Positive Composites Number Range 2.0 5 0.5-5.0 0.5 2 1.0-3.0 2.5 3 2.0-10 19 4 0.5-84 2.5 3 2.0-10 \RSENIC (AsaOs) Average Positive Composites Number Range <0.1 2 0.1 0 <0.1 2 0.1 0 1 0.1 :admium Average Positive Composites Number Range 0.01 4 0.01-0.02 <0.01 3 0.01 0.02 2 0.01-0.1 0.01 4 0.01-0.02 0.06 2 0.01-0.38 X. Oils, Fats, and Shortening (83-88% Fat) i Residues In Parts Per Million — Fat Basis DDT Average Positive Composites Number Range 0 0.004 3 0.007-0.009 O.OOI 2 T-0.004 1 0.003 0.005 2 0.011-0.018 ODE Average Positive Composites Number Range 1 0.031 0.003 3 0.002-0.008 1 0.005 0 0.004 4 0.004-0.008 FDE Average Positive Composites Number Range 1 T 0.001 3 T-0.004 0.002 2 T-O.OlO 0 0.007 4 0.004-0.022 Vol. 4, No. 3, December 1 ?70 99 TABLE 2a. — Levels of pesticide residues commonly found — by food class and region (June 1968-April 1969) — Continued Kansas City Los Angeles X. Oils, Fats, and Shortening (83-88% Fat) ' — Continued Residues In Parts Per Million— Fat Basis DIELDRIN Average Positive Composites Number Range 0.007 2 0.016-0.025 T 2 T-0.002 1 0.008 0 1 0.005 BHC Average Positive Composites Number Range 0 0.003 2 T-0.017 0.004 0.004-0.02 0 1 0.041 MALATHION Average Positive Composites Number Range 0.507 2 0.054-2.99 0 0 0 0.008 3 0.002-0.025 TOTAL BROMIDES Average Positive Composites Number Range 5.5 6 1.0-9.0 8.5 5 4.0-19 4.0 3 4.0-11 14 3 4.0-70 15 5 3.0-36 CADMIUM Average Positive Composites Number Range 0.02 6 0.02-0.03 0.02 5 0.02-0.03 0.04 5 0.02-0.11 0.02 5 0.02-0.04 0.04 6 0.01-0.13 XI. Sugars and Adjuncts ^ Residues In Parts Per Million DDT Average Positive Composites Number Range 0.024 2 T-0.141 0 0.001 2 0.003 1 0.004 0.002 3 0.002-0.004 DDE Average Positive Composites Number Range T 2 T 0 0.001 2 0.002 0 <0.001 2 0.001-0.002 LINDANE Average Positive Composites Number Range 0.002 2 T-O.OU 0 0.002 0 <0.001 2 0.001 ARSENIC (As20a) Average Positive Composites Number Range <0.1 3 0.1 0 0 1 0.1 1 0.1 TOTAL BROMIDES Average Positive Composites Number Range 3.5 4 3.0-8.0 2.5 3 3.0-9.0 3.0 3 3.0-8.0 4.5 4 1.0-18 23 5 4.0-58 CADMIUM Average Positive Composites Number Range 0.01 4 0.01-0.03 0.01 3 0.02-0.03 0.02 5 0.01-0.07 0.01 3 0.01-0.03 0.01 3 0.01-0.02 100 Pesticides Monitoring Journai TABLE 2a. — Levels of pesticide residues commonly found — by food class and region (June 1968-April 1969) — Continued Kansas City Los Angeles Minneapolis XII. Beverages ^ Residues In Parts Per Million "OTAL BROMIDES Average Positive Composites Number :admium Average Positive Composites Number Range 2.0 3 2.0-5.0 0.01 3 0.01-0.04 Six composite samples examined at each of five sampling sites: Boston. Kansas City, Los Angeles, Baltimore, and Minneapolis. ^OTE: Bromide, cadmium, and arsenic values are reported on an "as is" basis (no drying or isolation of fat) for Dairy Products; Meat, Fish, and Poultry; and Oils, Fats, and Shortening. TABLE 2b. — Pesticides found infrequently — fcy food class and region (June 1968-April 1969) No. Com- posites I. (a) Dairy Products (8-13% Fat) ' Residues In Parts Per Million — Fat Basis \rsenic (AsjOs) ^eptachlor Diazinon MCP Boston Los Angeles Boston Los Angeles Boston Minneapolis Minneapolis Minneapolis Boston 0.005, 0.013 0.010 0.1 0.1 0.008 0.008 0.047 II. (a) Meat, Fish, and Poultry (17-23% Fat) : Residues In Parts Per Million — ^Fat Basis Heptachlor PCP Malathion Aldrin Diazinon Endrin Toxaphene IMCP Boston Minneapolis Boston Los Angeles Boston Boston Minneapolis Los Angeles Los Angeles Baltimore 0.014 0.004 0.054 0.004 0.011 0.019 0.19 0.04 No. Com- posites III. (a) Grain and Cereal ' Residues In Parts Per Million Aldrin Boston 2 T Diazinon Los Angeles Miimeapolis T 0.002, 0.003 Heptachlor epoxide Kansas City Minneapolis T 0.001 Perth ane® Minneapolis 0.010, 0.038 PCP Boston Los Angeles 0.03 0.020 Methyl parathion Boston 0.033 Ronnel Kansas City T Heptachlor Minneapolis T Methoxychlor Kansas City T TCNB Boston T BHC Los Angeles T IV. (a) Potatoes ' Residues In Parts Per Million Heptachlor epoxide Boston Kansas City Los Angeles T, 0.009 0.010 0.004 Arsenic (AsiOa) Boston Los Angeles Miimeapolis 0.1 0.1 0.1 Chlordane Kansas City Los Angeles 0.043 0.026 Vol. 4, No. 3, December 1970 101 TABLE 2b. — Pesticides found infrequenlly — by food class and region (June 1968-April 1969) — Continued rV. (a) Potatoes " — Continued Residues In Parts Per Million V. (a) Leafy Vegetables' Residues In Parts Per Million VI. (a) Legume Vegetables' Residues In Parts Per Million Lindane Los Angeles 2 T Endosulfan sulfate Minneapolis 2 0.004, 0.011 Diazinon Los Angeles Minneapolis 1 1 T T TCNB Boston 1 0.007 TDE Minneapolis 1 0.001 Dieldrin Kansas City Los Angeles Minneapolis T 0.001 0.001, 0.003 Methyl parathion Boston Los Angeles Minneapolis 0.008 T 0.001, 0.025 Lindane Boston Kansas City Minneapolis T 0.007 0.002 Endrin Boston Minneapolis 0.017 0.002 Toxaphene Los Angeles T, 0.022, 0.33 Arsenic {AS2O3) Boston Los Angeles Minneapolis 0.1 0.1 0.1 Dacthal® Kansas City Minneapolis 0.032 T, 0.004 Diazinon Los Angeles Minneapolis T 0.003, 0.0O4 BHC Kansas City Minneapolis 0.004 0.001 Perthane® Minneapolis 0.142, 0.528 Heptachlor epoxide Minneapolis T Chlorbenside Minneapolis 0.029 Disulfoton Boston 0.002 2,4-D Boston 0.012 Arsenic (AS2O3) Boston Los Angeles 1 0.1 0.1 Toxaphene Los Angeles 2 0.04, 0.052 Carbaryl Baltimore Minneapolis T T Dieldrin Boston 0.006 Lindane Kansas City T Parathion Kansas City 0.035 Dacthal® Los Angeles 0.005 TCNB Boston T No. Com- posites VII. (a) Root Vegetables ' Residues In Parts Per Million VIII. (a) Garden Fruits' Residues In Parts Per Million IX. (a) Fruits Residues In Parts Per Million Dieldrin Boston Kansas City 1 2 T T, 0.007 Arsenic (AsiOs) Boston Minneapolis 2 1 0.1 0.1 TDE Boston Los Angeles 1 0.001 T Heptachlor Minneapolis 2 0.002, 0.003 Aldrin Minneapolis 1 0.001 Toxaphene Los Angeles 1 0.036 Arsenic (AS2O3) Boston Los Angeles 3 0.1 0.1 Endosulfan Kansas City Los Angeles Minneapolis 2 0.002 0.001, 0.007 T Endrin Los Angeles 2 T, 0.002 Diazinon Los Angeles Minneapolis T 0.004 Dacthal® Minneapolis T, 0.001 Heptachlor epoxide Minneapolis T PCP Los Angeles T Malathion Los Angeles T Dieldrin Boston Kansas City Minneapolis 1 1 1 T T 0.002 Heptachlor epoxide Minneapolis 2 T, 0.001 Endosulfan Minneapolis 2 0.002, 0.010 Lindane Los Angeles 1 T Malathion Minneapolis 1 0.004 Ovex Minneapolis 1 0.003 Carbaryl Minneapolis 1 0.3 X. (a) Oils. Fats, and Shortening (83-88% Fat) ■ Residues In Parts Per Million — Fat Basis Lindane Kansas City Los Angeles 1 2 0.002 0.002, 0.006 Heptachlor epoxide Boston Kansas City 1 1 T 0.012 Diazinon Minneapolis 2 0.002, 0.011 Arsenic (Asi'Os) Boston Minneapohs 1 1 0.1 0.1 Parathion Los Angeles 1 T PCP Los Angeles 1 T 102 Pesticides Monitoring Journai TABLE 2b. — Pesticides found infrequently — by food class and region (June 1968-April 1969) — Continued No. Com- posites No. Com- posites XI. (a) Sugars and Adjuncts ^ Residues In Parts Per Million XII. (a) Beverages Residues In Parts Per Million IDE Los Angeles Minneapolis 2 1 0.001 0.003 Dieldrin Boston Los Angeles 1 1 T 0.003 MCP Los Angeles 1 0.010 PCP Los Angeles 1 0.040 Arsenic (AS2O3) Boston Baltimore ' Six composite samples examined at each of five sampling sites: Boston, Kansas City, Los Angeles, Baltimore, and Minneapolis. NOTE: Bromide and arsenic values are reported on an "as is" basis (no drying or isolation of fat) for Dairy Products; Meats, Fish, and Poul- try: and Oils, Fats, and Shortening. TABLE 3 — Comparison of residues before and after processing by dietician (average PPM levels for residues found six or more times per food group) Potatoes Leafy Vegetables Legume Vegetables Root Vegetables Garden Fruits Fruits Average Retention of Residues (%) Pesticide IV V VI VII VIII IX Residues After Prep'n. Before After Before After Before After Before After Before After Before After DDT DDE TDE Dieldrin Lindane Endosulfan (total) Keith ane® Toxaphene Endrin Ethion o.ou 0.001 0.001 0.002 0.002 T 0.001 0.001 0.023 0.003 0.011 0.010 0.004 0.006 0.027 0.001 0.001 0.010 0.001 0.00 1 0.041 0.023 0.002 0.002 0.043 0.001 0.009 0.002 0.001 0.043 0.030 0.001 0.007 0.002 T 0.021 0.011 0.001 0.001 0.045 0.024 0.008 0.001 0.0O4 0.034 0.014 40 32 109 100 Not Calculable 55 76 49 50 58 TABLE 4. — Recovery studies, June 1968-April 1969 [( ) = Averages] Type of Spike Blank Total Number of Pesticide Food Level — Level — Recovered — Recovery Composite PPM PPM PPM Experiments Heptachlor epoxide 0.000 Fatty 0.005 0.004-0.005 2 Fatty 0.030 0.005 0.025 1 Non-Fatty 0.003 0.000-0.001 (T) 0.002-0.003 6 Non-Fatty 0.010 0.000 0.005-0.010 (0.008) DDT Fatty 0.050 0.000-0.023 (0.010) 0.041-0.069 (0.050) 6 Non-Fatty 0.003 0.000-0.002 (0.001) 0.003-0.004 (0.004) 4 Non-Fatty 0.010 0.000-0.007 (0.003) 0.005-0.017 (0.015) 6 Non-Fatty 0.030 0.000 0.012 1 TDE Fatty 0.010 0.000 0.009-0.010 2 Non-Fatty 0.003 0.000 0.003-0.005 4 Non-Fatty 0.050 0.000-0.016 (0.009) 0.052-0.076 (0.067) Vol. 4, No. 3, December 1970 103 TABLE 4. — Recovery studies, June 1968-April 1969 — Continued Type of Spike Blank Total Number of Pesticide Food Level— Level — Recovered — Recovery Composite PPM PPM PPM Experiments DDE Fatty 0.050 0.007 0.030 1 Non-Fatty 0.003 0.00O-0.0O2 (T) 0.0O3-O.0O5 (0.004) 5 Non-Fatty 0.010 0.000-0.006 (0.001) 0.009-0.020 (0.013) 7 Non-Fatty 0.100 0.000 0.089-0.115 (0.101) 6 Dieldrin Fatty O.OIO 0.000 0.006-0.015 2 Non-Fatty 0.003 0.000 0.003-O.004 7 Non-Fatty 0.050 0.000^.006 (0.001) 0.020-0.058 (0.049) 7 Aldrin Fatty 0.003 0.000 0.002 2 Non-Fatty 0.010 0.000-0.002 (0.001) 0.009-0.013 (0.010) 5 Non-Fatty 0.050 0.000 0.026-0.059 (0.048) 6 Endrin Fatty 0.010 0.000 0.008 1 Fatty 0.030 0.000 0.023 1 Non-Fatty 0.003 0.000 T-0.006 (0.003) 5 Non-Fatty 0.010 0.000 0.004-0.015 (0.011) 7 Non-Fatty 0.030 0.000 0.017 1 Malathion Fatty 0.05 0-0.015 (0.008) 0.04-0.06 (0.05) 2 Non-Fatty 0.05 0-O.057 (0.024) 0-0.087 (0.066) 6 Non-Fatty 0.1 0 0.061-0.112 (0.087) 13 ParathioD Non-Fatty 0.05 0-0.008 (0.003) 0.014-0.058 (0.O44) 12 Non-Fatty 0.1 0 0.076-O.16O (0.102) 6 Non-Fatty 1.0 0 0.88-1.01 (0.95) 3 Ethion Non-Fatty 0.05 0 0.015-0.070 (0.046) 12 Non-Fatty 0.1 0 0.023-0.110 (0.0«1) 12 2,4-DB Fatty 0.02 0 T-0.033 (0.015) 4 Fatty 0.05 0 0.02-0.06 (0.04) 3 Non-Fatty 0.02 0 0.02 (0.02) 2 Non-Fatty 0.05 0 0.04-0.06 (0.05) 3 Non-Fatty 1.0 0 0.69-1.26 (0.96) 4 2,4,5-TP Fatty 0.02 0 0-0 01 (0.003) 4 Non-Fatty 0.5 0 0.18-0.68 (0.48) 9 PCP Fatty 0.02 0 0-0.007 (0.003) 6 Non-Fatty 0.05 0 0-0.01 (0.003) 6 Non-Fatty 0.1 0 0-0.02 (0.003) 6 Carbaryl Non-Fatty 0.2 0 0.15-0.20 (0.18) 26 Non-Fatty 1.0 0 0.50-1.1 (0.88) 23 Amitrole Non-Fatty 0.05 0 0.02-0.10 (0.05) 23 Non-Fatty 0.1 0 0.05-0.14 (0.08) 22 104 Pesticides Monitoring Journa TABLE 4. —Recovery studies, June 1968-April 1969 — Continued Pesticide Type of Food Composite Spike Level — PPM Blank Level — PPM Total Recovered — PPM Number of Recovery Experiments rsenic (AsjOa) romides admium (Polarography and A. A.) ineb Fatty Fatty Non-Fatty Non-Fatty Non-Fatty Fatty Fatty Non-Fatty Non-Fatty Fatty Fatty Non-Fatty Non-Fatty Non-Fatty Non-Fatty 0.1 0.5 0.1 0.5 1.0 5.0 SO 5.0 50 0.1 0.05 0.01 0.1 5.0 1.0 (M).4 (0.16) 0-0.3 (0.06) 0-0.1 (T) 0-0.1 (T) 0-0.1 (T) 0-5.0 (1.2) 1.5-6.0 (3.2) 0-2.7 (1.1) 0^3 (8.7) 0-0.09 (0.02) 0-0.02 (0.01) 0-0.01 (T) 0-0.03 (0.01) 0 0 0.1-0.56 (0.25) 0.06-0.8 (0.40) 0.05-0.25 (0.11) 0.1-0.56 (0.43) 0.5-1.0 (0.82) 1.0-11.5 (5.9) 15-46 (35.2) 2.9-9.4 (6.0) 12.5-82.0 (50.8) 0-0.21 (0.11) 0.03-0.08 (0.05) O4).02 (0.01) 0.10-0.13 (O.U) 1-13 (3.5) 0.4-0.9 (0.74) 6 13 12 12 12 13 3 6 15 20 6 5 4 30 6 Vol. 4, No. 3, December 1970 105 Monitoring DDT Residues on Forage Plants Following a Forest Insect Control Program Gerald S. Strickler and Paul J. Edgerton' ABSTRACT The amount of DDT reaching understory vegetation grazed by cattle, deer, and elk, and the DDT residue levels in herbage samples of a sedge, lupine, and sagebrush were deter- mined for one prespray and three postspray sampling dates up to 1 year following aerial application of DDT for forest insect control. The DDT was applied at the start of the livestock grazing period. Ground level DDT dosage ranged form 3-78% of the designated %-lb/acre rate. Average prespray DDT residue was 0.54, 11.43, and 3.53 ppm for elk sedge, lupine, and sagebrush, respectively. Corresponding postspray averages were 74.04, 87.08, and 61.02 ppm immediately after appli- cation; 13.66, 13.55, and 6.62 ppm for 4 months: and 2.07, 0.41, and 1.22 ppm 12 months after application. Residues in species plot samples from the first two postspray dates were significantly related to ground level dosage. Cycling of DDT was not indicated; the greater elk sedge residue 1 year after spraying was attributed to differences in sampling. Associated DDT residues in cattle and big game are briefly discussed. Introduction DDT, at the rate of % of a pound in 1 gallon of fuel oil per acre, was applied to 66,000 acres of forest of ponderosa pine and Douglas-fir north of Bums, Oreg., in June 1965, to control an outbreak of Douglas-fir tussock moth, Hemerocampa pseudotsugata McD. Since a late spring application of DDT was necessary to con- trol the moth, forage plants grazed by livestock, deer, and elk were exposed to DDT spray residues at the beginning of the summer grazing season. We, therefore, initiated a study to measure the amount of DDT reach- ing understory vegetation following helicopter applica- tion and to monitor DDT residues on three forage plants for one prespray and three postspray sampling dates. The latter three sampling dates were immediately after spray application, approximately 4 months after spray ing when summer livestock grazing was terminated, an( 12 months after spraying just prior to livestock grazinj in the succeeding year. Crouch and Perkins (2) graphically presented a sum mary of the DDT residues obtained in this study in surveillance report of the insect control project. Th^ purpose of this paper is to make data available on canop; cover, spray deposition rate, and subsequent residui content of plants grazed by livestock, deer, and elk. Sampling Procedure The plants selected for herbage residue analysis wer elk sedge (Care.x geyeri Boott). tailcup lupine (Lupinu caudatits Kell.), and big sagebrush Artemisia tridentati Nutt.). Sixteen Vi-acre plots were located in the 23,000 acre Antelope Mountain unit of the spray project (Fig 1). The Antelope Mountain acreage encompassed par of a larger allotment grazed by cattle. Because elk sedgt and lupine are highly desirable forages for cattle, plot containing these species were fenced to prevent cattli grazing. All plots were accessible to deer and elk. Within elk sedge and lupine plots, 20- to 50-g (dn weight) subsamples of herbage were clipped from eacl of 20 randomly located points. On sagebrush plots simUaj subsamples were obtained from the crowns of 21 shrubs The area around each sampling point and shrub centei was divided into quarters. Clipping was confined to on£ quarter at each point or shrub during each of four sampl- ing dates: ^ Pacific Northwest Forest and Range Experiment Station, Forest Serv- ice, U.S. Department of Agriculture, La Grande, Oreg. Prespray First postspray Second postspray Third postspray May 26-June 9, 1965 June 21-27, 1965 Oct. 10-11, 1965 June 28-29, 1966 106 Pesticides Monitoring Journal FIGURE 1. — Diagram of the Antelope Mountain unit and location of sampling plots. (Insufficient forage prevented sampling of plots 3, 7, 16, 20, and 21.) SENECA, OREGON 28 Miles LEGEND Unit Boundary ^^ Road System Sample Plots • 0 Vi 1 mile >ize of the area clipped within each quarter varied with he distribution of the 20 to 50 g of herbage required or the sample. herbage samples consisted of leaf and twig growth of )ig sagebrush, all current growth of tailcup lupine, and loth current and older (2-3 years) green leaves of elk edge. Subsample clippings were composited, yielding ap- )roximately 1 to 2 lb of dry herbage per plot for DDT inalyses. To prevent possible transfer of DDT between species amples, clean plastic gloves were worn during clipping, ind shears and weight scales were rinsed thoroughly vith fresh acetone before each species sample was :lipped. Care was taken not to contaminate samples vith ground surface litter. The plastic sample bags were leld off the ground by wire holders. Composited samples were refrigerated on the day :lipped. stored at approximately 0 F, and shipped frozen o the Agricultural Research Service Laboratory in t'akima. Wash., for DDT residue analysis. All samples vere kept frozen until processed for analysis. fust before spraying, nine oil-sensitive dye cards (10) vere placed in a grid pattern within each Vi-acre sample slot. The cards were supported by wire holders (4) ipproximately 1.5 feet above ground. Cards were not Jlaced within 1 .5 feet of a tree bole or under sagebrush ;rowns. Cards were collected after spraying, and the imount of spray (gal/acre) reaching the card level was ;stimated (3,5). Percentage cover of tree canopy was Tieasured with a densiometer (9) at each dye-card loca- tion to determine if canopy cover differences were as- sociated with differences in spray deposition. Residue Analysis Elk sedge samples were chopped and mixed in a reel-type cutting machine. Samples of sagebrush and lupine were ground and mixed in a Toledo meat grinder. Small subsamples from each of the above species plot samples were composited and oven-dried to determine moisture content at the time of extraction. For each species two subsamples per plot, each weighing 25, 50, or 100 g net, depending on species and sampling date, were either tumbled for 1 hour or stee()ed over- night and then tumbled for 1 hour in measured amounts of n-hexane, filtered through cotton plugs into sample bottles, and refrigerated until analyzed. Aliquots of the extracts were shaken for 6 minutes with 4 g of a mixture of Florisil (60/100 mesh), Magnesol, and anhydrous sodium sulfate. The glassware was rinsed with a total of 75 ml of n-hexane which was combined with the filtrates. The filtrates were reduced in a warm water bath with an airstream to a volume of about 5 ml and transferred to test tubes. For those samples analyzed by an adaptation of a colorimetric method (S), the solvent was completely removed. For those analyzed by a Warner-Chilcott gas chromatograph, the filtrates were brought to dryness and the residues dis- solved in a measured amount of n-hexane and re- frigerated. Prespray, first postspray, and second postspray sedge and lupine samples were analyzed colorimetrically. The residue figures from the first and second postspray samples were reduced by the amount of "apparent" DDT in the corresponding prespray samples. The second postspray sagebrush samples and third post- spray samples of all species were analyzed by gas chro- matography and were not reduced by the "apparent" DDT in prespray samples. One untreated sagebrush sample collected near the laboratory when the second postspray samples were analyzed did not show any inter- ference of plant waxes with the gas chromatograph method, so no reductions were made. The third postspray residue values were reduced by values obtained from a complete reagent blank. Identity of pesticides was con- firmed by determination of extraction p-values (7). The gas chromatograph employed a 183-cm coiled glass column 3.8 mm i.d. packed with 10% DC 200 on Gas Chrom Q. The electron capture detector employed Strontium 90. Operating temperatures were: column-233 C, injector-250 C, detector-256 C, and the outlet-277 C. Vol. 4, No. 3, December 1970 107 The nitrogen carrier gas flow rate was 160/ minute, and nitrogen scavenger gas flow was 40 ml/minute at 2.82 kg/cm^ (40 psi). Standard DDT solutions were added to each species sample solution for each sappling period, and the per- cent recovery was determined. Average recovery effi- ciencies for the combined DDT isomers and DDE in each species ranged from 87-101% for the coiori- metric method and 78-85% for the gas chromatograph method. All residue values were corrected to 100% re- covery and to a dry-weight basis. Results Tree canopy cover, DDT deposition rate, and mean DDT residue* levels in the herbage samples from the 16 plots are presented in Table 1. Prespray lupine, sagebrush, and elk sedge samples con- tained 11.43, 3.53, and 0.54 ppm DDT residues, re- spectively. Residues in sagebrush and lupine were un- expectedly high and indicated previous DDT application; however, an examination of management history of the study area did not reveal any definite sources of con- tamination. There was no significant relationship between average canopy cover and average DDT deposition rate estimated from the dye cards. Variation in climatic conditions during the 2-week operation and in amounts of spray delivered by the helicopters resulted in different amounts of spray reaching the understory regardless of canopy cover differences. On the other hand, postspray plot sample residues were directly related to the average deposition rate. This relationship was significant (P ^ .05) for all but the third postspray elk sedge and sage- brush samples. Lupine, elk sedge, and sagebrush herbage had average DDT residue levels of 87.08, 74.04, and 61.02 ppm, respectively, immediately after spraying and 13.55, 13.66, and 6.62 ppm 4 months after spraying. Except for plot 8, the pattern of decreasing DDT levels was essen- tially the same for all sample plots. Cattle had broken into plot 8 following the first postspray sampling, but because of the abundance of elk sedge the light grazing which occurred did not prevent subsequent sampling. Since the second postspray sample from plot 8 was the only sample increasing in residue, we assume that the cattle contaminated the herbage, after having previously grazed in vegetation with high DDT residue. A similar incident, but with heavier grazing of herbage, precluded sampling of both elk sedge and lupine in plot 13. ' The term DDT residue is used in this paper with no distinction be- tween deposited material and that resuhing from decomposition or weathering processes or between materials within or on the plant. One year after spraying, average residue levels de- creased to 0.41, 1.22, and 2.07 ppm for lupine, sage- brush, and elk sedge, respectively. For lupine and sage- brush, these averages were lower than prespray averages, but elk sedge residue was about four times higher. Analysis of paired lupine and elk sedge samples clipped from the same plot showed that their DDT residue levels were significantly different (P ^ .05) only for prespray and third postspray samples. But, whereas lupine samples had the greater residue content before spraying, elk sedge residue content surpassed that in lupine 1 year after spraying. Discussion In this study, DDT residues in understory plants reached a high level immediately following aerial spray, de- creased rapidly the following 4 months, and were less than, or similar to, prespray levels 1 year after spraying. These findings are comparable to those reported for other forest insect control projects (6,7). The origin of the high DDT residues in the prespray samples is conjectural. Clarification might have been obtained if nonsprayed control samples from areas adjacent to the plots had been available for comparison at each date. The low residues 1 year after spraying indicate that cycling of DDT did not occur. It should he remembered, however, that this assumption is based on ( 1 ) a com- parison with surprisingly high prespray DDT residues in lupine and sagebrush and (2) our belief that the greater residue content in elk sedge compared with that in lupine and sagebrush 1 year after spraying was pri- marily due to differences in sampling. All clipped samples of lupine, which had the lowest residue content 1 year after spraying, were new growth. Clipped elk sedge samples, which had the highest residue content, necessarily contained a large percentage of 2- and 3- year-old green leaves which had been sprayed. Residue content of sagebrush was intermediate to that of lupine and elk sedge, and it was noted that in the process of stripping leaves from twigs some leaves and inflores- cences of the previous season were included in its samples. Thus, differences in residue content 1 year after spraying were most likely an artifact of sampling: higher residues were measured in samples containing more herbage which was present during spray application. We selected these species not only to provide information on plants of different growth form but also because they supply forage for cattle, deer, and elk. It should be noted that residue content of the first postspray samples of the three species tended to be more similar on a fresh-weight basis (field sample), especially for samples clipped from the same or adjacent plots. We. therefore. 108 Pesticides Monitoring Journal suspect that all understory plants available to grazing animals had DDT residue contents similar to those sampled. It was also apparent from the residues re- maining in the second postspray samples of these species that grazing animals were consuming forage with high DDT residue content during the 4-month grazing season. Companion studies showed that DDT residues in adipose tissue of cattle, deer, and elk also increased after spray- ing and subsequently declined, with average residues declining below the 7 ppm legal tolerance 120 days (deer) and 150 days (cattle) after spraying (2). How- ever, these residue values are considered low since both big game and cattle had access to and undoubtedly TABLE 1 . — Plot canopy cover, spray deposition rate at ground level, and DDT residues on elk sedge, lupine, and sagebrush Sample Plot Number Average Canopy Cover » (Percent) Average Deposition Rate' (Gal/acre) DDT Residue in PPM = First Postspray June 1965 Second Postspray Oct. 1965 Third Postspray June 1966 ELK SEDGE 2 76 0.21 0.68 62.50 16.90 1.86 5 63 0.40 1.71 123.40 15.00 2.79 6 46 0.36 (') 35.60 8.20 3.45 8 60 0.03 0.55 10.00 13.20 1.16 10 46 0.15 0.35 33.60 5.80 0.68 11 50 0.73 0.32 278.70 25.40 1.83 13 66 0.22 0.55 32.10 (*) 1.14 14 68 0.46 1.39 54.60 14.20 2.43 15 51 0.31 (») 86.60 13.90 3.32 18 55 0.27 (') 40.00 7.80 1.20 19 48 0.57 0.35 57.40 16.20 2.88 Plot average 57 0.34 0.54 74.04 13.66 2.07 Herbage moisture content (percent) 69 59 48 51 TAILCUP LUPINE 6 46 0.36 9.92 69.00 11.40 0.35 10 46 0.15 9.92 56.20 3.40 0.32 11 50 0.73 29.00 218.10 18.70 0.56 13 66 0.22 (') 62.20 (') (') 15 51 0.31 5.00 43.60 16.60 0.30 18 55 0.27 3.33 65.70 12.40 0.42 19 48 0.57 (•) 94.80 18.80 0.51 Plot average 52 0.37 11.43 87.08 13.55 0.41 Herbage moisture content (percent) 88 79 10 78 BIG SAGEBRUSH 1 46 0.24 2.98 45.10 6.20 1.68 4 33 0.38 4.48 37.80 6.00 1.06 9 18 0.17 2.98 27.30 4.10 0.64 12 23 0.78 1.98 156.20 13.20 2.11 17 26 0.19 5.23 38.70 3.60 0.60 Plot average 29 0.35 3.53 61.02 6.62 1.22 Herbage moisture content (percent) 56 67 48 68 Average of nine values per sample plot. Each value is the average of two analyses per herbage sample based on dry weight and 100% recovery. The colorimetric method used gives the sum of the two isomers of DDT and the metabolite DDE. The lower limit of detectability was 0.04 ppm for combined DDT isomers and DDE. The gas chromatograph method, used for the second postspray sagebrush samples and ail species for the third postspray samples, gives the sum of DDE. o,p'-DDT + TDE, and p.p'-DDT. The lower limits of detectability are 0.002, 0.005, and 0.04 ppm, respectively. Postspray values are corrected for "apparent" DDT found in prespray samples or in a complete reagent blank (third postspray samples). None detected. Not sampled because of insufficient herbage or accidental grazing by cattle. Vol. 4, No. 3, December 1970 109 grazed on untreated forage. The period and amount of grazing on the sprayed area, by big game particularly, is unknown. Nevertheless, in our study cattle contained average DDT residues well below the legal tolerance level 1 year after spraying and were grazing major forage plants no more contaminated with DDT than was measured before insect control operations. See Appendix for chemical names of compounds mentioned in this paper. This paper reports research involving pesticides. It does not contain recommendations for their use nor does it imply that the uses dis- cussed here have been registered. All uses of pesticides must be registered by appropriate State and/or Federal agencies before they can be recommended. Mention of trade names or commercial materials is for the con- venience of the reader and does not constitute any preferential endorsement by the U.S. Department of Agriculture over similar products available. A cknowledgments The authors thank Dr. Bohdan Maksymiuk, Pacific Northwest Forest and Range Experiment Station, for estimating spray atomization and spray deposits on the oil-sensitive cards, and piersonnel of the Malheur Na- tional Forest for materials and services they so freely offered. DDT residue analyses were under the supervi- sion of H. W. Rusk, L. I. Butler, and L. M. McDonough, Entomology Research Division. Pesticide Chemicals Re- search Branch Station, Agricultural Research Service, Yakima, Wash. LITERATURE CITED (1) Beroza, Morton, and M. C. Bowman. 1965. Identifica- tion of pesticides at nanogram level by extraction p- values. Anal. Cham. 37:291-292. (2) Crouch, Glenn L., and Randall F. Perkins. 1968. Sur- veillance report — 1965 Bums project, Douglas-fir tus- sock moth control. USDA Forest Serv. R-6 Rep. 20 p. (3) Davis. J. M.. and K. R. Elliott. 1954. Standards for estimating surplus spray deposits on oil-sensitive cards. Forest Serv., unnumbered pamphlet. U.S. Dep. of Agr. 7 p. (4) Maksymiuk. Bohdan. 1959. Improved holders for spray deposit assessment cards. J. Econ. Entomol. 52:1029- 1030. (5) . 1963. How to estimate the atomization of oil- base aerial sprays by the D-max method. USDA Forest Serv. Res. Note WO- 1, 6 p. (6) Mussehl, Thomas W., and Robert B. Finley, Jr. 1967. Residues of DDT in forest grouse following spruce budworm spraying. J. Wildlife Manage. 31:270-287. (7) Pillmore. Richard £.. and Robert B. Finley. Jr. 1963. Residues in game animals resulting from forest and range insecticide applications. Trans. N. Amer. Wildlife Conf. :S:409-422. (8) Stiff, Henry A., Jr., and Julia C. Castillo. 1945. A colorimetric method for the micro-determination of 2, 2, bis (p-chlorophenyl) 1,1,1 trichlorethane (DDT). Science 101(2626):440-443. (9) Strickler, Gerald S. 1959. Use of the densiometer to estimate density of forest canopy on permanent sample plots. USDA Forest Serv. Pacific Northwest Forest & Range Exp. Sta. Res. Note 180, 5 p. (10) While, H. W. 1959. How to make oil-sensitive cards for estimating airplane spray deposit. Forest Serv. Forest Insect Lab.. Beltsville. Md. U.S. Dep. of Agr. 3 p. 110 Pesticides Monitoring Journal Residues in Sorghum Treated With the Isooctyl Ester of 2,4-D ^ M. L. Ketchersid.= O. H. Fletchall.' P. W. Santelmann,' and M. G. Merkle ' ABSTRACT An isooctyl ester formulation of 2,4-D was applied to grain sorghum at acid equivalent rates of 1.25 or 2.50 lb/acre. The growth stage of the sorghum at the time of treatment varied from preemergence to the dough stage. Residues of 2,4-D in the grain and forage of sorghum were determined by gas chromatography with a sensitivity level of approxi- mately 0.2 ppm. No residues of 2,4-D were detected in the grain: residues in the forage ranged from less than 0.2 ppm to 5.25 ppm. The time interval between application of the herbicide and harvesting of the forage was the most critical factor in determining the residues. More 2,4-D remained in sorghum forage grown in Missouri than in forage grown in Oklahoma. Introduction Low volatile esters of 2,4-D have been used to control broadleaved weeds in grain sorghum [sorghum hicolor (L.) Moench] since the late 1940s. Registration of 2,4-D was obtained on the basis of no 2,4-D residue in the crop at harvest time. Since that time governmental agencies have required that finite tolerances be estab- lished for all herbicides used in the production of field crops. This report describes a chromatographic pro- cedure for detecting 2,4-D in sorghum forage and grain to a sensitivity of 0.2 ppm and summarizes the results of analyses of sorghum treated at various stages of growth in the field. ^ Contribution from the Agricultural Experiment Stations of Missouri, Journal Series No. 6012, approved 6-24-70; Oklahoma, Journal Series No. 2028, approved 6-9-70; and Texas, Journal Series No. 8572. ap- proved 6-2-70. - Department of Soil and Crop Sciences, College of Agriculture. Texas A&M University, College Station, Texas 77843. 3 University of Missouri, Columbia, Mo. * Oklahoma State University, Stillwater, Okla. Procedures for the determination of 2,4-D esters in plants have been published by Crafts (7) using radio- active tagging, Morre and Rogers (5) using three bioas- says, Szabo (6) using paper chromatography. Yip and Ney (7) using microcoulometric gas chromatography, and Hagin and Linscott (2) using electron capture gas chromatography. These studies indicate that the ester is hydrolyzed upon absorption by the plant and sub- sequently moved in the plant as the free acid. Klingman et al. (3) found that about 75% of the 2,4-D applied to forage as the 2-ethylhexyl ester had been hydrolyzed to the free acid within Vz hour after application. Morgan and Hall (4) studied the metabolism of 2,4-D acid by cotton and grain sorghum and found no 2,4-D in the grain. Either the 2,4-D was not translocated to the seed or it was destroyed or otherwise immobilized before reaching the developing fruit. Yip and Ney (7) used an alkaline hydrolysis of milk and an acid hydrolysis of forage to release possible bound forms of 2,4-D. The hydrolysis made it possible to analyze for the acid, ester, and bound forms of 2,4-D in a single sample. Treatment and Sampling Procedures MISSOURI At Columbia, Mo., grain sorghum (variety AKS-614) was planted at the rate of 4 lb/ acre approximately I'i inches deep on July 17, 1969. The soil was a Mexico silt loam which had been spring plowed and had re- ceived 500 lb/acre of 20-10-10 bulk fertilizer. The experimental design was a randomized complete block with four replications. Each plot consisted of two 40- inch rows, 32 feet long. The preemergent treatment and treatment at the 15- to 20-inch stage were applied with a plot sprayer mounted on a small garden tractor; the 30-inch stage Vol. 4, No. 3, December 1970 111 was treated with an oiling rig. Both types of equipment applied 40 gal/acre at 40 psi. The application dates were July 17, August 11, and August 27, respectively. The rainfall for July was 5.01 inches; August, 2.40 inches; and September, 6.82 inches. Even though the grain was immature, the plots were harvested on October 17, 1969, before the killing frost date. The heads and the stalks were separated, placed in plastic bags, and frozen. The samples were kept in frozen storage until shipment at which time they were packed in dry ice and shipped to Texas for analysis. OKLAHOMA OK612 grain sorghum was planted in 40-inch rows on a Norge loam soil near Perkins in northcentral Oklahoma. Planting was done with a commercial planter on June 5, 1969. The plots were three rows wide and 30 feet long in four replications. The treatments were made with a tractor-sprayer. The herbicide was applied in 30 gallons of water per acre at 30 psi. The treatment on June 10 was made to a dry soil when the air tempera- ture was 70 F and the wind speed was 3 mph. There was 0.75 inches of rain on June 1 1. Treatment at the 10-inch stage was made on June 21 when the air temperature was 80 F and the wind speed was 6 mph. Soil moisture was adequate and the sorghum was growing vigorously. There was 0.8 inches of rain on June 24. On July 28 the sorghum plants were flowering when treated; the air temperature was 80 F; there was no wind; and soil moisture was adequate. Treatment of plants in the dough stage was made on August 19; there was no wind; the air temperature was 90 F; and soil moisture was ad- equate. TTie sorghum was separated into grain or for- age samples, frozen, and shipped to Texas for analysis. A nalytical Procedures Forage samples were chopped into pieces 1 cm long or smaller and thoroughly mixed before taking a 25-g portion for analysis. Most of the grain was in the dough stage and required no grinding prior to blending. The 25-g samples of both forage and grain were heated to a boil in 400-ml beakers containing 25 ml of 0.05 n HCl. After cooling, each sample was transferred to a blending cup containing 100-500 ml of 2-propanol and blended in a Waring blendor for 10 minutes. The homogenate was filtered through glass wool in a Buchner funnel. The blending cup was rinsed twice with 25 ml of 2-propanol and the rinses added to the residue in the Buchner funnel. The filtrate was returned to the beaker and 100 ml of 1 N KOH solution added. This mixture was evap- orated on a warm hot plate to a volume of approximately 150 ml. The samples were kept under a fume hood over- night at 30 C. This hydrolysis converted all of the ester of 2,4-D to the potassium salt and possibly released bound 2,4-D from the plant filtrate. The basic solution was washed three times with petroleum ether. The aqueous phase was kept and the ether phase discarded. A number of these discards were analyzed to confirm that no ester of 2,4-D was in the ether. The sample was acidified by dropwise addition of concentrated HCl and extracted three times with 100 ml of petroleum ether. After evaporating the ether to dryness, the residue was methylated by heating the sample with 10 ml of 12.5% BF-j-methanol solution until BF, fumes were released. The methyl ester of 2,4-D was then extracted into 10 ml of hexane. Samples were analyzed on a Barber Coleman Model 5360 electron capture gas chromatograph equipped with a radium 226 detector. A 6-foot spiral column packed with 10% DC 200 on 100/200 mesh gas chrom-Q was used. Injector, column, and detector temperatures were 260 C. 200 C and 240 C, respectively. The carrier gas was nitrogen at a flow rate of 75 ml/min. Peak height of known concentrations of 2,4-D methyl ester were compared to the unknowns to determine the amount of herbicide present. Results and Discussion The alkaline hydrolysis of the isooctyl ester of 2,4-D to 2,4-D acid was essentially complete since equivalent weights of ester and acid yielded equal amounts of the methyl ester of 2,4-D (Fig. 1). There was a linear relationship between the peak height on the chromato- gram and the concentration of 2,4-D over a range of 0 to 1.0 fig/ ml. A direct transesterification of the isooctyl ester of 2,4-D to the methyl ester of 2,4-D with BF.j-methanol is possible, but hydrolysis of the isooctyl FIGURE 1. — Standard recovery curve showing the relative response of standard solutions of 2,4-D acid and 2,4-D isooctyl ester acid equivalent which were carried through the procedure using chemicals only. Two iil were injected for each sample. 112 Pesticides Monitoring Journal ester to the acid prior to methylation facilitates the re- moval of impurities from the sample. When an alkaline hydrolysis was used, the percent recovery of 2.4-D ester from either sorghum grain or forage was approximately 86%. The detection limits of the procedure are ap- proximately 0.2 ppm of 2,4-D when 25-g sorghum samples are used. If it is desirable to detect lower con- centrations of 2,4-D, larger samples must be taken. However, if larger samples are taken, larger quantities of solvents are required for the extraction, or the per- centage of recovery may be reduced. Also, background peaks are more troublesome when larger samples are taken. In Oklahoma, sorghum 10 to 12 inches tall treated with 1 .25 lb/acre of the isooctyl ester of 2,4-D contained 1 .06 ppm 2 days after application and 0.90 ppm 1 week after application. Forage samples taken 4 weeks after application or later contained no detectable residues (<0.2 ppm). Likewise, grain harvested from the treated plots contained no detectable residues (Table 1). TABLE 1.- — Residues of 2,4-D in sorghum forage and grain sprayed on June 21, 1969, with 2,4-D isooctyl ester at an acid equivalent rate of 1.25 lb /acre when sorghum was 10- to 12-inches tall (Perkins, Okla.) Date Residues in PPM Sampled Forage Grain June 23, 1969 June 28, 1969 July 19, 1969 Aug. 20. 1969 1.06 0.90 <.2 <.2 <.2 The residues in sorghum forage and grain at harvest following preemergent application with 1.25 lb/ acre of 2,4-D and treatment when the sorghum was 10 inches tall, when the sorghum was flowering, and when the sorghum heads were in the dough stage, are shown in Table 2. Application made when the sorghum was flowering or in the dough stage resulted in detectable residues of 2,4-D in the sorghum forage at harvest. No residues of 2,4-D were detected in the grain at any time, indicating that sorghum does not readily incorporate 2,4-D into the grain even if the herbicide is applied when the grain is being formed. TABLE 2. — Residues of 2,4-D in sorghum forage and grain sprayed at several growth stages with 2,4-D isooctyl ester at an acid equivalent rate of 1.25 lb/acre, and harvested September 19, 1969 (Perkins. Okla.) Treatment Residues at Harvest (PPM) Date Stage Forage Grain June 10, 1969 June 21, 1969 July 28, 1969 Aug. 19, 1969 Check Preemergence 10 inches Flowering Dough <.2 <.2 0.36 3.16 <.2 <.2 <.2 <.2 <.2 <.2 Table 3 shows data from sorghum grown in Missouri and treated at different stages of growth with various rates of 2,4-D. As with the Oklahoma samples, no 2,4-D residues were detected in the grain at anytime; however, 2,4-D persisted longer in the sorghum forage from Mis- souri than in the forage from Oklahoma. For example, Missouri sorghum 15 to 20 inches tall, treated with 1.25 lb/ acre of 2,4-D, and harvested about 2 months later, contained 1.19 ppm. Oklahoma sorghum treated at a similar rate when the sorghum was flowering, and harvested about 2 months later, contained only 0.36 ppm. Likewise, sorghum in Missouri treated 6 weeks prior to harvest with 2.50 lb/acre of 2,4-D contained almost twice the residue found in Oklahoma sorghum treated 4 weeks prior to harvest with 1.25 lb/ acre of 2,4-D. TTie hot, dry weather conditions in Oklahoma apparently increased the degradation of 2,4-D. TABLE 3. — Residues of 2,4-D in sorghum forage and grain sprayed with 2,4-D isooctyl ester at various rates and stages of growth, and harvested October 17, 1969 (University of Missouri, Columbia, Mo.) E Treatment Stage Acid Equivalent Residues at Harvest (PPM) Rate (LB/ACRE) Forage Grain July 17, 1969 Preemergence 2 <.2 <.2 Aug. 11 1969 15-20 inches 1.25 1.19 <.2 Aug. U 1969 15-20 inches 2.50 1.84 <.2 Aug. 27 Check 1969 30 inches (flowering) 2.50 0 5.25 <.2 <.2 <.2 See Appendix for chemical name of 2,4-D. A cknowledgment The authors thank the Thompson-Hayward Chemical Company for providing the chemical used in the study. LITERATURE CITED (1) Crafts, A. S. 1960. Evidence for hydrolysis of esters of 2,4-D during absorption by plants. Weeds 8:19. (2) Hagin, R. D. and D. L. Linscott. 1965. Determination of 4-(2,4-dichIorophenoxy)-butyric acid (2,4-DB) and 2,4- dichlorophenoxyacetic acid in forage plants. J. Agr. Food Chem. 13:123. (3) Klingman, D. L.. C. H. Gordon. G. Yip, and H. P. Burchfield. 1966. Residues in the forage and in milk from cows grazing forage treated with esters of 2,4-D. Weeds 14:164. (4) Morgan, P. W. and IV. C. Hall. 1963. Metabolism of 2,4-D by cotton and grain sorghum. Weeds 11:130. (5) Morre, D. J. and B. J. Rogers. I960. The fate of long chain esters of 2.4-D in plants. Weeds 8:436. (6) Szabo, S. S. 1963. The hydrolysis of 2,4-D esters by bean and com plants. Weeds 11:292. (7) Yip, G. and R. E. Ney, Jr. 1966. Analysis of 2,4-D resi- dues in milk and forage. Weeds 14:167. Vol. 4, No. 3, December 1970 113 RESIDUES IN FISH, WILDLIFE, AND ESTUARIES Organochlorine Pesticides in Nursing Fur Seal Pups Raymond E. Anas' and Alfred J. Wilson, Jr.^ ABSTRACT Samples of muscle, brain, liver, blubber, and ingested milk from five nursing fur seal pups, Callorhinus ursinus, were analyzed for organochlorine pesticides and polychlorinaled biphenyl compounds (PCB's). Of 25 samples of tissue and ingested milk, 25 contained DDE; 19, DDD; 22, DDT: and 9 contained dieldrin. All of the samples of ingested milk and tissues, with the exception of brain tissue, contained trace amounts of PCB's. Concentrations of DDE, DDD, and DDT in the liver and brain of 4-month-old nursing pups tended to be higher than those found in 4-month fetuses studied previously. Pesticides are known to occur in seals in Antarctica (5, 9), Scotland (6), and Canada (6). Adult and fetal northern fur seals, Callorhinus ursinus, collected on the Pribilof Islands, Alaska, in 1968 and off the Washington coast in 1969 also had detectable amounts of pesticides (2). This report records the amounts of organochlorine pesticides found in the tissues and ingested milk of nurs- ing northern fur seal pups collected on the Pribilof Islands in 1969. The pups were about 4 months old and had nursed approximately 12 times, but we do not know whether they had previously fed on marine organisms. In a study by Abegglen et al. (7), four pups collected from October 6-27, 1961, on St. Paul Island were found to contain remains of either fish or amphipods. ' National Marine Fisheries Service, Marine Mammal Biological Lab- oratory, Seattle, Wash. 98115. - Gulf Breeze Laboratory, Department of the Interior, Gulf Breeze, Fla. Of 20 pups sampled on November 14 and 15, 1966, on St. Paul Island, 9 contained milk. Of these nine, one pup showed positive evidence of having fed on marine organisms, and five additional pups may have fed on marine organisms (personal communication. H. Kaji- mura and M. C. Keyes, Marine Mammal Biological Laboratory. Seattle. Wash.). The amounts of pesticides accumulated from foraging is not known. However, since seal milk contains about 50*% fat by weight (5), it is a natural reservoir for pesticides. The seal pup, therefore, cannot avoid accumulating pesticides each time it nurses if the mothers milk contains pesticides. Knowledge of pesticides in seal milk and in tissues of nursing pups is important, because mortality of fur seal pups might be associated with accumulation of pesticides. Sampling and Analytical Procedures Samples of muscle, brain, liver, blubber, and ingested milk were collected from five nursing fur seal pups on November 10, 1969. The tissues were kept frozen from the time of collection until analysis. Blubber, brain, liver, and muscle tissues were analyzed for BHC, heptachlor, aldrin, heptachlor epoxide, toxa- phene, methoxychlor, dieldrin, endrin, and the o.p'- and p,p'-isomers of DDE, DDD, and DDT. Samples of the thawed tissues weighing approximately 10 g each were mixed with anhydrous sodium sulfate in a blender. Each mixture was extracted for 4 hours with petroleum ether in a Soxhlet apparatus. Extracts were concentrated and partitioned with acetonitrile. The acetonitrile was evap- orated just to dryness at room temperature and the resi- due eluted from a Florisil column (S) . Milk samples were analyzed by the method described by Giuffrida (4). 114 Pesticides Monitoring Journal Sample extracts were then identified and quantified by gas chromatographs equipped with electron capture de- tectors and 5' x Vs" o.d. glass columns. Operating conditions are outlined in Table 1 : Laboratory tests indicated recovery rates greater than 85% for pesticides found in tissues. Data in this report do not include a correction factor for percentage re- covery. The lower limit of sensitivity is 0.01 ppm (mg/ kg, wet weight). All residues reported are on a wet- weight basis except milk, which was calculated on a fat basis. Sample extracts had to be concentrated to small volumes (volumes less than that required to obtain 0.010 ppm sensitivity for pesticides) in order to evaluate the level of polychlorinated biphenyl compounds (PCB's). These compounds were present at levels that would not signif- icantly interfere with the quantitation of DDT and its metabolites. Thin-layer chromatography was used to separate and confirm the presence of these organochlo- rine compounds. Results Pesticides were found in every sample of tissue and ingested milk collected from the nursing fur seal pups. All 20 of the tissue samples contained DDE; 14 con- tained DDD; 17, DDT; and 5 contained dieldrin; all except those from the brain had trace amounts of polychlorinated biphenyls (Table 2). The highest con- centration of a single pesticide was 45 ppm of DDE in the blubber of one pup. Brain tissue contained the lowest concentrations of DDT and its metabolites. On the average, the levels of DDE, DDD, and DDT tended to be lower in brain and liver tissues of fur seal fetuses (2) than in those of nursing pups in the present study (Table 3). Samples of blubber and muscle were not collected from the fetuses. All of the samples of milk contained DDE, DDD, and DDT; four contained diel- drin. Trace amounts of PCB's were also found in all samples of ingested milk. Pesticides in the milk, especially residues of DDE, varied widely in amount, but nursing pups with the largest amounts of pesticides in the milk generally had the largest amounts of pesticides in body tissues. See Appendix for chemical names of compounds mentioned in this paper. A cknowledgments We thank Jerrold Forester and Johnnie Knight, Bureau of Commercial Fisheries, Pesticide Field Station, Gulf Breeze, Fla., for their assistance in the chemical analysis, and Lavrenty Stepetin, Bureau of Commercial Fisheries, St. Paul Island, Alaska, for collecting the samples. LITERATURE CITED (/) Abegglen. C. E., A. Y. Roppel, A. M. Johnson, and F. Wilke. 1961. Fur seal investigations, Pribilof Islands, Alaska. Report of field activities, June-November 1961. 148 p. (U.S. Fish Wildlife Serv., Marine Mammal Biological Laboratory, Seattle, Wash.) (2) Anas, R. E. and A. J. Wilson, Jr. 1970. Organochlorine pesticides in fur seals. Pesticides Monit. J. 3(4):198-200. (3) Ashworth. V. S., G. D. Ramaiah, and M. C. Keyes. 1966. Species diflference in the composition of milk with special reference to the northern fur seal. J. Dairy Sci. 49(10): 1206-1211. (4) Giiiffiida, L., D. C. Bostwick, and N. F. Ives. 1966. Rapid cleanup techniques for chlorinated pesticide resi- dues in milk, fats, and oils. J. Ass. Offic. Agr. Chem. 49(3):634-638. 15) George. J. L. and D. E. H. Frear. 1966. Pesticides in the Antarctic. J. Appl. Ecol. 3(Suppl.):155-167. (6) Holden, A. V. and K. Marsden. 1967. Organochlorine pesticides in seals and porpoises. Nature 216(5122): 1274-1276. (7} Bureau of Commercial Fisheries, Marine Mammal Bio- logical Laboratory. In press. Fur seal investigations, 1967. U.S. Fish Wildlife Serv. Spec. Sci. Rep. Fish. 597. (5) Mills. P. A., J. H. Onley, and R. A. Gailher. 1963. Rapid method for chlorinated pesticide residues in non- fatty foods. J. Ass. Otiic. Agr. Chem. 46(2):186-191. (9) Sladen, W. J. L., C. M. Menzie, and W. L. Reichel. 1966. DDT residues in Adelie penguins and a crabeater seal from Antarctica. Nature 210(5037):670-673. TABLE 1. — Operating conditions for analysis by gas chromatography Liquid phase 3% DC 200 5% QF-1 1:1 3% DC 200/ 5% QF-1 2% DEGS Solid support 60/80 Gas 60/80 Gas 80/100 Gas 80/90 Anakrom Chrom Q Chrom Q Chrom Q ABS Oven temperature 190 C 185 C 185 C 190 C Injector and detector temperature 210 C 210 C 210 C 210 C No flow rate 30 ml/minute 25 ml/minute 25 ml/minute 25 ml/minute Vol. 4, No. 3, December 1970 115 TABLE 2. — Pesticides in tissues and ingested milk of nursing northern fur seal pups, Pribilof Islands, Alaska, November 10, 1969 [nd = not detectable; T = trace] Tissue and Residues in ppm (Mo/kg, wet weight) i Pup Numbek DDE DDD DDT DiELDRIN PCS Muscle 1 8.1 0.33 0.34 0.038 T 2 0.58 0.060 0.068 nd T 3 0.19 0.015 0.022 nd T 4 1.0 0.051 0.084 nd T 5 0.069 nd 0.019 nd T Brain 1 0.34 nd 0.030 nd — 2 0.12 nd nd nd — 3 0,18 nd 0.013 nd — 4 0.058 nd nd nd — 5 0.012 nd nd nd — Liver 1 6.4 0.13 0.22 nd T 2 1.3 0.085 0.11 nd T 3 1.9 0.14 0.30 nd T 4 0.22 0.012 0.024 nd T 5 0.12 0.023 0.057 nd T Blubber 1 45 1.5 1.4 0.089 T 2 11 0.77 0.76 0.042 T 3 14 0.70 0.83 0.046 T 4 2.3 0.29 0.35 0.049 T 5 0.35 0.071 0.22 nd T Ingested milk 1 5.1 0.13 0.20 nd T 2 4.9 0.12 0.19 0.033 T 3 2.4 0.17 0.20 0.020 T 4 0.32 0.024 0.059 0.020 T 5 0.039 0.017 0.032 0.013 T Tissues only — residues in milk were calculated on a fat basis. TABLE 3. — Pesticides in liver ami brain (issues of fetal and nursing northern fur seal pups [nd — not detectable] Sample Size Residues in PPM (Mg/kg, wet weight) Tissue and Type of Pup DDE DDD DDT DiELDRIN Median Range Median Range Median Range Median Range Liver Fetus 1 7 nd nd-0.10 — nd — nd — nd Nursing 5 1.3 0.12-6.5 0.085 0.01-0.13 nd 0.02-0.22 - nd Brain Fetus ' 7 nd nd-0.04 — nd — nd — nd Nursing 5 0.12 0.01-0.34 — nd nd nd-0.01 — nd » Collected off Washington, February-March 1969 (2). 116 Pesticides Monitoring Journal International Cooperative Study of Organochlorine Pesticide Residues in Terrestrial and Aquatic Wildlife, 1967/1968 A. V. Holden' ABSTRACT A two-part collaborative study of organochlorine residues in wildlife was carried out by 17 laboratories in 11 countries. The first pari involved the analysis of four test samples con- taining organochlorine residues: one sample was a solution of standard chemicals, while the other three were cod-liver oil. chicken egg, and sprat (Clupea Sprattus) homogenate. This part provided a basis for comparison of the relative efficiency and accuracy of the laboratories in residue analysis. The second part of the program consisted of the sampling and analysis of four species of wildlife from areas believed to be free of pesticide usage in each country. The results indicate the degree of background contamination of the environment, as exemplified by the species selected, which were the starling, pike, marine mussel, and dogfish. The range of variation of residues among individuals of a natural population is much larger than that due to analytical errors or to differences between laboratories. Distributions within populations are non-Gaussian, necessitating large sam- ples for the detection of relatively small intcrpopulation differences. Introduction At a meeting sponsored by the Organization for Eco- nomic Cooperation and Development (O.E.C.D.) in Paris in June 1966, attended by representatives of 17 member countries, concern was expressed regarding the possible effects on the environment caused by the occurrence of pesticide residues in areas where there is no local usage of pesticides. To establish the presence of these residues in wildlife, a preliminary cooperative study was made voluntarily by laboratories in 1 1 member countries. The first O.E.C.D. Technical Conference on ' Freshwater Fisheries Laboratory, Pitlochry. Scotland (acting as coordinator). Pesticides in the Environment [sponsored jointly with The Natural Environment Research Council (London)] was held in .Scotland in September 1967 to discuss the results of this preliminary study. At this conference a decision was made to proceed with a further, more extensive study during the period 1967/1968. Reported here are the results of the extended study presented at the second O.E.C.D. Technical Conference on Pesticides in the Environment [sponsored jointly with T.N.O. (The Hague)], held in the Netherlands in September 1969. The main objective of the study was to establish the levels of pesticide contamination existing in selected species sampled at prearranged times in a prescribed manner in areas believed to be free of pesticide con- tamination arising from any local usage of organo- chlorine pesticides. The earlier study in 1966/1967 had demonstrated that certain types of pesticide residues were detectable in several species of wildlife in areas where no known usage of pesticides existed, but no attempt was made to compare the analytical abilities of the labora- tories involved. The second study therefore included a program of analysis of different types of samples ex- changed among the participating laboratories of several member countries of O.E.C.D., to establish a common basis for the comparison of their efficiency and accuracy with regard to both the qualitative and quantitative aspects of residue analysis. Each laboratory employed the techniques and apparatus which it normally used for pesticide residue analyses. A total of 1 7 laboratories took part in the analytical program, 16 in 10 member countries of O.E.C.D. and also the analytical laboratory of Euratom at the Euro- pean Community Commission Joint Research Center, Ispra, Italy. All had participated in the preliminary study in 1966/1967. The laboratories are identified in Table 1. Vol. 4, No. 3, December 1970 117 TABLE 1. — Laboratories participating in international co- operative study of organochlorine pesticide residues in terrestrial and aquatic wildlife, 1967/1968 COUNTKY Canada Euratom Finland Ireland Netherlands Norway Portugal Spain Sweden United Kingdom United States Participatino Laboratories Ontario Research Foundation, Sheridan Park, Ontario, Canada. Analytical Laboratory, Euratom, European Community Commission Joint Research Center, Ispra, Italy. State Veterinary Medical Institute, Box 10 368, Hel- sinki 10. Finland. The Agricultural Institute, Oak Park, Carlow, Ireland. Central Institute for Nutrition and Food Research (CIVO), Utrechtseweg 48, Zeist, Netherlands. National Institute of Public Health (RIV), Sterrenbos 1, Utrecht, Netherlands. Institute for Veterinary Pharmacology and Toxicology (Vet. Tox.), University of Utrecht, Biltstraat 172, Utrecht, Netherlands. Veterinary College of Norway, Department of Pharma- cology, Oslo 4, Norway. nacology, Quinta do Marques, Laboratory for Phytopha Oeiras, Portugal. Institute of General Organic Chemistry, Calle Juan de la Cierua 3, Madrid-6, Spain, Institute of Analytical Chemistry, University of Stock- holm, Roslagsvagen 90, Stockholm 50, Sweden. Laboratory of the Government Chemist (LGC), Corn- wall House, Stamford Street, London S.E. 1. Ministry of Agriculture, Fisheries and Food, Fisheries Laboratory (MAAF), Remembrance Avenue, Burn- ham-on-Crouch, Essex. The Nature Conservancy (NO, Monks Wood Experi- mental Station, Abbots Ripton, Huntingdonshire. Department of Agriculture and Fisheries for Scotland, Freshwater Fisheries Laboratory (FFL), Faskally, Pitlochry, Perthshire. U.S. Department of the Interior, Fish and Wildlife Service. Patuxent Wildlife Research Center, Laurel, Md. 20810, U.S.A. U.S. Department of Agriculture, Agricultural Research Service, BeltsviUe, Md. 20705 U.S.A. Interlaboratory Quality Evaluation Program for Pesticide Residue Analysis DESCRIPTION OF TEST SAMPLES Four types of test samples were circulated among the participating laboratories. These were analyzed by the methods in current use at the respective laboratories; details of the methods of analysis and confirmation are given in Table 2. As can be seen, the variations in technique were considerable. In addition, two samples containing polychlorinated biphenyls (PCB's) were provided for confirmation of the identity of unknown residues detected in some of the test samples. Sample No. 1 — a mixture of six organochlorine insecti- cides or derivatives, a S-ntl volume of hexane solution of the mixture being supplied in a glass ampoule. This sample was distributed by the Institute of Veterinary Pharmacology and Toxicology, University of Utrecht, Netherlands. On receipt of a report on the analysis of the sample from each laboratory, the Institute provided the correct analysis, as calculated from the amounts originally dissolved. This enabled analysts to correct any errors in standards or technique originating in their lab- oratoriRS before proceeding with the other tests and wildlife samples. Sample No. 2 — two ampoules, one containing a solution of the PCB formulation Clophen A. 50 in hexane at a stated concentration and the other a solution in hexane of five PCB isomers at stated concentrations, these hav- ing been isolated by preparative gas chrotnaiography and estimated by mass spectrometry. The solutions could be used for approximate quantita- tive estimation of some PCB residues in wildlife and were distributed by the Institute of Analytical Chemistry, University of Stockholm, Sweden. Clophen A. 50 (Bayer, Germany) is a commercial formulation comprising a series of PCB isomers and containing 50% chlorine by weight. It is similar to the Aroclors 1254, manufactured by Monsanto, which contains 54% chlorine by weight. Sample No. 3 — an ampoule containing cod-liver oil, dis- tributed by the Laboratory of the Government Chemist, London. Sample No. 4 — a sample containing 15 g of fresh chicken egg, uniformly ground with anhydrous sodium sulphate, supplied in an aluminum tube by the laboratory in Utrecht as for Sample No. 1. The eggs were produced in 1 week from 5 chickens which had been treated previously with a mixture of organochlorine insecticides for 1 month. The contents were homogenized and thoroughly mixed with anhydrous sodium sulphate for shipping. Sample No. 5 — a sample containing 15 g of a dried homogenate of 71 specimens of sprat (Clupea sprattus) caught in the Dutch Wadden Sea in June 1968. The homogenate was mixed thoroughly with anhydrous sodium sulphate for shipping in aluminum tubes and was supplied from Utrecht as for Sample No. 1. ANALYTICAL PROCEDURES The methods used varied widely, particularly in extrac- tion and cleanup. Some laboratories made a preliminary group separation of residues by column chromatography (CC) before analyzing by gas-liquid chromatography (GLC). Confirmation techniques included the prelimi- nary (CC) separation, two or more GLC columns, thin- layer chromatography (TLC), acidification to destroy dieldrin and endrin, and alkaline hydrolysis of TDE, DDT, and y-BHC, The techniques are summarized in Table 2. Extraction involved cold or hot solvents of various types, for short or long periods, in vertical columns using an adsorbent material, in Soxhlet extractors or 118 Pesticides Monitoring Journal sometimes in open dishes. Cleanup usually employed some form of liquid-liquid partition, commonly between hexane and acetonitrile or dimethylformamide, followed by passage through an adsorbent column. In a few cases the partition stage was omitted and fat removed on the column. TTie most common adsorbent was Florisil, and in many cases successive elutions with solvents of different polarity provided a pre-GLC sep- aration. Other adsorbents were alumina and silica. Some analysts obtained pre-GLC separations by thin- layer plates or by treatment with acid or alcoholic KOH. The GLC analysis generally employed two different types of column, providing further confirmation. DISCUSSION OF RESULTS Although distribution of the samples began in April 1968, many laboratories found difficulty in completing the analyses of the test samples by the end of 1968. This was partly due to their inability to give priority to the samples and partly to difficulties presented by sample No. 5. Some laboratories experienced difficulty with certain samples to which they were not accus- tomed and which required additional cleanup or separa- tion techniques. Sample No. 1 was the only one analyzed by all participants. The results of the analyses of test samples No. 1, 3, 4, and 5 are given in Tables 3-6. Where analyses by two methods were reported, the mean value is listed. For the mixture of pure pesticides fNo. 1) the analyses are given to one decimal place, but for the other test samples, which contained lower concentrations, the data are usually given to two decimal places. The manner of reporting inevitably differed widely, with some laboratories stating only the concentrations of residues found in appreciable amounts and others men- tioning levels of residues not fully confirmed, while some analysts reported pesticides to be absent at the level of detection obtainable by their current methods of analysis. The values reported were generally uncor- rected for losses incurred in extraction. As might be expected, agreement between the labora- tories on the analysis of the mixture of pure pesticides was reasonably good. In a few instances, analysts were encountering residues not normally found in their rou- tine work, e.g., heptachlor epoxide and endrin, and this necessitated preliminary investigation of the appro- priate GLC separation and the possibility of breakdown of endrin. The other exchange samples showed a wider variation between laboratories, both qualitatively and quantitatively. These samples required full processing (extraction, cleanup, and confirmation) which intro- duced varying degrees of error. Sample No. 1 Some analysts reported that they were doubtful of the purity of their own standards, in particular heptachlor epoxide and DDT. Assuming that the distributing lab- oratory employed only high-purity chemicals, others using standards of lower purity for calibration would be likely to report excessively high concentrations; but errors in dilution and errors due to GLC injection variations would be random. The results (Table 3) show that distributions of the values for all six pesticides were approximately normal, suggesting that any errors due to impurity of standards must be small. Only the values obtained by direct deter- mination have been used in the statistical evaluation of the data, presented in Table 7. This table gives the true value for each pesticide, the mean calculated from the analyses, standard deviation, coefficient of variation, and the number of laboratories which reported values within ± 5% of the mean. With the exception of endrin, the means of the concen- trations reported were very close to those stated to have been present by the distributing laboratory. No explana- tion has been found for the anomalous value of the mean for endrin, but the distributing laboratory, using the same standard for analysis, obtained a result similar to those reported by many other laboratories. Two lab- oratories, Ireland and Netherlands RIV, reported ex- ceptionally high values for endrin. Omission of these values produces a distribution which is closer to normal, with a coeflficient of variation of ± 11.9%. The mean value of the coefficient of variation for the six residues is then approximately it 9%. By comparison, the co- efficient of variation for replicate GLC injections by one analyst is. at best, usually of the order of ± 2% and often as high as ±: 5%. (A 2% error is equivalent to a one-division error for a GLC peak giving 50% of full-scale deflection on a 100-division recorder scale.) Examining the data in more detail, none of the 17 laboratories reported values consistently within ±5% of the means, although 5 reported only one value out- side these limits. The deviations from the means for individual laboratories were not random; however, Euratom reported five low values, Norway four low values, and Spain five high values (all in excess of 5% error). It is possible that the particular methods of diluting the original sample for analysis which were in use in the individual laboratories may have produced the biased results, and the methods of calculation may also have differed. For example, some analysts use a calibration graph derived from a series of concentra- tions of a standard, but others calculate from only one such concentration. The latter method is liable to give a significant error unless the standard and sample con- centrations are similar. Vol. 4, No. 3, December 1970 119 Sample No. 2 (PCB) The two samples provided were examined by most lab- oratories, but only three gave details of the results obtained. In the first sample 5 major peaks and a number of small peaks were detected, and in the second at least 14 peaks could be found. Most analysts used these peaks as references (by relative retention times) for the unknown peaks found in test sample Nos. 3 and 5. Sample No. 3 (cod-liver oil) This sample presented some difficulty in cleanup, but the proportion of PCB residues was less than the pesti- cide residues. Interference between these two groups was recognized by most analysts, and a variety of methods was used to overcome this difficulty. The 14 laboratories analyzing this sample reported 12 pesticide residues apart from PCB's. The frequency of reporting was as follows: Dieldrin 12 o,p' - DDT 6 p.p' -DDE 14 a-BHC 6 p.p'-TDE 13 y-BHC Endrin 2 Aldrin 1 4 Heptachlor epo.xide 2 p./ -DDT 13 ^-BHC 1 o,p'-TDE 1 PCB's 9 positive or suspected The laboratories were in general agreement on the presence of four of these compounds, dieldrin and the three members of the p.p' - DDT group. The distribu- tions of the data for these residues were approximately normal, and statistical analysis gives the following information: Mean Range SD CV Dieldrin 0.139 < 0.05 - 0.37 ±0.114 ±82% ^y-DDE 0.357 0.16-0.58 ±0.120 ±34% p,p' -TDE 0.541 0.13-0.80 ±0.205 ±38% p.p'-DDT 0.628 0.17-1.44 ±0.338 ±54% Mean, range, and standard deviation in ppm The standard deviation for a single determination is thus ± 0.1 to ± 0.3 ppm. the percentage error being higher for dieldrin and DDT than for DDE and TDE. The range of values reported is considered to be exces- sively large, but this seems to be only partly due to the problems caused by the presence of PCB's. Sample No. 4 (chicken egg) Analysts found this the easiest of the three "wildlife" samples to analyze, with no PCB interference and rela- tively high pesticide concentrations. Little cleanup was found to be necessary — some analysts using an unproc- essed extract in solvent for direct injection on a GLC column. The 16 laboratories analyzing this sample reported 10 residues. The number of laboratories reporting each residue was as follows: Dieldrin 16 p,p'-DDE 16 Endrin 14 Aldrin 1 BHC 5 o.p' - DDT 2 p.p'-TDE 12 HCB 5 p.p'-DDT 16 ^-BHC 1 The analysts were in general agreement on the pres- ence of dieldrin, p.p'-DDE. p.p'-DDT, and endrin, but a few did not report p.p'-TDE, which was in a smaller concentration. One laboratory probably incorrectly identified endrin as o.p'-DDT since endrin interferes with o.p'-DDT on some types of GLC columns. The y-BHC and hexachlorobenzene (HCB) peaks are prob- ably identical and were confirmed by some analysts as HCB, e.g., by separation on silica. Peaks at this early stage of a chromatogram are not always attempted by analysts. The data for the five major residues were approximately normally distributed, with statistical analysis giving the following results (values in terms of homogenate were not included) : Mean Range SD CV Dieldrin 0.162 0.08-0.24 ±0.046 ±28% Endrin 0.136 0.07-0.36 ±0.073 ±54% p.p'-DDE 0.127 0.08-0.23 ±0.040 ±32% p.p'-TDE 0.046 0.02-0.10 ±0.023 ±51% p.p'-DDT 0.445 0.14-0.62 ±0.140 ±31% Mean, range, and standard deviation in ppm One laboratory reported an exceptionally high value for endrin which, if omitted reduces the coefficient of varia- tion to ± 25%. and the standard deviation to ± 0.029 ppm. The agreement between analysts is thus appreciably better than that for the cod-liver oil sample, the coeffi- cient of variation being in the range of 25-30% with the exception of p.p'-TDE, which was in a significantly lower concentration. This measure of agreement, irre- spective of the type of extraction or other processing, is probably due to the absence of PCB residues and prob- lems of cleanup. Sample No. 5 (sprat) This sample proved to be the most difficult to analyze, due partly to the low levels of pesticide residues present 120 Pesticides Monitoring Journal in relation to PCB's and partly to difficulties experienced in cleanup. Fourteen laboratories analyzed this sample and reported 12 residues, in addition to PCB's, as follows: Dieldrin 14 p.p' - DDT 12 Heptachlor Endrin 13 y-BHC 8 epoxide 4 p,p' - DDE 12 a-BHC 3 Aldrin 2 p,p' - TDE 12 ;8-BHC 2 HCB o,p' - DDT 2 1 PCB's 1 3 positive Once again the laboratories were in general agreement on the presence of dieldrin, endrin, and the three mem- bers of the p,p'-DDT group, although interference by PCB peaks with p.p'-TDE and p.p'-DDT made their quantitative estimation difficult. As with test sample No. 4, y-BHC and HCB are probably the same residue, but few analysts are familiar with HCB. The data for the five major residues showed less agree- ment between analysts than for test samples No. 3 and 4; but, assuming norma! distribution of the values, statis- tical analysis gives the following results (values in terms of homogenate were not included) : Mean Dieldrin 0.122 Endrin 0.132 p,p' - DDE 0.068 p.p'-TDE 0.118 p.p'-DDT 0.113 Range SD CV 0.02 - 0.22 ± 0.060 ± 50% 0.09-0.21 ±0.039 ±29% 0.01-0.17 ±0.052 ±76% 0.04 - 0.28 ± 0.064 ± 54% 0.02-0.25 ±0.083 ±73% Mean, range, and standard deviation in ppm Two exceptionally high values were reported for DDE; the omission of these reduces the coefficient of variation to ± 56%. One very high TDE value was reported, and omitting this reduces the coefficient of variation to ± 37%. The interference of a PCB peak with DDT is the most likely reason for the high variation among the values for DDT, and with the omission of this residue the coefficient of variation is of the order of ±25% to ± 50%, slightly greater than the values for test sample No. 4. A study of the data for samples No. 3, 4, and 5 shows no evidence that any particular technique for extraction, cleanup, or pre-GLC separation consistently gave values higher or lower than average for a particular pesticide. The greatest degree of agreement was reached with test sample No. 4 (chicken egg) which required little or no cleanup and was free from PCB interference. This group of compounds, commonly found in marine samples, interfered in test samples No. 3 (cod-liver oil) and 5 (sprat), and most separation techniques in common use do not separate PCB's from the members of the p,p'-DDT group. Where such a separation is effective. as with a silica column, TDE and DDT values can be estimated more accurately. CONCLUSIONS The results of the analyses of the standard mixture of pesticides, which required only appropriate dilution be- fore GLC determination, show that at this stage of the procedure laboratories agreed on the identity of the common pesticide residues and on the amounts present with a coefficient of variation of 7-10%. In the analysis of wildlife samples, which required extraction, cleanup, and sometimes pre-GLC separation before GLC determination, the coefficient of variation was considerably greater, ± 30% to ± 60% for the more difficult samples, and no better than ± 25% to ± 30% for the least difficult. In the wildlife samples, agreement on the identity of residues was good only for the major residues found, which were dieldrin, p.p'-DDE, p.p'-TDE, and p.p'- DDT. These are normally the compounds which are of the most concern to ecologists. and the accuracy of the analyses is generally sufficient for most ecological studies. However, the detection of year-by-year trends in con- tamination levels or of small differences between popu- lations in different areas would require a greater ac- curacy. This could probably be achieved by greater efficiency of recovery in cleanup techniques, together with a method of separation of the interfering PCB's from the true pesticide residues where necessary. Organochlorine Residues in Wildlife Samples SAMPLING PROCEDURES This study involved the sampling and analysis of one or more of the following species — starling {Sturnus vul- garis), pike {Esox Indus), mussel {Myiihis edulis), and dogfish {Squalus acanthias) — or appropriate ecological equivalents. The species were to be sampled as far as possible at stated periods, from areas where there was no history of local pesticide usage. The preferred speci- men sizes were indicated in some instances as well as the methods of preparing samples for analysis. In some cases analyses of individuals or groups of individuals were planned to permit statistical evaluation of the re- sults. The detailed procedures for sample collection and prep- aration were outlined as follows: Starling, Sturnus vulgaris — adults of the species to be sampled on arrival at breeding areas. Where possible, 30 or 35 specimens to be taken, numbered individually, and the weight of each bird recorded. If such a large number of specimens is not available, or if analytical time is limited, 10 specimens should be numbered, and weighed individually. Remove 10 g of breast muscle from each bird for analysis. Vol. 4, No. 3, December 1970 121 For 10 specimens, analyze each bird individually, 2 in duplicate but not consecutively. For 30 or 35 specimens, analyze 10 individually as above, and analyze 20 (or 25) in 5 groups of 4 (or 5), mixing the 10-g aliquots in each group uniformly before analysis. Mussel, Mytilus eduUs — Specimens to be sampled in the estuaries of, or at the mouths of, large rivers during November or December 1967, or as soon as possible afterwards, but before the spawning period (spring 1968). A total of 10 large or 25 small specimens to be collected, numbered individually, and the overall shell length of each recorded. The soft tissue to be used for analysis. Where possible, 1 0 specimens should be analyzed indi- vidually, 2 analyses in duplicate but not consecutively. If specimens are too small for individual analyses, tis- sues should be mixed in five groups of five specimens and analyzed, two groups in duplicate but not consecutively. Pike, Esox Indus — Ten adults, 4 years of age or older, sampled in April or before spawning; length, weight, sex, and age recorded. A 100-g portion of lateral muscle to be taken at mid- section for analysis, each fish being analyzed separately. Samples from two fish to be analyzed in duplicate but not consecutively. Dogfish. Sqiialus acanthias — Large specimens to be taken as available; length, weight, sex, and age recorded if pos- sible. Remove lateral muscle and liver from each speci- men for analysis. (TTiis species was chosen because of its wide distribution in the marine environment. Its examination was in the nature of a reconnaissance, but it was hoped that those countries able to obtain and analyze specimens easily would do so.) Data from 10 individual specimens represent the mini- mum requirement for statistical analysis of the popula- tions sampled. Although not required, it was hoped that some of the participating laboratories would also analyze composited samples for comparison of data with results obtained from analysis of individual specimens. ANALYTICAL PROCEDURES The methods of analysis employed for the wildlife samples were usually the same as those for the exchange samples, but the following major differences were noted by participating laboratories. Canada Netherlands Method (a) used for pike, method (b) for starlings and mussels. For starlings. Vet. Tox. Laboratory used method of Onley and Bertuzzi( / ) , acetone/ methylcellosol ve extraction, DMF partition, transfer to 40-60 C petroleum ether, Florisil column, GLC analysis on 5% DC-11 and 5% QF-1 columns on Chromosorb W at 185 C. Spain For dogfish, Onley and Bertuzzi(/) cleanup method used. United Kingdom Starlings analyzed by NC laboratory. Mussels analyzed by MAFF, DAFS laboratories. Pike and dogfish analyzed by DAFS laboratory. United States Analyses at Laurel laboratory only. GLC analysis employed 10% QF-1 column at 160 C in place of 12% DEGS column. RESULTS Ten countries sampled and analyzed the starling and pike, and all sampled mussels. Euratom, Ireland, Norway, Spain. Sweden, and the United Kingdom examined dog- fish. There was difficulty, in some countries, in obtaining the full number of specimens required for analysis, par- ticularly of the starling, and the specified program for analysis of starlings (single specimens and groups of 4 or 5) was not always completed. Summaries of results of analyses for the four wildlife species are given in Tables 8 to 11. These indicate the sizes of the specimens analyzed; the mean values and ranges of values obtained for the four major pesticide residues — dieldrin. DDE, TDE, and DDT; the presence or otherwise of PCB residues; and any necessary qualifi- cations. Where residues were not detected, the limit of detection was stated. Unfortunately, analytical proced- ures at some laboratories did not permit determination of all four residues. All sampling sites were considered to be free of local pesticide usage. However, the movement of starling flocks may result in some ingestion of pesticides from other areas. Further, it is impossible to ensure freedom from pollution in estuaries where mussels are obtainable; and the catchment areas of lakes from which pike were sampled may also contain unknown sources of pesticides. Although the ranges of values for each pesticide found in individuals of a population vary considerably, the mean values can give some indication of the degree of contamination of the population. Certain conclusions may be drawn from the data in the tables, and these are discussed below. For the purpose of comparison, levels between 0.001 and 0.01 ppm could be considered as the minimum degree of contamination now found in the environment, and levels above 0.1 ppm could be regarded as evidence of local pollution. 122 Pesticides Monitoring Journal DIELDRIN Mean levels less than 0.01 ppm in starlings were re- corded only by Euratom. Finland, Portugal, Spain, and Sweden, and some U.S. and Portuguese samples had mean levels over 0.1 ppm. In the pike, all mean values were below 0.01 ppm e.xcept in Ireland, but in the dog- fish the values reported were above 0.01 ppm. For mussels almost half of the samples had less than 0.01 ppm, but in the Netherlands, Ireland, United Kingdom, Portugal, and the United States sample values up to 0.06 ppm were obtained. This insecticide is nevertheless a minor contaminant in most countries. DDE In starlings this residue was always in excess of 0.01 ppm, but much higher values were found in Canada, the United States, Italy (Euratom), the Netherlands, Portugal, and Sweden. In pike several countries reported levels below O.OI ppm: Euratom and Spain both found a high level of contamination. In mussels. Scandinavia reported values below O.OI ppm, and only in the United States did levels exceed 0.1 ppm. Of the six countries examining dogfish, Euratom and Spain reported values (in muscle) above 0.1 ppm and Sweden, Norway, Ireland, and the United Kingdom above 0.01 ppm. Also, the laboratories (Eura- tom. Sweden, and the United Kingdom) that examined liver reported values above 0.1 ppm. TDE This residue was not always reported, and no high values were found in the mussel or starling. Most countries reported values above O.OI ppm in the mussel, and in the dogfish all values reported exceeded 0.01 ppm with values exceeding 0.1 ppm in Italy (Euratom) and Spain. For pike. Spain and Euratom reported some values above 0.1 ppm, but most found levels below 0.01 ppm. DDT In starlings, levels were above 0.1 ppm in the Nether- lands and Spain and above O.OI ppm in most other countries. In pike, only Euratom reported values above 0.1 ppm. In dogfish, all six laboratories found levels exceeding 0.1 ppm. In mussels, residues in the range of 0.01-0.1 ppm were found in all countries, with the ex- ception of a low level (0.0036 ppm) in Norway and a high level (0.184 ppm) in Portugal. GENERAL The impression is gained that the Mediterranean area may contain a higher level of DDT group contamination than elsewhere in marine species, but the North Sea also has a significant level of pollution. Both Sweden and the Netherlands reported high DDT group levels in starlings, but pike generally showed only slight contamination from this chemical group. PCB These residues were not always reported. From the limited data reported, however, residues in starlings appear to be very small or below detection, and residues in pike were also at low levels. PCB's were present generally in mussels and in dogfish, with levels in mus- sels being somewhat higher than in other samples. It is not possible with current analytical techniques to identify individual PCB's, but those occurring in wildlife mainly elute on GLC columns after dieldrin. The PCB residues in this study were quantified, where mentioned, by com- parison with Clophen A. 50 (test sample No. 2). STATISTICAL ANALYSIS OF DATA A satisfactory statistical analysis of the data obtained from the wildlife samples was only possible where (a) the values measured were large enough to be useful, and (b) there were sufficient individuals in a sample to provide an indication of the appropriate type of distri- bution. These requirements limited the statistical analyses to the data for DDE concentrations in starlings and pike from most countries, although in some instances the data for other residues were also examined. The majority of the population distributions were found to be of a non-Gaussian type, but a lognormal distribu- tion was considered to be appropriate. Tables 12 and 13 give the results obtained from the DDE analyses of the various starling and pike samples. The variance in each case is in the logarithmic fonn, but the standard devia- tion is the antilog of the square root of the variance. The 95% confidence limits are calculated from the variance (log form) by deriving the antilog of twice the square root of the variance (log form). The geomet- ric mean is then divided and multiplied by this value, giving the lower and upper limits, respectively. The precision of the mean is calculated in a similar manner, from the variance (log form) divided by the number of observations, i.e., the variance of the mean. The antilog of twice the square root of this value (log form) is used to obtain the 95'~'c limits of the mean, by division and multiplication. The variance for the starling populations appears to be dependent on the level of contamination existing, and the standard deviations ranged from 1.4 to 3.2. These indicate that a wide range of values can be expected for the DDE contents of individuals in a population of starlings, the upper limit being as much as one hundred times the lower limit for 95% of the population. For the purpose of assessing trends in contamination levels or for comparing the degrees of contamination in popula- tions from two areas, the sample size will depend upon the percentage difference which it is required to detect, if it exists. The tables give the approximate number of specimens required to determine differences of 25%- 200% between two means, based on the analyses for both those countries with the higher variance and those with the lower variance. Vol. 4, No. 3, December 1970 123 The geometric mean values for DDE in the starlings fall into two groups, those from Canada, the United States, and the Netherlands giving an overall mean of 0.230 ppm with a variance (log form) of 0.22, and all the other values giving an overall mean of 0.072 ppm with a variance (log form) of 0.060. The range of values to be found among 95% of the population would in the first case be from 0.026 to 2.0 ppm and in the second case from 0.023 to 0.23 ppm. The DDE values in the pike were all low, with the sole exception of those from Euratom. Excluding the latter, the overall geometric mean is 0.0096 ppm with a variance (log form) of 0.052. The 95% limits in this case are from 0.0033 to 0.028 ppm. The natural range of values of a pesticide residue among individuals of a population is clearly large, although in theory this might be reduced if the individuals could be selected for age, sex, size, etc. Such a selection is nor- mally impractical, and the large error inherent in any estimate of the overall degree of contamination must be accepted. A comparison with the coefficients of varia- tion determined from the various exchange samples an- alyzed in the first part of this study, however, suggests that the errors due to analysis, to variations between techniques or laboratories, would be relatively unimpor- tant in any comparison of population means, either with- in or between different countries. The estimates of the numbers of individuals to be sampled in any population in order to detect reasonably large differences between populations are high. For ex- ample, the analysis of 50 individuals from each of 2 populations to enable detection of a 25% difference in the geometric means would entail a considerable an- alytical effort. It may be possible to reduce this effort to some extent by analyzing several homogenates of small numbers of individuals, such as 10 groups each of 5 individuals, but this requires a more detailed examina- tion of the population structure than is possible from the present data. TTie major obstacle to any further increase in the number of individuals sampled would, however, in most cases, be that such sampling would be unacceptable in the case of many wildlife species, especially those considered to be at risk. Any inter- national program of monitoring the environment for pesticide or other contamination would require con- sideration of the following: (a) the sample size required to detect a percentage change in the contamination level which may be biologically significant (b) whether a species can be sampled to this degree without suffering damage in the process (c) whether the analytical facilities are available to handle the large numbers of individuals involved. CONCLUSIONS The second O.E.C.D. study program has shown that the ability of analysts in different countries to detect and estimate the major pesticide contaminants of the en- vironment is reasonably good and sufficiently adequate for the purpose of monitoring wildlife populations. A further improvement can be expected as techniques be- come more uniform and as methods of separating in- terfering PCB compounds from organochlorine pesticide residues are more widely used. The estimates of the coefficient of variation to be ex- pected in the analysis of different types of samples for dieldrin and the DDT group of residues are: One analyst, replicate analyses, standard solution ± 2-3% One analyst, replicate analyses, wildlife sample ± 15-30% Between analysts, standard solution ± 7-10% Between analysts, wildlife samples requiring minimum processing ± 25-30% Between analysts, wildlife samples requiring maximum processing ± 30-60% The distribution between residues in individuals of a wildlife population is of a lognormal rather than a normal form, and the number of individual analyses re- quired is consequently large if a reasonably accurate estimate of the mean level of contamination is to be obtained. For the detection of small changes in the con- tamination level, as in an international monitoring pro- gram, the analytical effort required would be consider- able, and the size of the population sample could be unacceptably high or impracticable for many species. The detection of annual changes of the order of 50% would, however, be feasible. Current levels of the residues of dieldrin, DDE. TDE, and DDT in the wildlife species examined in areas free from the direct use of organochlorine pesticides, are generally in the range of 0.0 1 -0.1 ppm. although in a few countries levels above 0.1 ppm were found. Pike were the least contaminated, and starlings usually the most, presumably a reflection of the relative freedom of movement of the species in the environment. See Appendix for chemical names of compounds mentioned in this paper. A cknowledgments The author gratefully acknowledges the assistance of Miss M.B. van Lennep of the Central Organization for Applied Scientific Research (T.N.O.), the Hague, Netherlands, in assembling the data for the preparation of this report. The results of this study were originally presented for discussion at the Second O.E.C.D. Tech- 124 Pesticides Monitoring Journal nical Conference at Helvoirt, Netherlands, in Septem- ber 1969, and have been revised for presentation in this paper. The statistical analyses were made at the Marine Lab- oratory, Aberdeen, Scotland, by Mr. W. B. Hall, to whom the author is indebted for the information in that section of the paper. LITERATURE CITED (■/) Onley. J. H. and P. F. Bertuzzi. 1966. Rapid extraction procedure for chlorinated pesticide residues in raw ani- mal tissues and fat and meat products. J. Ass. OfBc. Anal. Chem. 49(2):370-374. TABLE 2. — Analytical techniques Methods of Extraction Cleanup Pre-GLC Separation G LC Confir- mation Cold Hot Time Parti- tion Column Other Packing Length Temp. (C) s 0 U .J H S Canada b) A/N- hexane a) ether- hexane 2 hrs. Florisil coldbath 6% and 20% ether in PE 6% QF-1 4% SE-30 on Chromosorb W 6 feet 200 -f- -1- Euratom a) A/N in column A/N-PE Florisil 6% and 15% ether in PE 1) 5% DOW-11 7.5% QF-1 on Gas Chrom Q 10 feet 175 -1- + b) A/N A/N-PE Florisil 6% and 15% ether in PE 2) 5% DOW-11 on Gas Chrom Q 5 feet 175 -1- + c)PE Florisil 6% and 15% ether in PE Finland ether in column a) H2SO1 b)TLC TLC 1) 4% SF-96 on Chromosorb W 2) 8% QF-1 on Chromosorb W 5 feet 176 180 Ireland H-acetone 2:1 H/DMF Alumina 1)5% DC-200on Aeropak 30 2) 5% QF-1 on Chromosorb W 5 feet 185 185 -1- Netherlands CIVO hexane H/DMF FlorisU 6% and 15% ether in PE 1)5% DC-200on Gas Chrom Q 2) 2% DC-200 3% QF-1 on Gas Chrom Q 6 feet 6 feet 200 200 4- -1- RIV A/N on FlorisU A/N to PE 6% and 15% ether in PE 5% DC-200 on Aeropak 30 5 feet 200 Vet. Tox. PE 6hrs. H/DMF Florisil a) hexane b) ether in hexane 5% DC-200 7% QF-1 on Gas Chrom Q 175 -t- Norway a) PE on column b) ether on column PE/ DMF H:SOi alcoholic KOH 1) 10% QF-1 on Chromosorb W 2) 4% SF-96 on Gas Chrom P 6 feet 5 feet 175-180 170-175 Portugal hexane 6 hrs. H/DMF Alumina a) first 50 ml b) next 40 ml 1) 10% DC-200 on Gas Chrom Q 2) 5% QF-1 on Gas Chrom Q 3 feet 3 feet 200 185 -t- Spain hexane A/N Florisil on alum- ina 5% DC-200 7.5% QF-1 on Chromosorb G 5 feet 190 + Sweden PE 3'/2 hrs. a) Silica b) 25% KOH c) AgC104 in fuming H=SO. 1) 1.75% SF-96 3.75% QF-1 on Gas Chrom P 2) SF-96 5 feet 5 feet 180-190 180-190 + + Vol. 4, No. 3, December 1970 125 TABLE 2. — Analytical techniques — Continued Methods of Extraction CLEAhfUP Pre-GLC Separation GLC Confir- mation Cold Hot Time Parti- tion Column Other Packing Length Temp. (C) i u United Kingdom LGC b) DMSO on column a) 2:1 hexane- acetone 3 hrs. DMF Alumina siUca l)SE-52 2) Apiezon L 3) XE-60 -f NC hexane and acetone alter- nately hexane DMF Alumina 1) Apiezon L on Gas Chrom Z 2) SE-30 on Gas Chrom Z 2V4 feet 188 + + MAFF hexane hexane DMF Alumina 1) 2% Oronite 128 0.2% Epikote 1001 on Gas Chrom Q 2) 2.5% SE-30 0.25% Epikote 1001 on Chromosorb G 4 feet 4 feet 168 168 FFL hexane 30 min. Alumina Silica 1) 10% DC-200 on Chromo- sorb W 2) 5% DC-200 7.5% QF-1 on Chromosorb W 5 feet 5 feet 200 200 U.S.A. Laurel PE 8 hrs. A/N hex- Florisil TLC 1)3% OV-17on Gas Chrom Q 2) 3% XE-60 on Gas Chrom Q 3) 12% DECS on Gas Chrom Q 6 feet 6 feet 6 feet 90 170 20O -1- Beltsville a)A/N hexane Florisil 1)5% DC-200 on Aeropak 30 6 feet 190 -1- b) chloro- form methanol 16 hrs. hexane and A/N TLC 2) XE-60 on Gas Chrom Q 5 feet 175 + Note: A/N = acetonitrile PE = petroleum ether H =: hexane Ether = diethyl ether DMSO = dimethylsulfoxide DMF = dimethylformamide TABLE 3. — Analyses of test sample no. 1 — standards CONCENTIUTIONS IN MO/LITER Laboratory Heptachlor Epoxide DiELDRIN Endrin P.P'-DDE P,p'-TDE p,p'-DDT Canada 5.0 5.2 5.0 4.7 9.9 10.2 Euratom 4.3 5.2 4.9 4.0 9.1 9.3 Finland '3.2 5.4 6.8 5.1 10.5 10.8 Ireland 4.8 5.9 7,9 M.O 10.2 10.4 Netherlands crvo 4.7 5.2 5.6 5.1 11.5 9.8 RTV 6.0 5.0 8.2 5.5 9.1 9.7 Vet. Tox. 4.4 5.3 5.2 5.0 10.0 9.3 Norway = NC 5.1 5.0 4.1 8.7 9.0 Portugal 3 5.3 35.1 35.7 35.2 3 10.5 8 9.1 Spain 5.8 5.5 6.7 6.0 10.7 11.8 126 PESTicroES Monitoring Journal TABLE 3. — Analyses of rest sample no. 1 — standards — Continued Concentrations in mg/liter Laboratory Heptachlor Epoxide DlELDRIN Endrin p,p'-DDE p,p'-TDE p,p'-DDT Sweden 3.8 5.5 6.6 5.1 10.4 10.2 United Kingdom LGC 4.7 5.2 4.9 5.1 10.1 9.9 MAFF 5.0 4.2 5.0 4.7 10.6 10.0 NC 4.9 5.5 5.5 4.7 9.7 9.3 FFL 4.9 4.8 5.9 4.5 10.2 10.0 United States Laurel 5.1 5.3 6.0 5.1 9.0 9.5 Beltsville 5.0 5.9 5.9 5.0 9.2 10.0 TRUE VALUES 4.95 5.24 7.05 4.87 10.04 9.95 Calculated indirectly, i NC = present, but not calculated. ' Means of 2 methods. TABLE 4. — Analyses of test sample No. 3 — cod-liver oil Concentrations in PPM Laboratory U X to 0 U X m u X a o I z Z < o X z Q UJ ED Q 9 H Q Q 0 0 o < ^ 03 Canada 0.075 0.10 0.087 0.32 0.41 0.50 0.055 (?) Euratom 0.58 0.72 0.79 (?) Finland 0.12 >0.41 0.62 0.48 0.11 1.39 Ireland 0.14 0.05 0.02 0.37 0.51 0.68 1.44 (?) NC Netherlands CIVO RIV Vet. Tox. <0.05 0.31 T Norway = 0.17 = 0.33 2 0.58 = 0.50 Portugal 0.07 0.11 0.016 0.50 0.66 0.63 Spain T 0.10 0.30 0.20 0.80 0.70 Sweden 0.09 0.40 0.49 0.55 0.085 •0.24 NC United Kingdom LGC 0.03 0.01 0.02 0.08 0.35 0.50 0.65 0.20 NC MAFF 0.056 0.16 0.33 ?0.20 0.37 0.80 1.10 NC FFL 0.16 0.35 0.39 0.41 NC U.S.A. Laurel <0.005 <0.05 0.16 0.13 0.17 0.04 NC BeltsvUle 0.18 0.21 0.25 0.25 0.22 1 + PCB interference. 2 Means of two methods. NOTE: ? = Suspected T = Trace NC = Present, but not calculated Vol. 4, No. 3, December 1970 127 TABLE 5.— Analyses of test sample no. 4 — chicken egg Concentrations IN PPM Laboratory U X 03 a y s u X O o X X z < X z Q z z Q Q D H Q a W Q H o X 2 UJ Z X S a u a. Canada 0.18 0.11 0.12 0.035 0.51 0.11 Euratom 0.20 0.16 0.11 0.02 0.51 Finland 0.21 0.15 0.14 0.028 0.59 Ireland 0.025 0.11 (?) 0.09 0.04 0.17 Netherlands CIVO 0.16 0.14 0.13 0.47 0.039 RIV 10.08 0.23 0.36 0.16 0.07 0.62 T Vet. Tox. 0.17 0.11 0.09 0.04 0.39 0.14 Norway 0.135 0.15 0.088 0.028 0.42 Portugal 0.03 0.05 0.18 0.11 0.17 0.03 0.39 Spain 0.04 0.14 0.13 0.10 0.05 0.50 0.21 Sweden = 0.044 = 0.039 = 0.034 = 0.004 -' 0.10 United Kingdom LGC 3 0.145 3 0.09 3 0.105 ■'■ 0.36 ' 0.035 MAFF 0.009 0.15 0.11 0.23 0.10 0.53 NC FFL 0.24 0.13 0.16 0.06 0.58 0.09 U.S.A. Laurel 0.08 0.07 0.08 0.49 Beltsville 0.1! 0.08 0.12 0.14 1 -f HCB. - Homogenale only. 3 Means of two methods. NOTE: ? = Suspected T = Trace TABLE 6 . — Ana 'yses 0 / test sample no. 5 — sprat Concentrations in PPM Laboratory X CQ u I ffi u I a o i X < X 0 i z Q z z in a Q Q H s 9 D Q H 6 o < ^j X S o Canada 0.016 0.19 0.13 0.083 0.10 0.087 NC Euratom 0.18 0.17 0.09 0.05 0.30 approx. Finland 0.005 0.12 0.13 NC 0.067 0.061 1.70 Ireland 0.02 0,02 0.04 0.02 0.04 0.21 0.15 0.14 0.20 NC Netherlands CIVO 0.10 0.09 0.07 0.80 RIV Vet. Tox. 0.11 0.10 0.025 0.04 0.57 Norway > 0.075 ' 0.12 1 0.068 1 0.093 ■ 0.050 NC Portugal 128 Pesticides Monitoring Journal TABLE 6. — Analyses of test sample no. 5 — sprat — Continued ' Means of two values. - Homogenate only. ^ Identity uncertain. NOTE: T = Trace; NC : Concentration IN PPM Laboratory U X m O X m U X o o X i X z < X z o 5 z z a a Q Q Q Q Q 1^ 0. 6 £ z < [^ n u a. Spain 0.098 0.022 0.046 0.024 T 0.025 0.043 0.11 T 0.65 Sweden 0.03 0.10 0.06 0.15 0.15 2.0 United Kingdom LGC ■-' 0.02 = 0.015 - 0.005 ■- 0.005 = 0.09 = 0.09 = 0.03 = 0.04 = 0.02 NC MAFF 0.014 0.1.1 0.12 0.17 0.28 0.25 NC FFL 0.22 0.18 0.072 »0.16 •■' 0.23 0.10 NC U.S.A. Laurel 0.01 0.005 0.20 0.10 <0.04 0.08 <0.07 >0.50 Beltsville 0.01 0.01 0.10 0.10 0.01 0.10 0.02 0.80 approx. . but not culculatcd. TABLE 7. — Sumnwry of analyses of test sample no. I Concentrations in mc/liter Residue No, OF True Proportion Determi- Concen- Means of Standard Coefficient WITHIN nations tration Analyses Deviation of Variation ±5% OF Mean Heptachlor epcxide 15 4.95 4.913 ±0.54 ill.O'Tf, 9 out of 15 Dieldrin 17 5.24 5.253 ±0.39 ± 7.4% 13 out of 17 Endrin 17 7.05 1 5.929 ±1.01 ±17.1% 5 out of 17 p.p'-DDE 16 4.87 4.931 ±0.49 ± 9.9% 11 out Of 16 p.P-TDE 17 10.04 9.965 ±0.75 ± 7.5% 9 out of 17 P,p-DDT 17 9.95 9.900 ±0.69 ± 7.0% 10 out of 17 See discussion of results for Sample No. 1 in text. TABLE 8. — Analyses of starlings Location Weight (GRAMS) Mean Values and Ranges in PPM Dieldrin DDE TDE p.p'-DDT PCB Canada New Brunswick singly (10) 71-96 0.045 0.349 0.021 0.030 (0.014-0.151) (0.058-1.28) (0.004-0.092) (0.004-0.166) 5 groups of 5 71-97 0.095 0.481 0.020 0.016 (0.023-0.262) (0.126-0.803) (0.014-0.025) (0.008-0.023) British Columbia Singly (10) 71-97 0.036 0.520 0.024 0.038 (0.004-0.125) (0.039-3.30) (0.006-0.124) (0.005-0.174) 5 groups of 5 74-100 0.035 0.306 0.011 0.008 (0.007-0.057) (0.111-0.636) (0.005-0.019) (0.004-0.012) Vol. 4, No. 3, December 1970 129 TABLE 8. — Analyses of starlings — Continued Location Weight (GRAMS) Mean Values and Range. IN PPM DlELDBIN DDE TDE P,p'-DDT PCS Canada — Continuec Ontario singly (10) 73-88 0.014 0.567 0.006 0.011 (0.003-0.040) (0.075-1.51) (0.002-0.012) (0.007-0.015) 5 groups of 5 79-105 0.009 0.423 0.017 0.022 (0.003-0.015) (0.146-0.861) (0.007-0.050) (0.004-0.056) Euratom Ispra singly (10) 72-93 <0.001 0.172 (0.05-0.35) 0.043 (0.03-0.06) 0.065 (0.04-0.12) 5 groups of 4 72-91 <0.001 0.164 (0.12-0.24) 0.048 (0.03-0.06) 0.066 (0.05-0.08) Finland Evo singly 5 80-96 0.013 «0.0I-0.03) 0.09 (0.07-0.11) <0.01 0.034 (<0.01-0.I1) 0.15-0.19 Ryttyla singly (18) 61-77 <0.01 0.085 (0.04-0.35) <0.01 0.01 (<0.01-0.03) 0.04-0.35 5 groups of 5 0.02 «O.0I-0.08) 0.10 (0.08-0.12) <0.01 0.018 (<0.01-0.04) 0.11-0.19 Netherlands Oude Molen singly (10) 72-80 0.029 (0.01-0.07) 0.29 (0.05-1.50) NC O.IO (0.06-0.23) 6 groups of 4 62-90 0.027 (0.02-0.04) 0.19 (0.11-0.31) NC 0.15 (0.10-0.18) Eext singly (10) 74-90 0.04 (0.01-0.09) 0.24 (0.03-1.29) NC 0.16 (0.07-0.41) 2 groups of 5 62-85 0.01 (<0.01-0.02) 0.23 (0.08-0.38) NC 0.12 (0.10-0,13) Beers Mill singly (10) 70-84 0.05 «0.01-0.13) 0.30 (0.09-1.26) NC 0.23 (0.12-0.41) 4 groups of 4 62-83 0.03 (0.02-0.04) 0.18 (0.08-0.26) NC 0.12 (0.08-0.16) Norway singly (24) 72-100 Method (a) 0.042 (0.01-0.14) 0.073 (0.02-0.17) <0.01 <0.001 Method (b) 0.032 (<0.01-0.17) 0.091 (0.03-0.18) <0.01 <0.001 <0.01-0.03 Portuga] Ourique singly (3) 84-89 0.003 0.044 0.003 0.007 (0.001-0.007) (0.035-0.150) (0.003) (0.006-0.008) Campo Maior singly (4) 78-90 0.012 1.03 0.0045 0.004 (0.004-0.024) (0.61-1,20) (<0.001-0.011) «0.001-0.015) S, Mansor singly (2) 0.245 (0.137-0.353) <0.001 <0.001 <0.00I Spain Maderuelo (Segovia) singly (5) 84-99 0.013 0.042 0.057 0.099 T (0.010-0.020) (0.022-0.063) (0.035-0.080) (0.055-0.150) 4 groups of 5 88-97 T 0.071 (0.030-0.140) 0.076 (0.062-0.096) ■' 0.241 ( V 0.047-0.860) Sweden Krankesjon singly d- (17) 25-97 0.013 (T-0.062) 0.124 (0.039-0.300) <0.003 <0.003 0.066 (0.027-0.180) ?(13) 50-97 0.012 (T-0.065) 0.069 (0.037-0.110) <0.003 <0,003 0.033 (0.017-0.056) Villingsberg singly (10) 72-91 0.005 (<0.001-0.017) 0.072 (0.015-0.130) <0.003 <0.003 0.132 (0.049-0.300) Kvismaren singly (10) 80-93 0.013 (T-0.037) 0.074 0.026-0.110) <0.003 <0.003 0.078 (0.030-0.200) Bcda Bruk s.ngly (10) 76-94 0.037 «0.001-0.200) 0.081 (0.024-0.230) <0.003 <0.001 0,114 (0.020-0.360) 130 Pesticides Monitoring Journal TABLE 8. — Analyses of starlings — Continued Weight (GRAMS) Mean Values and Range S IN PPM DlELDRlN DDE TDE p,p-DDT PCB United Kingdom Nature con- servancy singly (7) 81-98 0.027 (0.01-0.09) 0.049 (0.02-0.08) — (0.055) «0.01-0.10) 4 groups of 5 78-94 0.045 (0.01-0.09) 0.053 (0.03-0.08) — 0.030 (0.01-0.07) United States Patuxent singly (10) 83-93 0.11 «0.02-0.42) 0.32 (0.07-0.92) <0.007 0.05 «0.02-0.26) T 5 groups of 5 82-86 <0.01 «0.01-0.09) 0.17 (0.11-0.29) <0.01 <0.01 «0.01-0.03) NOTE: T = Trace NC = Present, but not calculated. ? = Suspected TABLE 9. — Analyses of mussels Length of Mean Values and Ranges in PPM Weight (GRAMS) Shell (centi- meters) Location Dieldrin DDE TDE p,p'-DDT PCB Canada Portage Island (10) 6.3-6.8 0.001 0.016 0.012 0.010 «0.0001-0.001) (0.006-0.041) (0.004-0.016) (0.002-0.053) St. John (10 groups of 5) 2.6-3.2 0.001 0.027 0.016 0.057 «0.0001-0.002) (0.009-0.091) (0.008-0.030) (0.020-0.104) British Columbia (10) Fat O.S'-'c 4.0-5.0 0.002 0.021 0.010 0.038 (<0.0001-0.006) (0.007-0.106) (0.002-0.062) (0.015-0.126) Euratom Magra-Tillaro, 7 groups of 5 6.5-16.5 4.2-6.5 T <0.0O5 0.015 (0.012-0.021) 0.026 (0.020-0.030) 0.078 (0.046-0.099) Absent Ireland Clonakilty (10) singly 6.0-8.0 0.03 (0.01-0.05) 0.02 «0.01-0.07) <0.01 <0.01 (10) group 6.2-7.3 0.03 0.02 Kinsale (10) singly 6.9-8.3 0.016 «0.01-O.03) 0.01 «0.01-0.02) <0.01 <0.01 (10) group 0.01 0.02 East Ferry: (10) singly 7.1-7.5 0.04 (0.02-0.06) 0.02 (0.01-0.07) <0.01 <0.01 (10) group 0.04 0.05 Finland 6 groups of 10 5.02-6.71 2.8-3.1 <0.001 <0.01 «0.01-0.01) <0.0I (<0.01-0.01) 0.017 incl. o.p-DDT (<0.01-0.03) 0.06 Netherlands 1967 IJmuiden 5.5-14.5 0.016 (0.010-0.020) 0.020 (0.013-0.025) NC NC 0.44 Scheveningen 12.7-15.1 0.042 (0.031-0.057) 0.032 (0.022-0.041) NC NC 0.47 Grevelingendam W 10.0-24.3 0.009 (0.006-0.011) <0.009 NC NC 0.26 Grevelingendam E 13.5-18.8 0.006 (0.004-0.007) 0.022 (0.018-0.031) NC NC 0.46 Vol. 4, No. 3, December 1970 131 TABLE 9. — Analyses of mussels — Continued Weight (CRAMS) Length of Shell (CENTI- METERS) Mean Values and Ranges in PPM Location DlELDRIN DDE TDE p,p'-DDT PCB Netherlands — Continued 1968 Wadden Sea 24.6 <0.007 NC NC NC Schiermonnikoog 18.1 0.013 NC NC NC Den Helder 26.5 0.018 NC NC NC IJmuiden 25.2 0.018 NC NC NC Scheveningen Hoek van Holland 17.9-18.4 21.7 0.046 (0.039-0.053) 0.034 NC NC NC NC NC NC . Mean 0.46 Grevelingendam N 29.4 0.015 NC NC NC Grevelingendam E 34.2 0.010 NC NC NC WestkapeUe 29.4 0.012 NC NC NC Norway singly 52-113 7.3-8.9 0.0021 0.0015 0.0033 0.0036 (0.001-0.003) (0.001-0.002) (0.002-0.006) «0.00I 0.020) 3 groups of 3, 5. 6 2.2-3.2 T 0.005 <0.001 0.009 0.004 Portugal (0.004-0.006) «0.001-0.027) Aveiro 5.8-8.3 0.004 0.00° 0.004 0.012 Setubal 6.6-9.2 (0.001-0.005) 0.006 (0.004-0.013) 0,024 (O.0O2-0.006) 0.020 (0.005-0.017) 0.059 Cascais 3.0-6.1 (0.003-0.010) 0.034 (0.020-0.034) 0.036 (0.009-0,035) 0,059 (0,037-0,084) 0,184 Spain (0.022-0.047) (0.020-0.062) (0.039-0,083) (0,128-0.278) Vigo 12.03-40.30 7.52-9.20 0.001 0,014 0,046 0.053 Barcelona Sweden 2.74-5.11 4.16-5.30 <0.001 0.033 0,304 0.140 T Fiskebackskil 10.51-20.87 6.5-7.9 <0.0OI 0.004 <0.01 0.010 0.048 Graddo Asko 4.5-6.6 3,14-3.36 <0.001 <0.001 (0.001-0.013) 0.003 0.008 <0.01 <0.01 (0.004-0.026) 0.010 0.015 0.032 0.036 Lundakrabukten 4.53-6.19 3.63-3.76 <0.001 (0,002-0,017) 0,007 <0.01 (0.006-0.023) 0.007 0.029 United Kingdom (0,005-0,009) (0,003-0.011) R. Roach 0.0230 0,0254 0.0304 0.011 R. Crouch (0.012-0.047) 0.0204 (0,018-0,033 0.0230 (0.019-0.055) 0,0328 (0,008-0,018) 0,005 Loch Linnhe 5.0-6.0 (0.012-0.036) 0.00335 (0.005-0.072) 0.0214 (0,010-0.073 0.0157 (0,002-0,012) <0,016 LochgoUhead 5.8-7.3 (0.002-0.006) 0.0168 (0.013-0.037) 0.0240 (0.009-0.032) 0.0279 «0,01-<0,02) <0.018 T United States (0.006-0.035) (0.007-0.087) (0.007-0.042) «0.0I-0.3) Modiolus demissus, groups 8.20-14.20 4.1-4.9 0.02 0.13 0.10 0.08 T «0.02-0.04) (0.08-0.18) (0.07-0.13) (0.05-0.12) NOTE: T = Trace; NC = Present, but not calculated (due to PCB interference). TABLE 10. — Analyses of pike Location Weight (KILO- GRAMS) Length (centi- meters) Ace (YEAR) Means and Ranges in PPM DlELDRIN DDE TDE p.p'-DDT PCB Canada Saskatchewan Quebec 1.02-2.61 1.14-3.79 53.4-67.3 58.5-76.2 4-5 4-6 0.000 «0.0001-0.001) 0.000 «0.0001-0.001) 0.024 (0.007-0.049) 0.009 (0.005-0.016) 0.010 (0.003-0.023) 0.003 (0.001-0.006) 0.035 (0.013-0.079) 0.013 (0.005-0.030) 132 Pesticides Monitoring Journal TABLE 10. — Analyses of pike — Continued Location Weight (KILO- GRAMS) Length (CENTI- METERS) Age (YEAR) Means and Ranges in PPM DIELDRIN DDE TDE p,p'-DDT PCS Euratom Ispra 0.49-6.00 39-100 2-6 <0.005 1.36 (0.26-4.13) 0.18 (0.06-0.53) 0.41 (0.08-1.57) Finland Lintuselka 1.06-L73 55-65 4 <0.005 <0.01 <0.01 0.018 (0.01-0.03) Hoikanlahti 0.61-1.45 48-60 4-6 <0.005 <0.01 <0.01 <0.01 Ireland L. Carra 1.56-8.16 53-94 5-9 0.044 (0.01-0.22) 0.020 «0.01-0.05) T T Netherlands 0.262-0.364 30.9-34.5 0.0022 (0.0010-0.0034) 0.0057 (0.004-0.007) T T Mean 0.046 Norway Method (a) 0.31-3.05 36-86 2-11 <0.001 0.0094 (0.004-0.019) 0.0021 (0.001-0.006) 0.0026 (0.001-0.006) Method (b) <0.001 0.0138 (0.005-0.045) <0.001 0.0041 (T-O.OlO) 0.002-0.011 Spain Santillana (2) 0.73-1.20 42.5-54.0 2 <0.001 0.026 (0.018-0.034) 0.049 (0.023-0.076) 0.010 «0.001-0.020) T Buendia (1) Garcia Sola (2) 2.10 66.5 3 <0.001 0.090 0.028 0.036 T 4.65-1.76 78.5-60.5 2-4 <0.001 0.815 0.310 0.018 T (0.790-0.840) (0.130-0.490) «0.001-0.037) Sweden Lake Bolmen 0.600-0.975 39-44 T 0.011 (0.005-0.016) <0.01 0.031 (0.015-0.045) (included o,p'-DDT) 0.024 United Kingdom Loch TuUa 0.67-1.43 46-55 0.0011 «0.001-0.003) 0.0146 (0.006-0.021) 0.0046 (0.002-0.008) 0.0043 (0.002-0.008) Loch Choin 0.45-1.67 40-62 <0.001 «0.001-0.001) 0.0057 (0.003-0.009) 0.0034 (0.002-0.006) 0.0037 (0.002-0.008) Loch Skiach 1.85-6.13 64-93 <0.00I «0.001-0.002) 0.0166 (0.008-0.040) 0.0024 (0.001-0.005) 0.0055 (0.003-0.010) Loch Glassie 0.45-1.48 42-64 <0.0018 «0.001-0.002) 0.0036 (0.002-0.007) 0.0032 (0.001-0.011) 0.0044 (0.002-0.008) United States Garrison Res- ervoir (2) 2.72-3.49 73.7-79.2 4 esticides were separated and re- moved in four fractions as described by Mulhem (4). The four fractions would contain, if present, the follow- ing: I — dieldrin, lindane, heptachlor epoxide, endrin, 4,4'-dichlorobenzophenone (DCBP), methoxychlor, and dicofol; II — p.p'- and o.p'-DDD; III — o,p'-DDE, p.p'- and o,p-DDT; IV — p,p'-DDE, heptachlor, aldrin, and mirex. TTie TL fractions were analyzed separately by gas chromatography (GC) using three columns of different polarity; the operating parameters are shown in Table 2. The four fractions were analyzed on the OV-17 column, and the jjesticides detected in fractions II through IV were confirmed on the XE-60 column. The residues de- tected in fraction I were confirmed on the DEGS column. TABLE 2. — Chromatographic operating conditions using electron capture detection Columns, Glass 6"xVi" O.D. A B C Liquid Phase 3% OV-17 3% XE-60 12% DEGS Support Gas Chrom Q Gas Chrom Q Anachrom SD Mesh Size 100/120 60/80 100/110 N„ Flow Rate (ml/min) 100 100 85 Temperature (C) 190 170 190 Retention Time of 14.0 16.3 11.0 dieldrin (minutes) In addition to the TL zonal separations and dual-column gas chromatographic confirmation, the residues in 10% of the samples were confirmed by TLC. This TLC con- firmation also showed that the DCBP detected by GC was not the GC breakdown product of dicofol. Some samples were also treated with alkali to confirm DDT. The polychlorinated biphenyl (PCB) compounds were present in fraction IV and in considerably smaller amounts in fraction III. Inspection of the GC profiles for fraction IV from these specimens indicated that the maximum ratio of PCB's (Aroclor 1254 as reference) to DDE encountered was 1:1. Therefore, a study was made to determine if PCB's could interfere in the analysis of DDE. Standards were prepared to contain the following ratios of PCB to DDE : 1 : 1 , 2 : 1 , 3 : 1 , and 5:1. The standards were zoned on the TL plate, as described above, and DDE was analyzed on the OV-17 column. PCB compounds did not interfere in the quantitative determination of DDE unless the ratio exceeded 2:1. To determine the recovery efficiency of the analytical procedure. 20-g aliquots of an eagle carcass homogenate containing trace quantities of pesticides were spiked to contain the following: 15 ppm DDE. 7.5 ppm DDD, 4.5 ppm DDT. 5.0 ppm dieldrin. 4.0 ppm heptachlor epoxide, 2.0 ppm DCBP, and 3.5 ppm o,p'-DDT. The average recoveries were: 95% DDE. 102% DDD, 110% DDT. 106% dieldrin, 112% heptachlor epoxide, 75% DCBP, and 107% o.p'-DDT. In addition, experimental quail tissues containing biologically incorporated carbon- 14-labeled p,p'-r>DT and dieldrin were analyzed by liquid scintillation techniques. The average recovery from all tissues was 80% for DDT and 79% for dieldrin (5). The residue levels reported for the eagle samples were not corrected for recovery: the lower limit of sensitivity was approximately 0.05 ppm wet weight. Results and Discussion RESIDUES The chlorinated pesticide residues found in 69 bald eagle carcasses and brains are summarized in Table 3. Median values are presented instead of means because of the skewness of the data. All eagle carcass samples contained DDE; 68 contained dieldrin; and 64 con- tained DDD. In addition, 39 samples contained DDT; 34 contained heptachlor epoxide; and 24 contained DCBP. All samples contained PCB compounds; their presence was confirmed in six samples by gas chroma- tography-mass spectrographic (GC-MS) analysis. The characterization of the individual PCB compounds for two of these samples has been reported by Bagley et al. (1). The median value of DDE in the carcass samples was lower in 1968 than in the preceding years. However, it cannot be concluded that a general average decrease did, in fact, occur due to ( 1 ) the wide range in DDE levels, (2) the wide distribution of collection sites, and (3) the relatively small number of samples collected in any one area over the 3-year period. For example, 16 States were represented in only 1 of the 3 years, and 142 Pesticides Monitoring Joltrnal TABLE 3. — Pesticide residues in bald eagles, 1966-68 [T = <0.05 ppm] Year Residues in ppm i Compound Carcass Brain Median Range Na Median Range N2 P,p'-DDE 1966 1967 1968 11.80 16.55 4.92 0.5-136.0 0.6-263.0 0.4-104.0 21 22 26 1.50 1.81 0.92 T- 35.9 T-149.0 T- 113.9 21 21 26 p,P'-DDD 1966 1967 1968 1. 10 1.09 0.85 T- 13.8 <0.1- 79.2 T- 24.8 21 20 23 0.17 0.70 1.00 T- 1.5 T- 14.4 T- 3.8 14 14 12 p,p'-DDT 1966 1967 1968 0.20 0.20 0.15 T- 1.3 T- 14.1 T- 0.5 14 15 10 T T 0.42 T T- 20.3 0.1- 0.7 4 3 2 Dieldrin 1966 1967 1968 0.59 0.60 0.47 T- 2.1 <0.1- 8.2 T- 22.3 21 21 26 0.10 0.27 0.77 T- 0.6 T- 9.5 T- 7.0 17 16 17 Heptachlor epoxide 1966 1967 1968 0.07 0.08 0.08 T- 0.25 T- 0.34 T- 0.21 13 12 9 0.02 0.19 0.07 T- 0.04 T- 0.2 T- 0.21 5 5 5 Dichlorobenzophenone 1966 1967 1968 1.20 0.71 0.53 0.3- 5.0 0.1- 3.5 T- 6.4 5 9 10 0.20 0.60 0.45 T- 2.0 T- 1.9 0.2- 1.2 4 5 6 ^ Calculated on a wet-weight basis. 'Number of specimens that contained residues; the median is based on this number. NOTE: A total of 21 birds were collected in 1966, 22 birds in 1967, and 26 birds in 1968. 5 States were represented in only 2 of the 3 years. Four contiguous States — Iowa, Michigan, Minnesota, and Wis- consin— were represented in each year of the reporting period. The concentrations of DDE in individual carcass samples from these States are shown in Table 4. Analysis of variance of the log-transformed data showed no sig- nificant difference (P=0.05) between years, either for the four States combined or when Minnesota or Wis- consin were tested separately. A trend may become more evident upon the analysis of samples collected in 1969 and 1970. One of the principal findings of these analyses is that eight specimens had concentrations of dieldrin in the TABLE 4. — DDE residues in carca'i-2,3-quinoxalinedithiol cyclic ^.i'-dithiocarbonate methylmercuric dicyandiamide (CsHr.N^Hg) l,r-dimethyl-4,4'-bipyridinium pentachloronitrobenzene pentachlorophenol 5-propyl butylethylthiocarbamate l,l-dichloro-2,2-bis (p-ethylphenyl) ethane 0,0-diethyl 5-(ethylthio) methyl phosphorodithioate l-chIoro-diethylcarbamoyI-l-propen-2-yl dimethyl phosphate mixture of ethylene bis dithiocarbamic acid and zinc salts and sulfides 0,0-dimethyl i'-phthalimidomelhyl phosphorodithioate C2.1H-O0 2-(2,4,5-trichlorophenoxy) propionic acid 2-chloro-4,6-bis ( ethylamino ) -j-triazine NaAsOj NaCIOs terpene polychlorinates (65-66% chlorine) S 2,4.5-trichIorophenoxyacetic acid 2,2-bis (p-chlorophenyl ) -1 , 1-dichloroethane tetraethyl pyrophosphate p-chlorophenyl 2,4,5-trichlorophenyl sulfone technical chlorinated camphene (67-69% chlorine) octachlorocamphene dimethyl (l-hydroxy-2.2,2-trichloroethyl) phosphonate alpha, alpha, fl/p/?rt-trifluoro-2.6-dinitro-jV..V-dipropyl-p-toIuidine 5-propyl dipropyllhiocarbamate Zn ZnaPa zinc ethylenebis[dithiocarbamateI zinc dimethyldithiocarbamate of 0,0-dimethyl O.p-nitro= 166 Pesticides Monitoring Journal APPENDIX Chemical Names of Compounds Mentioned in This Issue''' ALDRIN AMITROLE DCBP DICOFOL DDD (TDE) DDE DDT (including its isomers and dehydrochlorination products) DIELDRIN ENDRIN HEPTACHLOR HEPTACHLOR EPOXIDE LINDANE METHOXYCHLOR MIREX ZINEB 2,4-D 2,4-DB 2.4,5-TP Not less than 95% of 1,2,3.4. 10,10-hexachIoro-1.4.4a.5,8.8a-hexahydro-1.4-<'nrfo-eio-5.8-dimethanonaphthalene 3-amino-l,2,4-triazole 4,4'-diclilorobenzophenone 4,4'-dichloro-fl- ( trichloromethy 1 ) benzhydrol l,l-dichIoro-2,2-bis(p-chlorophenynethane; technical DDD contains some o,p'-isomer also. l,l-dichloro-2.2-bis(p-chlorophenyl) ethylene l.l,l-trichloro-2,2-bis(p-chlorophenyl)cthane; technical DDT consists of a mixture of the p,p'-isomer and the o,p'-isomer (in a ratio of about 3 or 4 to 1 ) Not less than 85% of l,2,3,4,10,I0-hexachIoro-6,7-epoxy-l,4,4a,5.6.7.8.8a-ociahydro-1.4-c:)<;o-i'.ro-5.8-dimethano= naphthalene l,2,3,4,10,10-hexachloro-6,7-epoxy-l,4,4a.5,6.7.8,8a-octahydro-1.4-enrfo-<')irfo-5.8-dimethanonaphthalene 1,4.5 ,6,7.8,8-heptachloro-3a.4,7,7a-telrahydro-4,7-methanoindene l,4,5,6,7,8,8-heptachloro-2,3-epoxy-3a,4,7,7a-tetrahydro-4.7-methanoindan 1,2,3,4.5,6-hexachlorocyclohexane, 99% or more gamma isomer 1, 1, l-trichloro-2.2-bis(p-methoxyphenyl) ethane dodecachlorooctahydro-1.3.4-metheno-l//-cyclobutatc(fIpentalene zinc ethyIenebis[dithiocarbamate] 2.4-dichlorophenoxyacetic acid 4-(2,4-dichlorophenoxy) butyric acid 2-(2,4,5-trichIorophenoxy ) propionic acid ♦Does not include chemical names of compounds mentioned in the papers by (1) P. E. Corneliussen and (2) Lynn J. Stevens el al. Since the lists of compounds mentioned in these papers were extensive, they have been included in tables in the individual papers. Vol. 4, No. 3, December 1970 167 Information for Contributors The Pesticides Monitoring Journal welcomes from all sources qualified data and interpretive information which contribute to the understanding and evaluation of pesticides and their residues in relation to man and his environment. The publication is distributed principally to scientists and technicians associated with pesticide monitoring, research, and other programs concerned with the fate of pesticides following their application. Additional circulation is maintained for persons with related in- terests, notably those in the agricultural, chemical manu- facturing, and food processing industries; medical and public health workers; and conservationists. Authors are responsible for the accuracy and validity of their data and interpretations, including tables, charts, and refer- ences. Accuracy, reliability, and limitations of the sampling and analytical methods employed must be clearly demonstrated through the use of appropriate procedures, such as recovery experiments at appropriate levels, confirmatory tests, internal standards, and inter- laboratory checks. The procedure employed should be referenced or outlined in brief form, and crucial points or modifications should be noted. Check or control samples should be employed where possible, and the sensitivity of the method should be given, particularly when very low levels of pesticides are being reported. Specific note should be made regarding correction of data for percent recoveries. Preparation of manuscripts should be in con- formance to the Style Manual for Biological Journals, American Institute of Biological Sciences, Washington, D. C, and/or the Style Manual of the United States Government Print- ing Office. An abstract (not to exceed 200 words) should accompany each manuscript submitted. All material should be submitted in duplicate (original and one carbon) and sent by first-class mail in flat form — not folded or rolled. Manuscripts should be typed on 8'/2 x 11 inch paper with generous margins on all sides, and each page should end with a completed para- graph. All copy, including tables and references, should be double spaced, and all pages should be num- 168 bered. The first page of the manuscript must contain authors' full names listed under the titles, with affiliations, and addresses footnoted below. Charts, illustrations, and tables, properly titled, should be appended at the end of the article with a notation in text to show where they should be inserted. Charts should be drawn so the numbers and texts will be legible when considerably reduced for publication. All drawings should be done in black ink on plain white paper. Photographs should be made on glossy paper. Details should be clear, but size is not important. The "number system" should be used for litera- ture citations in the text. List references alpha- betically, giving name of author/ s/, year, full title of article, exact name of periodical, volume, and inclusive pages. Pesticides ordinarily should be identified by common or generic names approved by national scientific so- cieties. The first reference to a particular pesticide should be followed by the chemical or scientific name in parentheses — assigned in accordance with Chemical Abstracts nomenclature. Structural chemical formulas should be used when appropriate. Published data and information require prior approval by the Editorial Advisory Board; however, endorsement of published in- formation by any specific Federal agency is not intended or to be implied. Authors of accepted manuscripts will receive edited typescripts for approval before type is set. After publication, senior authors will be provided with 100 reprints. Manuscripts are received and reviewed with the under- standing that they previously have not been accepted for technical publication elsewhere. If a paper has been given or is intended for presentation at a meeting, or if a significant portion of its contents has been published or submitted for publication elsewhere, notation of such should be provided. Correspondence on editorial matters or circulation mat- ters relating to official subscriptions should be addressed to: Mrs. Sylvia P. O'Rear, Editorial Manager, Pesti- cides Monitoring Journal, Division of Pesticide Com- munity Studies, Food and Drug Administration, 4770 Buford Highway. Bldg. 29. Chamblee, Ga. 30341. Pesticides Monitoring Journal The Pesticides Monitoring Journal is published quarterly under the auspices of the WORKING GROUP, Subcommittee on Pesticides, President's Cabinet Committee on the Environment, and its Panel on Pesticide Monitoring as a source of information on pesticide levels relative to man and his environment. The WORKING GROUP is comprised of representatives of the U. S. Departments of Agricul- ture; Defense; the Interior; Health, Education, and Welfare; State; and Transportation. The Pesticide Monitoring Panel consists of representatives of the Agricultural Research Service. Consumer and Marketing Service, Federal Extension Service, Forest Service, Department of Defense, Fish and Wildlife Service, Geological Survey, Federal Water Quality Administration, Food and Drug Administration, Environmental Health Service, National Science Foundation, and Tennessee Valley Authority. Publication of the Pesticides Monitoring Journal is carried out by the Division of Pesticide Community Studies of the Environmental Protection Agency- Pesticide monitoring activities of the Federal Government, particularly in those agencies repre- sented on the Pesticide Monitoring Panel which participate in operation of the national pesticides monitoring network, are expected to be principal sources of data and interpretive articles. How- ever, pertinent data in summarized form, together with interpretive discussions, are invited from both Federal and non-Federal sources, including those associated with State and community monitoring programs, universities, hospitals, and nongovernmental research institutions, both domestic and foreign. Results of studies in which monitoring data play a major or minor role or serve as support for research investigation also are welcome; however, the Journal is not intended as a primary medium for the publication of basic research. Manuscripts received for publication are reviewed by an Editorial Advisory Board established by the Monitoring Panel. Authors are given the benefit of review comments prior to publication. Editorial Advisory Board members are: Reo E. Duggan, Food and Drug Administration, Chairman Anne R. Yobs, Environmental Protection Agency Andrew W. Briedenbach, Environmental Health Service Thomas W. Duke, Environmental Protection Agency William F. Stickel, Fish and Wildlife Service Milton S. Schechter, Agricultural Research Service Paul F. Sand, Agricultural Research Service Mention of trade names or commercial sources in the Pesticides Monitoring Journal is for identification only and does not represent endorsement by any Federal agency. Address correspondence to: Mrs. Sylvia P. O'Rear Editorial Manager PESTICIDES MONITORING JOURNAL Environmental Protection Agency 4770 Buford Highway, Bldg. 29 Chamblee, Georgia 30341 CONTENTS Volume 4 March 1971 Number 4 Page RESIDUES IN FISH, WILDLIFE, AND ESTUARIES Residues of polychlorobiphenyls in biological samples 169 A. Richardson, J. Robinson, A. N. Crabtree, and M. K. Baldwin Dieldrin and endrin concentrations in a Louisiana estuary 177 D. R. Rowe, L. W. Canter, P. J. Snyder, and J. W. Mason Monitoring ecological conditions associated with wide-scale applications of DMA 2,4-D to aquatic environments 184 T. A. Wojtalik, T. F. Hall, and Larry O. Hill PESTICIDES IN AIR Volatilization of soil-applied DDT and DDD from flooded and nonflooded plots 204 G. H. Willis, J. F. Parr, and S. Smith PESTICIDES IN SOIL Soil persistence of fungicides — experimental design, sampling, chemical analysis, and statistical evaluation 209 W. J. Polzin, I. F. Brown, Jr., J. A. Manthey, and G. W. Probst PESTICIDES IN WATER Chlorinated hydrocarbon pesticides in Iowa rivers, 216 Lauren G. Johnson and Robert L. Morris APPENDIX Chemical names of compounds mentioned in this issue 220 ACKNOWLEDGMENT 221 ANNOUNCEMENT Please note that the Information for Contributors has been revised to include an invitation for the submission of "brief' reports for publication in the Journal. It is the consensus of the Editorial Advisor>' Board and the Monitoring Panel that a useful purpose will be served in providing a place for the publication of data of a preliminary nature or studies of limited scope. Frequently, although such studies do not warrant publication as full reports, they do provide timely and informative data that should be brought to the attention of the readers. These papers will be identified as BRIEFS at the end of each issue. RESIDUES IN FISH, WILDLIFE, AND ESTUARIES Residues of Polychlorobiphenyh in Biological Samples^ A. Richardson, J. Robinson, A. N. Crabtree, and M. K. Baldwin ABSTRACT The presence of polychlorobiphenyh (poly chlorinated bi- phenyls) in samples of agnatic origin has been demonstrated and confirmed by combined gas-liquid chromatography — mass spectroscopy. The relevance of this finding to the de- termination of organochlorine insecticides is discussed. Introduction During the past 15 years the detection and quantitation of organochlorine insecticides in biological material has received considerable attention. With the introduction of the electron capture detector, used in conjunction with gas-liquid chromatography, it has become possible to detect residues of less than one part per hundred mil- lion. However, the electron capture detector is not specific for organochlorine compounds (1). and even if it were, the occurrence of polyhalogenated hydrocar- bons other than organochlorine insecticides in the en- vironment would make the detection and quantitation of chlorinated pesticides difficult. The presence of chlorinated compounds, other than in- secticides, as residues in environmental samples has been suspected for some years. Roburn (2) showed that the total organochlorine content of some wildlife samples was greater than the organochlorine insecticide content determined by gas-liquid chromatography and inferred the presence of other organochlorine compounds. Eidel- man (3) examined Norwegian cod liver oil and reported the presence of compounds "in the region of DDE, TDE and DDT" on gas chromatograms. Paper chromatog- raphy also indicated the presence of halogenated com- pounds which were not known insecticides. Harrison (4) reported the presence of compounds in biological samples which interfered with the determination of p,p'-DDT and p,p'-TDE. Robinson et al. (5) detected From the Tunstall Laboratory, Shell Research Limited, Sittingboume, Kent, England. Vol. 4, No. 4, March 1971 compounds of unknown identity in many marine or- ganisms: these compounds, which also gave a response with the microcoulometric detector, were of low polarity and had retention times similar to or longer than that of p.p'-DDE. The identification of a group of polychlorinated com- pounds by Jensen et al. (6,7) and Widmark (8) in the Swedish environment using a combination of gas chro- matography and mass spectroscopy first drew attention to polychlorobiphenyls as environmental pollutants. The presence of these materials has also been reported in Britain (9.10) and North America (11): the conclusions drawn by these reports were based upon the chromato- graphic behavior of the compounds in the samples ana- lyzed and should, therefore, only be regarded as tenta- tive. Koeman et al. (12). however, examined samples collected in Holland by mass spectrometer and produced clear evidence of the presence of polychlorobiphenyls, together with other unidentified organochlorine com- pounds, and subsequently Bagley et al. (13) have used a similar technique to identify polychlorobiphenyls in bald eagle samples in the United States. More recently, Reynolds (14) described a procedure for the separation of the more polar chlorinated insecti- cides from the polychlorobiphenyls, but his report stated that no positive identification had been obtained from polychlorobiphenyls in wildlife. Holden et al. (15) also described a similar procedure, but p.p'-DDE is not sep- arated from the polychlorobiphenyls by the procedure of Reynolds (14) or Holden et al. (15). The composition of commercial blends of polychlorobi- phenyls is not known, but mass spectral examination of a particular commercial product, Aroclor 1254, indi- cated the presence of chlorinated biphenyls containing three to seven chlorine atoms per molecule (16). A synthetic mixture of mono- and di-chlorobiphenyls has 169 been separated on capillary columns (17), and there is also a report on the separation of the components of a commercial sample of chlorinated biphenyls in the dis- tillation range 325-360 C by programmed temperature, capillary GLC (18). Recently we investigated the presence of organochlo- rine insecticides in the environment. Since the largest residues in the United Kingdom occur in birds asso- ciated with the aquatic and marine environment, tissues of the shag and the heron have been analyzed for con- tent of organochlorine insecticides. Herring oil has also been examined because of the obvious importance of fish in the diet of these birds. The results of these monitoring surveys will be published later. Experimental Design The method used routinely for the determination of chlorinated pesticides in environmental samples is given below. The extraction and cleanup procedure used is a modification of that described by de Faubert Maunder et al. (19). It is suitable for the determination of -y-BHC, aldrin, heptachlor. p.p'-DDE, p.p'-TDE. p,p'-DDT, heptachlor epoxide, endrin, and dieldrin at concentra- tions of one part per hundred million in animal tissues, fats, oils, and eggs. The procedure consists of extrac- tion with hexane or hexane/ acetone followed by cleanup with dimethylformamide, hexane partition, liquid-solid chromatography on Florisil, and determination of the chlorinated insecticides by gas-liquid chromatography with electron capture detection (20). APPARATUS Separating funnels: Volumetric flasks: Flat bottom flasks: Graduated stoppered cylinders: Chromatographic column: 100-ml, 250-ml, and 1- liter capacity. 10-ml. 20-ml, 25-ml, 50-mI, and 100-ml capac- ity. 150 ml. 25-ml capacity. The column consists of thick wall glass tubing 55 cm long X 0.5 cm bore. A female S29 ball joint is fused at the bottom. The outlet of the column is made from a female S13 joint. The top of the column is connected to a S29 male joint which fits to an air line. Soxhlet extraction apparatus High speed homogenizer Centrifuge Beakers 170 REAGENTS Sodium sidpJmte, anhydrous granular. General purpose reagent. Wash with hexane prior to use. Hexane petroleum fraction SBP 60-70. Redistill and collect fraction boiling between 64 and 66 C. Acetone. General purpose reagent, redistilled. Sand. Horticultural sharp sand, hexane washed. Diethyl ether. Analytical reagent, redistilled. Florisil 60/100 mesh. Activate by heating overnight al 120 C. Deactivate by addition of 3% v/w distilled water into conical flask, stopper, and shake for 30 minutes. Distilled water. Potassium oxalate. General purpose reagent. Methanol. Analytical reagent grade. All reagents and apparatus should be examined by gas- liquid chromatography of a hexane extract to ensur&|2 freedom from contamination. S EXTRACTION Animal fats, muscle tissue, kidney, liver, and brain Freeze sample by storing in polystyrene box witf Cardice. With sharp knife chop up frozen sample, mi) well, and weigh 4 g into a 250-mI beaker. Add 10 j of sand and grind with a heavy glass rod with flattenei end; add sufficient sodium sulphate and grind again t( give a dry uniform granular mass. Warm the grounc material successively with 50, 30, 30, and 30 ml volumei of 2:1 v/v hexane:acetone on a steam bath stirring care fully until the solvent boils gently. Decant the extract: into a 150-ml flat-bottom flask and evaporate the bulkec extracts to near dryness on a steam bath. Add 25 m of hexane and re-evaporate to remove all traces ol acetone. Transfer to a 100-ml volumetric flask by filter- ing through a filter funnel containing a glass wool plug covered with sodium sulphate and when cool make up to the mark with hexane. Treat a 25-ml solution of this by DMF partition. Eiigs. Weigh the egg contents, or the whole egg if con- tents are dehydrated. If the eggs are fluid, incorporate sufficient sodium sulphate to give a granular mass. If they are dehydrated and solid, grind them with sand and sodium sulphate. Transfer the granular mass to a suitable extraction thimble and extract for IVi hours in a Soxhlet extractor with acetone:hexane (1:2). Evap- orate the contents of the extraction flask to near dryness on a steam bath. Add 25 ml of hexane and evaporate to remove all traces of acetone. Transfer to a 100-ml volumetric flask by filtering through a filter funnel con- taining a glass wool plug covered with sodium sulphate and when cool make up to volume with hexane. Pesticides Monitoring Journal Fish Oil. Dissolve 4 g of fish oil in hexane in a grad- uated 100-ml flask and make up to volume with hexane. Take 25 ml for DMF partition. LIQUID-LIQUID PARTITION PROCESS Place 10 ml of DMF in a 100-ml separatory funnel. Allow a little of the DMF to run through into a lower 100-ml separatory funnel to lubricate both taps with clean DMF. This ensures that any leaking taps can be discarded without loss of sample. Place 25 ml of the hexane extract of the biological sample into the upper 100-ml separatory funnel, shake vigorously for 1 min- ute, allow the layers to separate, and run off the DMF phase into the lower 100-ml separatory funnel. Shake the hexane extract with two further 10-ml portions of DMF and add the DMF extracts to the first DMF ex- tract. Discard the hexane phase. Wash the combined DMF extracts with 10 ml of hexane saturated with DMF. The hexane used for this washing is then ex- tracted with an additional 10 ml of DMF. This extract and the original 30 ml of DMF extract are added to a 250-ml separatory funnel containing 180 ml of 2% aqueous sodium sulphate solution and 10 ml of hexane. Shake the aqueous sodium sulphate/ DMF/ hexane mix- ture and allow to settle for 40 minutes. Run off the aqueous phase and wash the hexane with two further 20-mI portions of distilled water. Dry the hexane ex- tract by using a small quantity of sodium sulphate be- fore the next cleanup procedure. LIQUID-SOLID CHROMATOGRAPHY Using a plug of hexane-washed cotton wool as a sup- port, add 3 g of Florisil to the column and top with 1 cm of sodium sulphate. Run on to the column the hexane solution obtained from the DMF partition and wash in with a little hexane. Using a 25-ml graduated flask as a receiver, elute with hexane to a volume of 25 ml. Replace the receiver with another 25-ml flask and elute with 25 ml of 10% diethyl ether in hexane. The first fraction will contain any y-BHC, heptachlor, aldrin, p.p'-DDE and PCB's, p.p'-DDD, and p,p'-DDT which may be present in the original extract. The second fraction contains any heptachlor epoxide, dieldrin, and endrin which may be present. GAS-LIQUID CHROMATOGRAPHY Instrument conditions for first fraction: Instrument conditions for second fraction: Column: Column packing: Carrier gas: Inlet pressure: Detector: Temperature: 1 m X 3 mm i.d., all glass 3.8% SE-30 on Diatoport S 80/100 mesh No, oxygen free 1.75kg/cm2 Electron capture 184 C Column: Column packing: Carrier gas: Inlet pressure: Detector: Temperature: 1.5 m X 3 mm i.d., all glass 2% Oronite polybutene -f 0.2% Epikote 1001 on Celite 85/100 mesh No, oxygen free 1.75 kg/cm^ Electron capture 184 C Standard solutions: Standard 1 (a) — A hexane solution containing 0.004 |Lig/ml of y-BHC, 0.004 /ig/ml aldrin, and 0.04 jxg/ml p.p'-DDT. Standard 1 (h) — A hexane solution containing 0.004 iu,g/ml of heptachlor, 0.02 jxg/m\ /^.p'-DDE, and 0.02 /xg/ml p.p'-DDD. Standard 2 — A hexane solution containing 0.01 jMg/ml of heptachlor epoxide, 0.02 /Ag/ml dieldrin, and 0.02 |Lig/ml endrin. METHOD Inject 20 fj} of sample into the appropriate column and record the chromatogram. The sample may then be diluted or concentrated to give a peak of the same height as in the appropriate standard. Because of the decrease in sensitivity of the detector during prolonged running, it is impracticable to con- struct a calibration curve for the estimation of insecticide present. To overcome this difficulty it is essential to inject a standard after every two samples and to average the two standard peak heights for calculating the con- centration of the insecticide in the biological samples between the standard samples. Calculations: Let mass of sample = f^ g Initial volume of hexane containing IF g = 100 ml Volume taken for cleanup = 5 ml Final volume after cleanup (20 or 25 ml) = Z ml Volume of standard = volume of sample injected ^ 20 jul Concentration of standard z= C fig Dilution or concentration of final volume = — Mean peak height of standard at Rx = Ps cm Peak height of sample at Rt = Pe oti Then ppm of each component ^ Pp X 100 X Z X D X C Ps X W X 5 X Z ppm Vol. 4, No. 4, March 1971 171 EXTRACTION AND CLEANUP OF SAMPLES FOR GAS-LIQUID CHROMATOGRAPHY— MASS SPECTRAL ANALYSIS Fish Oil. The crude herring oil (220 g) was dissolved in 500 ml of hexane and dried over anhydrous sodium sulphate. Eggs. The total egg weight was 67 g, and the final hexane extract volume prior to the DMF partition was 100 ml. CHROMATOGRAPHIC CLEANUP Fish Oil. The solution in hexane was concentrated to 50 ml and percolated through a column 1 cm wide containing 20 g of activated Florisil. The column was eluted with hexane until there was no further elution of peaks corresponding to those occurring in the first fraction of routinely analyzed samples. The resulting eluate was concentrated to 10 ml and passed down an activated Florisil column (6 g), eluted with hexane, and fractions collected until no further peaks were eluted, as shown by GLC electron capture. This process was repeated three times. The hexane eluate obtained as above contains p.p'-DDT, p,p'-DDE, p.p'-DDD, polychlorobiphenyls, and non- oxygenated "drin" insecticides. The more polar com- pounds such as dieldrin, endrin. and heptachlor epoxide are not eluted under these conditions. The solvent was evaporated from the eluate, and the residue was nitrated with a mixture of nitric and sulphuric acid 4:1 v/v. The acid mixture was poured into water and extracted with hexane. The resulting hexane solution was again chromatographed through activated Florisil, and the eluate was evaporated to 1 ml. Heron Egg. Purification of the heron egg extract, fol- lowing DMF partition, was achieved by chromatography on a deactivated Florisil column (25 g) using hexane (150 ml) as eluent. The concentrated eluate (10 ml) was further purified by chromatography on activated Florisil (6 g) using hexane as eluent. Portions of the above samples were examined by gas- liquid chromatography with electron capture detection, microcoulometric detection, and mass spectrometric and flame ionization detection. Standards of Aroclor 1254 were also examined by GLC with the same column conditions and detection pro- cedures. Gas chromatographic operating conditions for elec- tron capture detection and microcoulometric detection were as follows: Instrument: Column 1: Inlet pressure: Temperature: Carrier gas: Electron capture detector: Microcoulometric detector: Electrolyte: Oxygen flow rate: Pye 104 152 cm X 0.3 cm glass tube packed with 3.8% SE-30 on Diatoport S 1.25 kg/cm2 185 C No, oxygen free 63 Ni source with pulsed DC supply and 150 mi- crosecond pulse interval Inlet temperature 200 C Furnace temperature 900 C 70% V glacial acetic acid: 30% v water 60 ml/ minute The other columns mentioned were operated under the following conditions: Column 2: 152 cm X 0.3 cm glass tube packed with 2% GEXE60 on 80/100 mesh Gas Chrom Q Temperature: 185 C Inlet pressure: 1.4 kg/cm^ Column 3: 152 cm X 0.3 cm glass tube packed with 2% WFl on 100/120 mesh Gas Chrom Q Temperature: 185 C Inlet pressure: 1 kg/cm- The GLC/ mass spectrometer operating conditions were: Instrument: Chromatographic column: Carrier gas: Flow rate: Temperature: Ionization voltage: Pye 104 chromatograph with flame ionization de- tector and 1:1 split be- tween detector and mass spectrometer, an AEI MS 12 152 cm X 0.3 cm glass column packed with 2% Silicone SE-30 and 0.2% Epikote 1001 on 85/100 mesh Diatoport S Helium 40 ml/minute 184 C 70 volts Results and Discussion Gas chromatography on SE-30 columns followed by electron capture detection gave complex chromatograms for the hexane-eluted fractions of herring oil and heron 172 Pesticides Monitoring Journal egg samples (Fig. 1 and 2). Comparison of relative retention times of these compounds with those obtained by the gas chromatography of Aroclor 1254 (Fig. 3) further suggested the possibility that chlorinated bi- phenyls may be present, and this was supported by the detection of peaks with similar retention values by the microcoulometric detector (Fig. 4). Positive identification of the components responsible for interference in the determination of the organochlorine insecticides was made by mass spectroscopy coupled with gas chromatography. An example of part of the mass spectra obtained for two peaks from heron eggs is shown in Fig. 5 in comparison with two peaks of the same relative retention values on SE-30 derived from Aroclor 1254. The assignment of molecular weights and number of chlorine atoms per molecule for both samples is given in Table 1 . This indicates the molecular weight (CI ^ 35) and number of chlorine atoms per molecule of chlorinated compounds eluted at the quoted relative retention volumes (RRV p,p'-DDE = 1 .00). Where more than one component was present in a peak this is indicated by a second or third row of values. For comparison, the molecular weights and number of chlorine atoms per molecule of polychlorobiphenyls is given in Table 2. Apart from confirming the occurrence of polychlorobiphenyls in environmental samples it also shows that on SE-30 columns p,p'-DDE has a similar retention time to that of a pentachlorobiphenyl. It is of interest that an extract of Coho salmon from Lake Michigan extracted and cleaned up in the same way as the two samples of European origin also showed inter- ference by a pentachlorobiphenyl on the p.p'-DDE when examined by mass spectroscopy following gas chroma- tography on SE-30. FIGURE 2.—GLC/EC heron egg extract SE 30 1 r . 1 1 1 A A FIGURE 3. — Comparison of relative retention times, Aroclor 1254, heron egg, and herring oil FIGURE 4. — GLC/MC herring oil and heron egg extract FIGURE \.—GLC/EC herring oil extract Vol. 4, No. 4, March 1971 173 FIGURE 5. — Comparison of mass spectra obtained from heron egg extract and Aroclor JMMtMjJ\ Heron egg mass spectrum of peak 2 Heron egg mass spectrum of peak 8 288 (4CI) ^^^ '^^'' 358 (6CI) Aroclor 1254 mass spectrum of peak 9 Retention time data for some organochlorine insecticides and metabolites, and Aroclor 1254 were obtained on three liquid phases, silicone gum SE-30, GEXE60 and QF-1. These are compared in Fig. 6. The main interest is in the compounds of the DDT group since in general they occur, together with dieldrin, most widely in wild- life samples. However, since dieldrin can be easily separated from the polychlorobiphenyls by liquid-solid 174 FIGURE 6. — Comparison of relative retention times. Rode = 1.00 of Aroclor 1254 and chlorinated insecticides on GEXE60, QF-1. and SE-30 Ml J I I II II _ iiiii III II II I rr r" T if I J I II III T\\ I nil I chromatography, the quantitation of dieldrin in the pres- ence of these compounds presents no difficulty. There are considerable difficulties, on the other hand, in the case of DDT-type compounds and the PCB's. Eluate 1 of our procedure contains all the less polar compounds, i.e., lindane, aldrin, heptachlor, /?,p'-DDT, p.p'-DDD, p,p'-DDE, and the polychlorobiphenyls. On all three of the stationary phases studied in this investigation there are some similarities in the retention times of some of the polychlorobiphenyls and one or more of the DDT- type compounds. Preliminary experiments have indi- cated the feasibility of separating PCB's from p,p'-DDT and /j,p'-DDD by liquid-solid chromatography in a man- ner similar to Reynolds (14) and Holden (15), but we have not. to date, been able to separate p,p'-DDE in this manner. The use of thin layer chromatography as described by Reichel et al. (21) has been unsuccessful in achieving the separation also. The presence of halogenated compounds other than or- ganochlorine insecticides in environmental samples pre- sents a complex analytical problem which may not be solvable by gas chromatography. It would seem desir- able to carry out a separation based on the chemical differences between the two groups of compounds such as resistance of PCB's to oxidation and/ or alkaline hy- drolysis and treat the DDT compound or compounds produced in this manner as a measure of DDT and its congeners. Pesticides Monitoring Journal TABLE I. — Molecular weight and number of chlorine atoms per molecule of components separated by GLC Relative retention volumes (p,p'-DDE = 1.00)^ Molecular weights and corresponding number of Cl atoms per molecule AROCLOR 1254 0.42 256(3) 290(4) 0.50 290(4) 0.53 324(5) 0.70 324(5) 0.82 324(5) 1.00 358(6) 324(5) 1.07 358(6) 1.30 358(6) 1.58 358(6) 1.85 358(6) 2.25 392(7) 358(6) HERRING OIL 0.50 290(4) 0.70 324(5) 0.82 324(5) 324(5) 316(4) 1.30 358(6) 1.58 358(6) 1.85 392(7) 2.22 392(7) HERON EGG 0.35 0.45 0.69 0.81 1.00 1.30 1.52 1.84 2.05 2.42 2.76 3.07 3.75 256(3) 290(4) 324(5) 290(4) 324(5) 316(4) 324(5) 358(6) 324(5) 316(4) 358(6) 324(5) 358(6) 392(7) 358(6) 392(7) 392(7) 392(7) 426(8) Mass spectra not obtained. TABLE 2. — Molecular weight of polychlorinated biphenyls No. OF CHLORO- SUBSTITtJENTS Corresponding molecular weight 222 256 290 324 358 392 426 460 494 At the present time there are no results to indicate the relative metabolic biodegradation rates of polychlorobi- phenyls, and neither are pure components of the poly- chlorobiphenyls available. It is therefore impossible us- ing current analytical procedures to quantitatively determine these materials in the environment with any degree of confidence. Generally, in the North American continent. p.p'-DDT and its metabolites predominate over the PCB's, and therefore the determination of DDT compounds presents less difficulty than in the case of some European samples where the PCS concentration is much higher than the total DDT content. In the latter cases any results for the least polar chlorinated insecticides, particularly p,p'-DDE, must be considered as tentative until a method of separating the two groups of compounds is available. See Appendix tor chemical names of compounds mentioned in this paper. A cknowledgmeni The authors wish to thank Dr. W. Kelly for performing the mass spectra analyses. LITERATURE CITED (1) Lovelock, J. E. 1960. An ionisation detector for per- manent gases. Nature 187:49-50. (2) Roburn, J. 1965. A simple concentration cell technique for determining small amounts of halide ions and its use in the determination of residues of organochlorine pesticides. Analyst 90:467-475. (3) Eidelman, M. 1963. Determination of micro quantities of some chlorinated organic pesticide residues in edible fats and oils. J.A.O.A.C. 46:182-186. (4) Harrison, R. B. 1966. The detection and determination of organochlorine pesticide residues in wildlife with special reference to endrin. J. Sci. Food Agr. 17:10-13. (5) Robinson, J., A. Richardson, A. N. Crabtree, J. C. Coulson, and G. R. Polls. 1967. Organochlorine resi- dues in marine organisms. Nature 214:1307-1311. (6) Jensen, S. 1966. A new chemical hazard. New Sci. 32: 612. (7) Jensen, S., A. G. Johnels, M. Olssem, and G. Otter- lind. 1969. DDT and PCB in marine animals from Swedish Waters. Nature 224:247-250. (8) Widmark, G. 1967. Possible interference by chlorinated biphenyls. J.A.O.A.C. 50.1069. (9) Holmes. D. C, J. H. Simmons, and J. O'G. Tatton. 1967. Chlorinated hydrocarbons in British wildlife. Na- ture 216:227-229. (10) Holden, A. V. and K. Marsden. 1967. Organochlorine pesticides in seals and porpoises. Nature 216:1274-1276. Ill) Riseborough, R. W.. P. Ricche, D. B. Peakall, S. G. Herman, and M. N. Kirven. 1968. Polychlorinated bi- phenyls in the global ecosystem. Nature 220:1098-1102. (12) Koeman, J. H., M. C. Ten Noever de Braun, and R. H. de Vos. 1969. Chlorinated biphenyls in fish, mussel and birds from the River Rhine and the Netherlands Coastal Area. Nature 221:1126-1128. (13) Bagley. G. E., W. L. Reichel, and E. Cromartie. 1970. Identification of polychlorinated biphenyls in two bald eagles by combined gas-liquid chromatography — mass spectrometry. J.A.O.A.C. 53:251-261. (14) Reynolds, L. M. 1969. Polychlorinated biphenyls (PCB's) and their interference with pesticide residue analysis. Bull. Environ. Contamination Toxicol. 4:128- 143. (15) Holden, A. V. and K. Marsden. 1969. Single stage clean-up of animal tissue extracts for organochlorine residue analysis. J. Chromatogr. 44:481-492. (16) Shell Research Ltd. 1956. The use of chlorine isotopes in the interpretation of mass spectra — The analysis of polychlorinated aromatic hydrocarbons. Thornton Re- search Centre, Res. Rep. R.722. (17) Weingarten. H., D. W. Ross, J. M. Schlater, and G. Wheeler, Jr. 1962. Gas chromatographic analysis of chlorinated biphenyls. Anal. Chim. Acta 26:391-394. Vol. 4, No. 4, March 1971 175 (18) Perkin-Elmer Ltd. 1964. Analysis of chlorinated bi- (20) iJ/cfzarrf^of!, /I. 7967. The determination of organochlo- phenyls with open tubular (Golay) columns. Gas rine insecticides in animal tissues and fluids. Paper Chromatogr. Appl. Sheet No. GC-DS-011. presented at Vlth International Plant Protection Con- (19) de Faubert Maunder, M. J., A. Egan, E. W. Godly, gress, Vienna. E. W. Hammond, J. Roburn, and J. Thompson. 1964. (21) Reichel, W. L., T. G. Lamont, E. Cromartie, and L. N. Clean-up of animal fats and dairy products for the Locke. 1969. Residues in two bald eagles suspected of analysis of chlorinated pesticide residues. Analyst 89: pesticide poisoning. Bull. Environ. Contamination Toxi- 168-174. col. 4:24-30. 176 Pesticides Monitoring Journal Dieldrin and Endrin Concentrations in a Louisiana Estuary D. R. Rowe,' L. W. Canter/ P. J. Snyder,^ and J. W. Mason * ABSTRACT The general objective of this study was to determine the endrin and dieldrin concentrations in water, bottom sedi- ment, and oysters in an estiiarine area of Louisiana. Sampling was conducted on approximately a semimonthly basis from October 1968 through May 1969 in Grand Bayou, Haclcberry Bay, and Creole Bay. Creole Bay is about 14 miles above the Gidf of Mexico. Samples of oysters, sediment, and water were analyzed for residues of dieldrin and endrin using electron capture gas chromatography. Identification was made first on a non- polar column and then on a polar column. The median concentration of dieldrin and endrin present in the oysters collected from Grand Bayou. Hackhcrry Bay. and Creole Bay was 1.3 ppb and less than 1 ppb. respec- tively, while the maximum concentration was 3.4 ppb and 2.4 ppb. Water samples from all stations on every sampling date contained less than 1 ppb of both dieldrin and endrin; the highest level of dieldrin detected in the bottom sediment was 4 ppb, and the maximum concentration of endrin was less than 5 ppb. Levels in oyster samples collected in this study were com- pared to those for samples collected in the same general area between 1964-66. Introduction The use of pesticides has been a principal factor in in- creased agricultural production as well as control of insect vectors of disease. However, some pesticide resi- dues often persist in the environment and can eventually appear in water resources far from the area of initial application. In addition, aquatic organisms have the Ogden College of Science and Technology. Western Kentucky Uni- versity, Bowling Green, Ky. 42101. Department of Civil Engineering and Environmental Science, The Uni- versity oi Oklahoma, Norman. Okla. The Franklin Institute, Boston, Mass. Riverside Research Laboratories, Belle Chasse, La. ability to concentrate trace substances at levels many times greater than those present in the aqueous environ- ment. Because of their stationary habits and sensitivity to the environment, oysters are of particular concern as concentrators of pesticides. Oysters and other mol- lusks are. in fact, so uniquely efficient in concentrating chlorinated hydrocarbon pesticides that methods for using mollusks to monitor pesticide pollution have been investigated (1). Fortunately, the oyster is capable of purging itself of absorbed pesticides and other toxins when the pollutant is removed from the environment. Physiological recovery of the oyster can occur if the damage has not been too extensive (2). The general objective of this study was the determina- tion of endrin and dieldrin concentrations in the water, bottom sediment, and oysters of an estuarine area of Louisiana. Of major concern were the concentrations in oysters, specifically Crassostrea virginica, the American oyster which is sometimes referred to as the eastern oyster. As shown in the inset to Fig. 1, the study area was near Barataria Bay located some 40 miles south of New Orleans. In 1963 a major fish kill occurred in the lower Missis- sippi River with endrin identified as the causative agent. The possibility of pesticides reaching the estuarine waters was considered quite possible, especially in view of the westerly tides which may carry Mississippi River water into Barataria Bay. Because of the public health im- plications of pesticides in water and shellfish, an investi- gation was conducted in 1964-1965 by the Gulf Coast Shellfish Sanitation Research Center in cooperation with the Louisiana State Board of Health, Louisiana Wildlife and Fisheries Commission, Louisiana Stream Control Commission, and the Federal Water Pollution Control Vol. 4, No. 4, March 1971 177 Administration (3). Oysters, water, and bottom sedi- ment from oyster-growing areas west of the lower Mis- sissippi River were analyzed for dieldrin, endrin, and other chlorinated hydrocarbon pesticides. In most in- stances, chlorinated pesticides were not detected or were present in very low concentrations. The median con- centrations of endrin and dieldrin in the oyster samples were less than 10 ppb. Hammerstrom et al. (3) reported the concentrations of dieldrin and endrin detected in oysters during a 1965- 1966 study of southern Louisiana estuaries. The highest concentration of endrin reported was 70 ppb, and the median value was less than 10 ppb. The highest con- centration of dieldrin was 90 ppb, and the median value was less than 10 ppb. Bugg et al. (4) reported a 1964-66 study of pesticides in the South Atlantic and Gulf of Mexico. Oysters were collected from estuarine areas of South Carolina, Geor- gia, Florida, Mississippi, Texas, and Louisiana. Pesticide concentrations were determined by electron capture gas chromatography, and chlorinated pesticides were either not detected or were present in relatively low concen- trations. The Louisiana samples had a median concen- tration of dieldrin of 10 ppb. FIGURE \.— Study art 178 Study Area As shown in Fig. 1, the estuarine area selected for the study is approximately 40 miles south of New Orleans; sampling sites are located in Grand Bayou, Hackberry Bay, and Creole Bay. Creole Bay is about 14 miles above the Gulf of Mexico. The entire watershed above the study area eventually drains through Grand Bayou, Hackberry Bay, and Creole Bay. This watershed encompases the region be- tween Bayou Lafourche on the west and the Mississippi River on the north and east. The calculation of the amount of fresh water reaching the study area from the watershed is very complicated and, in fact, is not attempted by the U. S. Geological Survey. Their hydro- graphical studies are not applicable to areas affected by tidal backwater and interconnected drainage (Sauer, V. B., U.S. Geological Survey. Baton Rouge, La. per- sonal communication, 1969). The study area is typically estuarine in nature, with average depths of 4 to 5 feet, and stratification of salinity levels. The 1963 total marketable oyster harvest for Barataria Bay and vicinity was estimated to be 111,000 barrels (3.73 ft.Vbarrel) (5). Agricultural use of pesticides is common above Bara- taria Bay, especially along Bayou Lafourche and the Mississippi River where the ground is high and fertile. The principal crops are sugar cane, soybeans, and rice. Dieldrin, endrin, and DDT have been used in the past to control insect pests such as the sugar cane borer. Dieldrin, never extensively used, is not presently ap- plied in this area. The use of endrin has been curtailed since 1968 upon recommendation of the Louisiana State Department of Agriculture, although small amounts of endrin may have been applied in some areas during 1969. The major economic poisons in cur- rent use in the drainage area are carbaryl for sugar cane; toxaphene, DDT, and silvex for soybeans; and propanil, 2,4-D, and 2,4, 5-T for rice. Sampling Procedures and Environmental Analyses As shown in Fig. 1,13 sampling stations were selected in Grand Bayou, Hackberry Bay, and Creole Bay. The stations were divided into four groups, with Group A (Stations 1-7) being closest to the Gulf, Group B (Sta- tions 8-11) further up in the estuarine area, and Group C (Station 12) and D (Station 13) being still further upstream. Sampling was conducted 12 times on an approximately semimonthly basis from October 1968 through May 1969. At about noon on each sampling date approxi- mately 10 oysters (the average edible portion per oyster Pesticides Monitoring Joltrnal was 10 g), 50 g of bottom sediment, and 32 oz of water were collected at each station. The oysters and bottom sediment were collected using a dredge; the sediment samples were obtained at a depth of 3 to 4 inches; and surface water samples were collected by standard tech- niques. Field measurements, using instrumental analyses, were made on each sampling date at each station for the fol- lowing environmental parameters: temperature, salinity, dissolved oxygen, and pH. Temperature and salinity measurements were made with a commercial salinom- eter, dissolved oxygen with a dissolved oxygen probe, and pH with a pH meter. Tlie turbidities of the water samples were measured at the Riverside Research Lab- oratories with a turbidimeter. Analytical Procedures The oyster, sediment, and water samples were analyzed for residues of dieldrin and endrin using electron capture gas chromatography. A Micro-Tek Gas Chromatograph Model MT-220 was used, and identification was made first on a nonpoiar column (9) (DC 200 on Anakrom ABS) and then on a polar column (4% SE-30 and 6% QF-1 on Anakrom ABS). The sediment samples were ground to a fine powder, and at least a 15% moisture content was maintained in order to enhance the release of the volatile pesticides. Periodically, samples were spiked with either dieldrin or endrin; the spiked samples were then carried through the appropriate analytical procedure and percent recov- ery of the pesticides calculated. Recovery was 82% or better for the oysters, 95% or better for the water, and 50% or better for the sediment. The concentrations re- ported herein have been corrected for recovery. The chemical standards used in these analyses were ob- tained from the Pesticides Research Laboratory, USPHS, Perrine. Fla. The sample size used for GLC injection was selected to provide detection and confirmation of residues at or above 1 ppb for the oysters and water and 2 ppb for the sediment. A statistical evaluation of the number of sampling sta- tions and number of oysters per station indicated that approximately 15 stations with 10-15 oysters per sta- tion would be adequate to secure meaningful data. Results of sample analyses for pesticides were expressed in one of four ways: (1) as an identifiable concentration, (2) as a concentration less than a given value, (3) as a trace amount, or (4) as having less than the minimum detectable limit for the method. The first three ways are considered as positive responses for the presence of pesticides; however, only the first one is considered as a positive, identifiable response. The levels of pesti- cides found were reported in parts per billion (|Lig/kg) of drained weight for the oysters and on the air-dried soil weight. Residues in the water samples were reported in parts per billion (|U,g/liter). The analytical procedures used for oysters, including the extraction of dieldrin and endrin and their detection by gas chromatography, were minor modifications of methods published in the Guide to the Analysis of Pes- ticide Residues (6). The analyses of water and sediment samples were carried out in accordance with established procedures (7,8). The oyster and soil extracts were cleaned up by filtering through columns made up of successive layers of an- hydrous sodium sulfate and Florisil. The Florisil was specially prepared for use with the micro-modification of the Mills cleanup method. The columns were washed with hexane which would elute the PCB's but not the pesticide residues (9). The pesticides were then eluted from the column with diethyl ether in petroleum ether. In any event the concentrations obtained for dieldrin and endrin represent maximums since PCB's, if present, would exert a positive interference. The eluate from the columns was then concentrated by evaporation to a small volume by a stream of air at room temperature. The volume was then made up to 10 ml, and an aliquot of 5 ju.1 injected into the gas chromatograph. Environmental Data A summary of the environmental data collected during the study period is contained in Table 1. The tempera- ture, pH, and dissolved oxygen values were very sim- ilar at each of the four groups of stations. Salinity values decreased as the distance from the Gulf of Mexico increased; whereas, the turbidity increased as salinity decreased. The overall environmental conditions during the study period were conducive to satisfactory oyster growth and reproduction. TABLE 1.— Environmental data summary Mean Value Group A Group B Group C Group D Temperature (C) 17.5 17.6 17.0 17.2 pH (units) 8.0 8.0 8.1 8.1 Salinity (Voo) 15.4 12.5 10.1 8.6 Turbidity (JTU) 58 67 73 126 Dissolved Oxygen (mg/llter) (% saturation) 8.8 99 8.8 98 8.9 97 8.8 % Vol. 4, No. 4, March 1971 179 TABLE 2.- —Pesticide detection in environmental samples Pesticide Medium No. OF Sam- ples WITH Positive Con- centration No. OF Sam- ples WTTH Concentration less than a given value No. of Sam- ples WITH Trace Amounts Total No. OF Positive Samples Total No. OF Samples Percent of PosmvE Samples Dieldrin En drill Water Sediment Oysters Water Sediment Oysters 0 0 93 2 0 3 2 8 20 2 7 62 2 1 0 4 1 12 4 9 113 8 8 77 148 45 113 148 44 111 3 20 100 5 18 69 Dieldrin Concentrations A summary of the data for water, sediment, and oyster samples positive for dieldrin is contained in Table 2. A total of 148 water samples were analyzed for the presence of dieldrin during the survey. Responses of less than the minimum detectable limit for the method (1 ppb) were obtained in 144 samples (97% of the total water samples). Two samples yielded positive results of less than 0.2 ppb, and trace quantities were detected in two other samples. A total of 45 sediment samples were analyzed for the presence of dieldrin. Responses of less than the mini- mum detectable limit were obtained for 36 samples (80% of the total sediment samples). Positive results were obtained as follows: one sample contained less than 4 ppb, one sample less than 2 ppb, and six sam- ples less than 1 ppb. A trace amount of dieldrin was detected in one sample. A total of 11 3 oyster samples were analyzed for the presence of dieldrin, and all yielded positive results. Twenty samples gave positive results with levels less than a given concentration, and 93 samples exhibited positive, identifiable concentrations. Of the 20 samples, 4 contained levels of less than 2 ppb, 3 had less than 1.1 ppb, 7 had less than 1 ppb, 1 had less than 0.9 ppb, and five had less than 0.5 ppb. The mean concen- tration in the 93 samples with identifiable concentra- tions was 1.4 ppb, and the median value was 1.3 ppb. The maximum measured concentration was 3.4 ppb. The variation of dieldrin concentrations in oysters col- lected at each sampling date is shown in Table 3. In general, the concentrations were greater in the groups of stations which were further removed from the Gulf of Mexico (Groups C and D). Endrin Concentrations A summary of the data for water, sediment, and oyster samples positive for endrin is contained in Table 2. A total of 148 water samples were analyzed for the pres- ence of endrin during the survey. Responses of less than the minimum detectable limit for the method (1 ppb) TABLE 3. — Dieldrin concentrations in oysters Sampling Average Dieldrin Concentration in PPB Group A Group B Group C Group D 10-21-68 0.8 1.0 1.0 _ 11-17-68 0.8 0.7 1.7 2.2 12-1-68 • • — 12-15-68 0.8 0.4 1.5 1.1 12-29-68 1.1 1.4 — — 2-2-69 1.3 1.6 • • 2-16-69 1.9 1.7 2.2 1.7 3-2-69 1.5 1.8 1.2 3-24-69 1.9 1.8 1.1 1.5 4-14-69 2.0 2.0 2.2 2.7 5-11-69 1.1 1.1 — 1.1 NOTE: • = less than the detectable limit of 1 ppb; — = no sample. were obtained in 140 samples (95% of the total water samples). Two samples yielded identifiable levels of 0.09 and 0.2 ppb; two samples were positive with levels of less than 0.2 ppb; and trace quantities were detected in four other samples. A total of 44 sediment samples were analyzed for the presence of endrin. Responses of less than the minimum detectable limit were obtained in 36 samples (82% of the total sediment samples). Five samples were positive with levels of less than 5 ppb; and two samples had less than 4 ppb. A trace amount of endrin was detected in one sample. The sediment contained 31% sand (over 50 microns in diameter), 25% silt (2 to 50 microns in diameter), and 16% clay (less than 2 microns in diam- eter) with the balance being organic material and solu- ble compounds. A total of 11 1 oyster samples were analyzed for the presence of endrin. Responses of less than the minimum detectable limit for the method (1 ppb) were obtained in 34 samples (31% of the total oyster samples). A total of 62 samples were positive with levels of less than a given concentration: 3 samples exhibited positive, iden- tifiable concentrations: and 12 samples contained trace quantities. Of the 62 samples, 1 sample yielded a posi- tive response of less than 6 ppb, 10 samples less than 5 ppb. 5 samples less than 4 ppb, 1 sample less than 2 ppb. 32 samples less than 1 ppb, 1 sample less than 0.8 ppb, and 12 samples less than 0.5 ppb. The mean con- centration in the three samples with identifiable concea- 180 Pesticides Monitoring Journal trations was 1.8 ppb, with the lowest measured value being 0.9 ppb and the highest 2.3 ppb. Discussion and Conclusions The dieldrin and endrin concentrations in C. virginica oysters, water, and bottom sediment samples collected during 1968-1969 from a southeastern Louisiana oyster- growing estuarine area were low; however, they must be evaluated in terms of the tolerance levels in food for these pesticides. The overall mean concentrations of dieldrin and endrin in over 100 oyster samples was 1.4 ppb and less than 1 ppb, respectively; and the median values were 1.3 ppb and less than 1 ppb, respectively. Dieldrin was detected in a positive, identifiable concen- tration in 93 of 113 samples; whereas, endrin was found in like manner in only 3 of 1 1 1 samples. The maximum concentration of dieldrin and endrin present in the oysters collected from Grand Bayou, Hackberry Bay. and Creole Bay was 3.4 ppb and 2.3 ppb, respectively. Water samples from all stations on every sampling date contained less than 1 ppb of both dieldrin and endrin. The highest level of dieldrin detected in the bottom sedi- ment was less than 4 ppb. The maximum concentra- tion of endrin in this medium was less than 5 ppb; how- ever, in the case of endrin the only definite maximum value detected was 1.9 ppb. Upon comparison of pesticide concentrations found in this study with those from similar surveys conducted in 1964-1966 and 1965-1966, it seems evident that the pesticide influx into the study area has decreased since the earlier work (3,4). A comparison of the results is contained in Table 4. The 1968-1969 median dieldrin concentration is less that the 1964-1966 level by a factor of 7, and the maximum is less by a factor of 26 than the 1965-1966 level. The 1968-1969 maximum endrin concentration is 29 times less than the 1965-1966 level. In order to determine the major source of pesticides in the study area, an attempt was made to evaluate the effect of the measured environmental conditions on the level of pesticides in the oysters at any given date. Due to the lack of sufficient definite results for endrin con- centrations in the oysters, only the time variations of dieldrin concentrations were examined. Considering the recent nondetectable levels of dieldrin and endrin measured in the Mississippi River at New Or- leans, it was considered that the major influx of pesti- cides to the study area was in runoff from the drainage basin. Pesticides in the runofl' could come from soil residues or from rainout or washout. Rainfall data from two rain gauge stations located in the drainage basin were obtained for the period from October 1968 through May 1969. One station (Diamond 4N W) was located approximately 15 miles northeast of the center point of the oyster-sampling area; and the other station (Galliano) was located about 15 miles northwest of the center point. The measurement of rainfall over 7-day periods prior to the collection of samples is summarized in Table 5. TABLE 5. — Rainfall in drainage area 7-DAY RAINFALL Sampling 7-DAY Date Rainfall Diamond 4 KW Galliano (INCHES) 10-21-68 0.42 _ 0.21 11-17-68 0.50 1.30 0.90 12-1-68 1.52 2.00 1.76 12-15-68 1.15 0.49 0.82 12-29-68 0.36 0.65 0.50 1-19-69 3.18 2.50 2.84 2-2-69 0.20 0.10 0.15 2-16-69 1.75 1.33 1.54 3-2-69 0.12 0.28 0.20 3-24-69 0.63 1.42 1.02 4-14-69 3.34 1.68 2.51 5-11-69 4.77 4.48 4.62 TABLE 4. — Pesticides in oysters Residues IN PPB Survey Dieldrin Endrin Median High Median High 1964-1966 (4) 1965-1966 (J) 1968-1969 10 <10 1.3 90 3.4 <10 <1 70 2.4 The efli'ect of the 7-day rainfall on the measured envi- ronmental characteristics on a given sampling day was examined. No apparent effects on water temperature, pH, or dissolved oxygen were found. However, as would be expected, the levels of salinity and turbidity at each group of stations varied depending on the rainfall just preceding sampling. These variations are shown in Fig. 2. The decrease in pesticide concentrations found in oys- ters is due to a decrease in the amounts of chlorinated hydrocarbons applied in the selected drainage area as well as to decreasing pesticide concentrations in the nearby Mississippi River. Recent pesticide concentra- tions detected in the Mississippi River at New Orleans were generally near zero and considerably less than amounts detected previously. In general, turbidity increased and salinity decreased as the distance separating the stations and the Gulf of Mexico increased. Considering turbidity only, there was a greater variation between Group A stations (closest to Gulf) and Group D stations (farthest from the Gulf) for 7-day rainfalls between 1 and 5 inches than for rains between 0 and 1 inch. Vol. 4, No. 4, March 1971 181 If the major pesticide influx into the sampling area was from land drainage, then the dieldrin concentra- tions in the oysters should be greater following heavier rains. This was found to be true as is indicated in Table 6 and shown in Fig. 3. Pesticide concentrations in the oysters increased with increasing distance from the Gulf of Mexico, and the average levels at all groups of stations were higher after heavier rains during the 7-day period preceding sampling. If westerly tides forcing Mississippi River water into the area caused the greatest influx of dieldrin, the trends indicated in Fig. 3 should be reversed. TABLE 6. — Influence of rainfall on dieldrin in oysters FIGURE 3. — Variation of dieldrin in oysters with rainfall 7-DAY Residues of Dieldrin in PPB (INCHES) Group A Group B Group C Group D 0-1 1-5 1.0 1.4 1.2 1.3 1.4 1.8 1.6 1.8 FIGURE 2. — Variation of turbidity and salinity with rainfall 182 A Group The influence of water temperature on dieldrin uptake by oysters was examined, and the results are shown in Table 7. On December 15, 1969, the average water temperature at all stations was 8.5 C; the average diel- drin concentrations were less at all groups of stations than they were at any other temperatures. The 7-day rainfall prior to December 15 was in the 0- to 1-inch range. At about 8 C the American oyster ceases to feed and is essentially dormant (10); therefore, the low levels of dieldrin in the oysters observed on December 15, 1969, can be attributed to oyster inactivity as well as to the low level of rainfall in the 7-day period preceding sampling. TABLE 7. — Influence of temperature on dieldrin in oysters Water Number OF Days Dieldrin Concentration in PPB (C) Group A Group B Group C Group D 8.5 10-15 15-25 1 4 4 0.8 1.5 1.6 0.4 1.6 1.6 1.5 2.2 1.6 1.1 1.4 1.8 Although levels of dieldrin and endrin in the media investigated by this study are of great concern, the use of these pesticides has declined since 1965. However, at the time of this study, DDT was being utilized in this drainage area; and data for DDT could be gathered from the extractions and chromatographs from this investigation if funds were available. Peaks did appear on the chromatographs; however, efforts at that time could not be directed to analyzing this information. See Appendix for chemical names of compounds mentioned in this paper. This investigation was supported by PHS Research Grant No. 1 RO 1 U I 00346 — 01 from the National Center for Urban and Indus- trial Health. Pesticides Monitoring Journal LITERATURE CITED (1) U.S. Department of the Interior. 1965. Effects of pes- ticides on fish and wildlife — 1964 research findings of Fish and Wildlife Service. Circ. 226. 77 p. (2) Butler, P. A., A. J. Wilson, Jr., and A. J. Rick. I960. Effect of pesticides on oysters, Proc. Nat. Shellfish. Ass., 51:23-32. (3) Hammerstrom, R. J. et al. 1967. Study of pesticides in shellfish and estuarine areas of Louisiana, Public Health Service Publ. 999-UIH-2. 26 p. (4) Bugg, J. C, E. Higgins, and E. A. Robertson. 1967. Chlorinated pesticide levels in the eastern oyster (Cras- sostrea virginica) from selected areas of the South Atlantic and Gulf of Mexico. Pesticides Monit. J. 1(3):9-12. (5) Louisiana State Board of Health, 1964. Louisiana oyster water survey area III-B, Barataria Bay and vicinity. (6) Burchfield, H. P., D. E. Johnson, and E. E. Slorrs. 1965. Guide to the analysis of pesticide residues. Vol. 1, (Sec. 11: A. 3 a) U. S. Dep. of Health, Educ. and Wel- fare, Washington, D.C. (7) Warnick, S. L. and A. R. Gaufin. 1965. Determination of pesticides by electron capture gas chromatography. J. Amer. Water Works Ass.57(8): 1023. (8) Warnick, S. L., R. F. Gaufin, and A. R. Gaufin. 1966. Concentration and effects of pesticides on aquatic environments. J. Amer. Water Works Ass. 158 (5) : 601. (9) Reynolds, L. M. 1969. Polychlorobiphenyls (PCB's) and their interference with pesticide residue analysis. Bull. Environ. Contamination Toxicol., 4{3):128-143, 1969. (10) Galtsoff, Paul. 1964. The American Oyster, Crassos- trea virginica Gmelin. Fish. Bull. Fish Wildlife Serv. Vol. 64: 70, 152-lSO, 404-405, 407. Vol. 4, No. 4, March 1971 183 Monitoring Ecological Conditions Associated With Wide-Scale Applications of DMA 2,4-D to Aquatic Environments^ T. A. Wojtalik, T. F. Hall, and Larry O. HiU ABSTRACT Over 18,000 surface acres of Nickajack and GuntersviUe Reservoirs were treated with about 170,000 gallons of di- methylamine salt of 2,4-D during April — June 1969 to con- trol invading Eurasian watcrmilfoil (Myriophyllum spicatum L.). The DMA 2,4-D was applied at the rates of 20 and 40 lb of 2,4-D acid equivalent (a.e.) per acre. Representative habitat types were selected and monitored for 2,4-D con- tent in water, plankton, and sediment and for plankton species composition, distribution, abundance, change, and response. The applications of liquid DMA 2,4-D applied at a rate of either 20 or 40 lb a.e. per acre to milfoil colonics achieved excellent control within 3 to 4 weeks but did not seriously affect other submersed aquatics, and the treated water had no apparent effects on most marginal plants. No harmful or distinguishable response to the herbicide was observed in zooplankton. phytoplankton, benthic macroinvertebrates, or fish. Water from treated areas continued to be used for domestic purposes during this period without user com- plaints. Since the lower rate of application achieved good results in large block applications, it was concluded that reduced amounts of liquid 2.4-D may be used efficiently in such areas, provided that hydrolic flows and other charac- teristics are carefully considered. Dimethylamine salt of 2,4-D appears to be a noncumulative herbicide in that only small amounts are translocated along and through food chains or food webs. Plankton sorbed large amounts and retained it for extended periods. Finished drinking water from municipal treatment plants on occasion contained 2,4-D. Division of Environmental Research and Development, Tennessee Val- ley Authority, Muscle Shoals, Ala. Introduction Photogrammetric study and field inspections in the fall of 1968 showed that more than 18,000 acres of Nicka- jack and GuntersviUe Reservoirs on the Tennessee River were densely colonized by Eurasian watermilfoil (Myrio- phyllum spicatum L.) (Fig. 1). The colonies had formed in shallow water zones extending to depths as great as 16 feet below top summer pool level. Herbicidal treatment of all known surviving colonies was scheduled for the spring of 1969 after special water level manipulation produced an extra 2-foot winter drawdown. The objec- tives of the herbicidal application were: (1) to restore watermilfoil-choked areas to more desirable open water- use areas, and (2) to help prevent the vegetative spread of this nuisance macrophyte through the Tennessee River system. The herbicide used was a liquid concentrate of a di- methylamine salt (DMA 2-4-D) which contained 4 lb of 2,4-D acid equivalent (a.e.) per gallon. Helicopters ap- plied the herbicide at rates of 20 to 40 lb a.e. per acre. The applications of the liquid herbicide which is less costly and possibly more direct acting than the granular herbicide achieved excellent control within 3 to 4 weeks but did not control other submersed aquatics, and the treated water had no apparent effects on marginal plants. Spraying began on GuntersviUe Reservoir on April 1, 1969, and was terminated on May 29, 1969. Nickajack was treated June 2-5, 1969. Because of the magnitude of the watermilfoil control program, a broad monitoring project was developed on GuntersviUe Reservoir, start- ing in March before treatment and continuing with post- treatment sampling and evaluation through April 1970. 184 PESTicroEs Monitoring Journal FIGURE 1. — Main river reservoirs with heavy infestations of Eurasian watermilfoil in March 1909 A L A B A M The possible effects of the herbicidal applications on certain aquatic macrophytes, other aquatic organisms, and potentially on man were studied and evaluated. Study Areas Five study areas having extensive colonies of watermil- foil were established in Guntersville Reservoir. Four of the study areas were treated with herbicide: an embay- ment which was landlocked except for a limited flow through a road culvert; an island and shallow water channel complex on an overbank exposed to main river flow; a large embayment with limited flow-through from a tributary stream; and a slough off the channel on the overbank enclosed between channel edge islands and the shoreline. These areas were located at Jagger Branch, Ossa-win-tha, North Sauty Creek, and Comer Bridge, respectively (Fig. 2). The fifth area, Sublett Ferry Slough, which was similar to Jagger Branch, was left untreated and used mainly for observing watermilfoil colonization progression and phenological aspects. Sampling Techniques Three sampling sites designated as A, B, and C were selected in each of the four treatment areas, field- marked with styrofoam floats, and plotted on vertical aerial photographs which showed the watermilfoil in- festations. In each study area at each collection time, composite water samples were made up by stratum source using equal volumes from sites A, B, and C for each layer. Samples were placed in 1 -liter acid-washed glass bottles and then kept on ice until extracted within 24 hours (usually 8-10 hours). Water samples were collected before treatment and dur- ing the posttreatment period at approximately 1, 8, and 24 hours; at 2 and 4 weeks; and at approximately 2, 3, 6, and 12 months. These samples were analyzed for the herbicide with a Varian Aerograph chromatograph equipped with a concentric tritium foil detector after processing by the stepwise procedure given in Supple- ment A. The acid concentrations can be converted to DMA 2,4-D concentrations by multiplying the acid values by 1.2. Plankton, chlorophyll, and carbon- 14 samples were col- lected by standard techniques. Processing and data computations were by procedures commonly used by limnologists. Plankton tows below the surface were made with a Vi -meter Wisconsin-type net of No. 20 bolting cloth Vol. 4, No. 4, March 1971 185 FIGURE 2. — Location of main study areas in evaluation of DMA 2,4-D applications for control of Eurasian watennilfoil at Gunlersville Reservoir, March-October 1969 Jagger Branch— Sprayed at 40 lb 2,4-D a.e./A on April 1, 1969 Ossa-win-tha— Sprayed at 40 lb 2,4-D a.e./A on April 15, 1969 North Sauty Creek— Sprayed at 20 lb 2,4-D a.e./A on May 17, 1969 Comer Bridge — Sprayed at 40 lb 2,4-D a.e./A on May 14, 1969 Sublett Ferry — Untreated in 1969 ••♦Sampling Stations •Reference Stations and a 150-micron mesh bucket. The large volumes of plankton collected were placed in 1 -liter acid-washed bottles and stored on crushed ice. Upon arrival at the analytical laboratory 8 to 10 hours later, all except the Jagger Branch 1-, 8-, and 24-hour samples were immediately filtered through 0.3-micron glass fiber fil- ters and washed with 20-30 ml of distilled water. The filtrates and filters were subsequently analyzed sepa- rately. A set of three samples was collected with an Ekman dredge from each of the sampling sites A, B, and C within each study area. Each set of samples was later composited, providing a single composite sample for each of the three sampling sites. Immediately after col- lection, a surface 1-inch skim was removed with a plexiglass skimmer and placed in labeled double plastic bags. Plastic bags are recognized as less than ideal con- tainers, but it was felt that the bulk volume of the sample could saturate the plastic retention surfaces and still provide a "normal" residual content without inter- ference from compounds potentially added by the plas- tic. The samples were iced until arrival at the laboratory where they were placed in freezers until analyzed by procedures given in Supplement B. The material ana- lyzed for 2,4-D was subsampled, and an aliquot was dried at 105 C, weighed, and incinerated at 600 C in a muffle furnace to obtain an estimate of volatile solids and ash-free dry weight. The estimates of 2-4-D content and volatile solids or ash-free dry weight from the same samples provided data on the correlation of 2,4-D present before treatment, after treatment but before plant breakup, and after breakup and remnant accumu- lation. Before plant control a given amount of organic material and volatile solids existed in the milfoil beds. After 2,4-D treatment, as the plants responded and died, volatile solids, organics, and 2,4-D increased in the sedi- ments. Higher accumulations could be expected after the plants fragmented or lost their leaves. The efficacy of an herbicide is influenced by water tem- perature, hydrogen ion concentration, light penetration, dissolved oxygen, and alkalinity. Data were obtained on these parameters for top, middle, and lower water strata at each of the three sites of the four study areas prior to treatment and up to nine times during the 1-year evaluation following treatment. The samples were col- lected in a manner and time sequence to provide match- ing data for herbicidal concentrations and milfoil re- sponse. Temperatures were determined with a Whitney thermistor equipped with a graduated suspension line. Hydrogen ion concentrations were measured with an Orion pH meter. Light penetration was determined with a submarine photometer. Dissolved oxygen was meas- ured by the azide modification of the Winkler method. Alkalinity was measured as parts per million of cal- cium carbonate by titration of the samples with N/50 sulfuric acid to an end point of 4.8. Samples of Eurasian watermilfoil were taken from three sites of each of the four study areas during the pretreat- ment and posttreatment sampling periods. Surface and bottom strata samples were taken at each site with an Ekman dredge (0.05 m-) at selected times during the 1-year sampling cycle. At least five bottom and five surface samples were collected and inspected for water- milfoil during each sampling period. Supplementary ob- servations were also made on watermilfoil colonies at an intermediate level. Submersed aquatics in all study areas were also sam- pled with a dragchain sampling device. Thirty to sixty 186 Pesticides Monitoring Journal samples were taken at each treatment area per inspec- tion. Observations were made of the persistence of watermilfoil in the untreated Sublett Ferry Slough and in the adjacent main river. Samples of watermilfoil for biomass determinations were collected in September 1969 at the untreated Sublett Ferry area for comparison with posttreatment watermilfoil populations in the four treated study areas. Supplemental Observations In conjunction with the sampling activities, general ob- servations were also made for possible response to ex- posure by filamentous green algae, other macrophytes, macroinvertebrates, and zooplankters in the different reservoir treatment areas. Observations made on fish were of a more limited nature because of their behavioral change and mobility as the watermilfoil decomposed. Possible herbicidal toxicity to fish was investigated by confining bluegill, Lepomis macrochirus Rafinesque; redear sunfish, Lepomis microlophus Gunther; and fat- head minnows, Pimephales promelas Rafinesque in nylon and wooden cages placed in treated and untreated areas 24 hours before treatment and observed for 96 hours after treatment. Ten live specimens of bluegill and red- ear sunfish were taken from the cages at the end of the test and frozen for later analysis of residual 2,4-D in the whole fish (Supplement C). In addition to the cage test, native fish were collected before and after treatment at three locations. Four gill nets (two each, I'/a- and 2-inch bar mesh) were set for one night at each station. Selected species were re- moved, frozen, and shipped to the laboratory for 2,4-D analysis. Each month, gill netting and electro-fishing were used to collect gizzard shad, channel catfish, largemouth bass, and redear sunfish to determine if 2,4-D was increasing in the flesh of the fish in treated portions of the lower reservoir. Herbicidal treatment began on the lower end of the reservoir and proceeded upstream. Divers hand-picked mussels from their places of resi- dence in the riverbed (Fig. 3). Several individuals of a species were collected at each station in March before treatment and in June and December after all treatment ended in upstream reservoirs. It was impossible to select mussels by species and size and, thus, the mussels ob- tained for analysis did not represent the same species and the same sizes at all stations during each sampling period, nor did they represent the same species and sizes FIGURE 3. — Downstream and upstream collection areas of commercial mussels analyzed for 2,4-D content ALABAMA Vol. 4, No. 4, March 1971 187 on successive sampling dates. The meats were removed from the shells, labeled, frozen, and shipped to the laboratory for 2,4-D analysis (Supplement D). Samples of the epiphytic and benthic macroinvertebrates inhabiting watermilfoil beds and the sediments under beds were obtained with the same Ekman dredge used for soil sampling. The Ekman dredge was used to clip the milfoil at the surface to a depth of 20 cm (8 inches), again between 60 and 80 cm (24 to 32 inches), and finally from the bottom upward to 20 cm (8 inches). Of the study areas, only Jagger Branch contained sur- face-breaking watermilfoil or watermilfoil within 15 to 30 cm (6 to 12 in) of the bottom when treated. A minimum of five samples of benthic materials was ob- tained from each site. A, B, and C, producing a mini- mum total of 15 for the area. In Jagger Branch during pretreatment and the first month of posttreatment sampling. 15 epiphytic samples at both the surface and at mid-depth were also taken. The samples were washed in a 30 mesh/ inch screen in the field, preserved in alcohol, then sorted in the laboratory after a second washing in a 30-mesh screen. The macroinvertebrates were identified and enumerated, and their distribution in horizontal and vertical locations was plotted. Discussion Pretreatment monitoring was conducted during March 1969. Jagger Branch was the most enclosed of the four study areas and appeared to offer the most stringent test conditions since relatively little herbicide dilution was expected (Fig. 4). Waterborne soluble dimethyl- amine salt was monitored as it penetrated macrophyte beds following surface application and during the post- treatment period as its residual concentration decreased and disappeared. After 8 hours, residual material in the water filtrate was assumed to be that neither actively adsorbed nor absorbed by the milfoil, other aquatic life, or silt. No attempt was made to measure the passage of water masses at the sampling points. A vertical stratification of 2,4-D occurred within Jagger Branch between the time of application and the 8-hour sample. Nearly 5 mg/ liter was present at the surface while only 1.5 mg/liter occurred at the level of the root- crowns following treatment at a rate of 40 lb 2,4-D a.e. per acre. Within two weeks the 2,4-D content in water was uniformly 0.65 mg/liter and at 1 month it was 0.001 mg/liter. These concentrations eliminated surface-breaking colonies within a month after treat- ment and prevented regrowth for at least 12 months. A site adjacent to Jagger Branch was intensively moni- tored for 4 weeks. No acute toxicity was evident to any nontarget form of aquatic life while levels of 2.4-D ex- FIGURE 4. — Enclosed Jagger Branch study area showing sampling station A, B, and C, the reference station on the main reservoir, and the confined hydraulic situation for the intake of the test water treatment plant, Gunters- ville Reservoir ceeded 5 mg/liter for 5 days and 1 mg/liter for an ad- ditional 3 days. No apparent gross injury was found on unsprayed marginal plants even though their root systems were in, at, or near the waterline of herbicidal treatment areas. A strong decrease in pH (2.1 units) occurred between 1 and 14 days after treatment in watermilfoil beds which were in the early stages of breakup at Jagger Branch (Table 2); however, plankton counts showed no signifi- cant population change at any time during the study. The hydrogen ion concentration returned to its normal level approximately 1 month after treatment, when breakup was almost complete. Oxygen levels also de- creased during the period of breakup with a minimum of 5.9 mg/liter. Following watermilfoil breakup and settling of suspended matter, light penetration at 1 meter below the water surface increased from 16 to 66%. The release of nutrients from milfoil upon breakup was not fol- lowed by explosive growth of algae or other plants. The Ossa-win-tha area is associated with offshore islands adjacent to the main channel where conditions are fa- vorable for rapid herbicide dilution. Less than 0.87 mg/ liter was present within 24 hoiirs after treatment at a PESTicroEs Monitoring Journal rate of 40 lb DMA 2,4-D and there was less than 0.005 mg/liter within 14 days. Even with lower 2,4-D concen- tration and shorter exposure time, only scattered root- crowns survived at 6 months, and only a few floating fragments remained 12 months after treatment. The Comer Bridge study area is on the flooded overbank of the old channel where it receives lateral inflow from the main river and upstream inflow from a large tribu- tary. These inflows produced high dilutions of the 2,4-D application at 40 lb per acre; content in water was 0.44 mg/liter or less within 24 hours. Again, even with lim- ited persistence and low concentrations, watermilfoil control was achieved for 6 months to 1 year. The North Sauty study area was in a large lateral arm of the reservoir with downstream flow through narrow bridges. Application at a rate of 20 lb 2,4-D per acre achieved persistent water concentrations of 0.72 mg/ liter or more for 2 weeks. Living watermilfoil plants were eliminated within 1 month, and control persisted for more than 1 year into the posttreatment period even though abundant viable milfoil seed was present in the area. Chemical parameter values other than 2,4-D concentra- tions at Ossa-win-tha, Comer Bridge, and North Sauty Creek were similar to those at Jagger Branch. Oxy- gen levels were adequate at all four locations before, during, and after the period of breakup. General obser- vations gave no evidence of fish kills or acute toxicity of the herbicide to other aquatic forms at any of the study areas. Changes in macroinvertebrate populations shown in preliminary analyses appeared to result from observed emergence of maturing insects and/or loss of the watermilfoil substrate, epiphytic food, and cover for snails and some insects as the watermilfoil macrophytes died. As well as could be determined under field conditions, herbicidal concentrations in the lake water had no ap- parent control effect on most other aquatic plants. Direct observations revealed that plants survived lakewater 2,4-D concentrations at the 40 lb a.e. per acre rate, exclusive of direct spraying on their aerial portions. Plants surviving and flourishing included all common marginal woody species; common grasses, rushes, and sedges; submersed species such as coontail, crispy-leaf pondweed, and Chara; filamentous green algae such as Cladophora and Spirogyra; and common phytoplankton. One macrophyte, Justicia americana (L.) Vahl, showed a marked decline in areal extent following treatment, but even some colonies of this species persisted in treat- ment areas. The data suggest that concentration and residence time are closely relateii to water flow rates. Jagger Branch was the site with the most restricted water exchange, while North Sauty Creek had restricted water exchange but a larger watershed. Both of these embayments had higher concentrations and longer 2,4-D residence times than were found at Ossa-win-tha and Comer Bridge where high flow-through rates resulted in greater water exchange, greater dilution, and a more rapid loss from sites of application. The loss at these sites was such that less than 1.0 mg/liter remained at 24 hours. It appears that reduced amounts of liquid 2,4-D may be used eff'ectively in applications over large areas, pro- vided that careful consideration is given to the hydraulic flows and other characteristics of watermilfoil habitats. The authors believe that milfoil control may be achieved through underwater injections near the rootcrowns if the absorbed herbicide is transported upward in water- milfoil from the application zone. The heavy application was used to assure that the herbicide applied to the water surface would reach all parts of the milfoil for a duration sufficient to assure control. It is evident that relatively uniform concentrations in the water reached all levels in the watermilfoil beds especially in the en- closed embayments. In the center of the enclosed embay- ment away from sprayed weed beds the 24-hour after- treatment water concentrations of 2,4-D were uniformly 0.32 mg/liter throughout the 4-meter depth. It is also evident that treatment rates under "low-flow" conditions in the enclosed embayments were excessive even at the 20-lb per acre rate. Control in these sites was so effective that no living watermilfoil could be found for at least 1 year after treatment. In contrast, the 40-lb rate was less effective at Comer Bridge where flow-through conditions associated with the main river were accompanied by some survival of watermilfoil. As a general rule, watermilfoil control was attained more readily with the herbicide in minimal dilution embayments than in main stream high dilution areas. Other plants present in the treated areas continued to grow without the stress of competition with milfoil. None reached significant population levels. These plants included Potamogeton crispus, P. nodosus, Ceratophyl- lum demersum, Najas minor, N. guadaltipensis. and Chara. Removal of liquid DMA 2,4-D from water by plankters that adsorb or absorb the herbicide is an important form of loss of the applied material. The plankters removed approximately 24% of the herbicide within 1 hour and a proportional amount during the next 7 hours, while the 2,4-D content of raw water was increasing 19 times. Data 24 hours after application indicated that nearly 100% of the 2,4-D extracted from the raw water sam- ples was apparently bound within plankters. Filtrates from the plankton samples contained only limited amounts of herbicide the first day and relatively insig- VoL. 4, No. 4, March 1971 189 nificant amounts at 30 and 60 days. Filtrates sampled at 1 hour after treatment contained 73% of the water- borne 2,4-D but less than 25% at 8 hours and less than 10% at 24 hours. The apparent differences, as shown in Table 8, among the analyses of water, plankton, and plankton filtrates result from the fact that the plankters were collected by towing nets which concentrated and possibly damaged the dispersed plankton in the surface meter of water. When the plankton tow samples were filtered by vacuum filtration, some leakage of 2.4-D from the plankters into the filtrate may have occurred. From the viewpoint of nontarget organism accumulation of 2,4-D, the plankters readily accumulated the herbicide and retained it for more than 6 months, while the water with its small content of widely dispersed plankters lost its DMA 2,4-D load in approximately 2 weeks. Maximal herbicidal content of raw water occurred after only 8 hours. In fish taken from the treatment areas, only the grazing and filter-feeding gizzard shad and one of its predators (largemouth bass) showed a slight 2,4-D content. At no time were unusual tastes or odors reported in fish. Fishermen, boat dock operators, and Alabama conser- vation officials reported that 1969 was one of the best fishing years ever for crappie, redear sunfish, bluegill, and largemouth bass in Guntersville Reservoir. The peak of the fishing season coincided with the treatment period, and no complaints were received from fishermen that the treatment was affecting fishing. In addition to sport fishing, thousands of pounds of commercial fish, (channel catfish, smallmouth buffalo, carp, and drum) were har- vested by commercial fishermen; herbicidal treatment did not affect either their catch or market acceptance during the operation. Because of the observed uptake of 2,4-D in plankton, it was expected that filter-feeding and possibly grazing fish, macroinvertebrates, and crustaceans would concen- trate the herbicide. No large beds of commercial mussels occur in the major portions of Guntersville and Nicka- jack Reservoirs that were treated, nor does the Asiatic clam, Corhicula, yet occur in abundance in the embay- ments. However, since small Corhicula are the main food of redear sunfish, spawning redear sunfish were sampled in one study area. They were analyzed for 2,4-D content after stomach analyses confirmed Corhicula were con- sumed. As shown in Table 11, the redear did not accu- mulate DMA 2,4-D. Caged fish tests for acute toxicity were conducted in two areas but gave ambiguous results, because the fish were damaged by mechanical handling. Frequent observations revealed no significant effects upon either individual or populations of juvenile and mature fish in the study areas. Gill net catches at Stations 2 and 3 showed more fish caught per net-night after treatment (Tables 9 and 10). Commercial mussels taken from colonies located on overbank edges or channel slopes downstream from Guntersville Reservoir (Table 14) apparently accumu- lated 2,4-D with high efficiency. A trend of progressive downstream dilution of the waterborne 2,4-D was ap- parent. Mussels in beds in the path of the waterborne herbicide filtered and accumulated either the soluble 2,4-D or particles containing or having the 2,4-D ad- hering to them. During the interval of actual application, the amounts of 2.4-D accumulated were less than 1 mg/kg wet weight, and the amounts progressively de- creased downstream for about 214 miles to TRM 135, at which point a significant increase in 2,4-D was found. Subsequent investigation revealed that these colonies are downstream of the confluence of the Beech and Tennessee Rivers which drain an area where many people used 2,4-D compounds to treat stumps, brush, and weeds as part of tributary area development and new use of the Beech watershed. It appears that runoff contamination from such areas reached the Beech River, and through it entered the Tennessee River. Again a progressive dilution and decreasing accumulation trend was found from TRM 1 35 to Kentucky Dam, a distance of 114 river miles. A small local contribution of the herbicide to the area below Kentucky Dam may be occurring, but efforts to document its source have given negative results. One anomalous situation was noted in the 2,4-D values obtained from the mussels. A pretreatment mussel sam- ple collected below Guntersville Dam contained the highest 2,4-D concentration found at any time during the monitoring. Several aliquots run from the same sample indicated the same results, and simultaneous determinations of extraction coefficients for spiked mus- sel samples confirmed the high value. No significant amount of granular 2,4-D was applied by TVA to Guntersville Reservoir to control Eurasian watermilfoil or to test various formulations of 2,4-D in 1968. At present, the source of the 2,4-D in the mussel sample of March 1969 is unknown. The 2,4-D added to the waters of Nickajack and Gun- tersville Reservoirs was apparently rapidly diluted in downstream receiving waters since the mussels in colo- nies at all points below Guntersville Dam (TRM 349) showed rapidly decreasing flesh concentrations of 2,4-D. It is assumed that mussels do not selectively remove 2,4-D over and above the amounts received in the filtering-feeding system directly from the water. Mussels from each successive colony downstream appear to concentrate 2,4-D in the same proportion as received. Either increases in distance from the treatment area or increases in posttreatment time, or both, are accom- panied by lower 2,4-D levels in the mussels. 190 Pesticides Monitoring Journal ' The maximum concentration of 2,4-D in mussels after treatment approached 1 mg/kg wet weight. Most con- centrations were generally much lower during the i treatment interval and thereafter than those found in I pretreatment samples. Higher pretreatment concentra- 1 tions in downstream areas are attributed to local non- TVA additions of herbicide prior to TVA's large-scale upstream treatments. Water treatment plants, regardless of intake location. , processing, or special holding methods, apparently re- moved little DMA 2,4-D from the raw water. The 2,4-D concentrations at water intakes can be re- duced by applying DMA 2,4-D outside the water treat- ment plant buffer zone and then changing to granular applications of butoxyethanol ester (BEE) of 2.4-D within the buffer zone. The reductions can be accom- plished without change in application rate and are largely related to the ester being relatively insoluble in water, whereas salt is highly water-soluble. finable chronic toxicity; however, population changes of genera and species were apparent. These changes were apparently due to the rapid collapse of the milfoil which provided a substrate, epiphytic food in the form of diatoms (as much as 75% of the ash-free dry weight of milfoil samples), as well as cover from fish predators. Extremely large hatches of damselflies, dragonflies, may- flies, midges, and other dipterans were observed during the time of actual spraying and for 4 months after spraying. Large cladocerans, snails, some diptera, and some lepi- doptera larvae were absent 2 to 4 weeks after treatment mainly because there were no tangled stems and leaf mats to afford a suitable substrate. The same forms could be collected by plankton towing or benthic sam- pling, but only scattered individuals occurred after milfoil leaves fell from their stems. Preliminary analysis of macroinvertebrate samples from Jagger Branch indicated neither acute toxicity nor de- See Appendix for chemical names of compounds mentioned in this paper. TABLE 1. — Content of 2,4-D acid equivalent in Guntersville Reservoir water samples during 1969 Concentration in Mo/Liter Pre- treatment Posttreatment Area Hours Weeks Months 1 8 24 2 4 2 3 6 Jagger Branch > (Treated 4/1/69) Top— A-(-B+C Mid— A+B+C Low— A-(-B+C Ref 1+2 Plankton filtrate 0.023 0.008 0.005 0.006 0.25 1.2 0.22 0.014 0.22 4.8 3.1 1.4 0.20 1.8 1.5 1.1 0.003 0.23 0.63 0.67 0.66 0.077 0.001 0.001 < 0.001 < 0.001 0.028 0.023 0.020 0.013 0.042 0.036 0.023 0.009 <0.001 < 0.001 0.003 <0.001 Comer Bridge ^ (Treated 5/14/69) Top— A-l-B+C Mid— A+B+C Low— A+B+C Ref 1+2 Plankton filtrate <0.001 0.002 0.001 <0.001 0.96 0.46 0.52 0.005 0.58 0.043 0.044 0.039 0.021 0.053 <0.001 <0.001 0.002 0.002 0.024 0.021 0.015 < 0.001 0.002 0.002 0.004 0.009 <0.001 - < 0.001 <0.001 0.011 <0.001 0.002 Ossa-win-Iha ' (Treated 4/15/69) Top— A+B+C Mid— A+B+C Low— A+B+C Ret 1+2 Plankton filtrate 0.064 1.5 0.002 0.039 1.4 1.2 1.5 0.025 0.41 0.63 0.63 0.87 0.55 0.33 < 0.001 0.005 0.001 <0.001 0.17 0.001 0.001 66.0 14.0 2.0 75.6 37.8 5.3 75.6 37.8 5.3 85.0 18.8 5.6 72.5 16.2 4.7 80.0 72.8 12.1 7.0 82.1 84.0 66.0 24.0 Temperature — F Top Mid Low Reference i 53.0 53.0 53.0 59.2 56.8 55.5 55.5 66.3 59.0 55.7 60.0 60.3 57.8 55.3 56.0 61.0 61.0 60.7 62.0 68.3 67.8 67.2 65.5 79.5 77.8 75.7 77.0 93.0 89.2 86.3 86.0 Oxygen — mg/liter Top Mid Low Reference ' 9.4 9.6 9.5 13.6 12.3 11.0 10.0 13.8 12.7 12.4 10.5 12.7 11.6 11.7 10.9 6.5 6.6 5.9 8.9 6.3 6.2 6.1 6.4 8.1 7.9 7.8 8.6 8.6 8.2 7.5 8.6 9.0 9.0 9.1 8.3 Alkalinity — mg/liter Top Mid Low Reference ' 54.0 54.0 54.0 62.7 60.0 60.7 47.0 60.0 57.7 58.0 47.0 63.7 57.7 62.0 49,0 61.3 59.7 61.7 50.0 64.7 65.3 64.0 49.0 56.7 57.7 57.3 47.0 57.3 55.7 56.0 44.0 56.0 56.0 55.0 55.0 pH Top Mid Low Reference i 6.9 6.9 6.9 8.4 7.9 7.9 6.8 9.3 9.1 8.7 6.7 8.5 7.7 7.4 7.2 6.4 6.4 6.3 7.2 8.1 8.0 8.0 8.0 7.8 7.7 7.4 7.6 8.7 8.6 8.3 7.9 7.5 7.4 7.5 7.9 • All reference samples were collected from nearby untreated areas (such as shown in Fig. 4) at a depth of 1 meter or simila samples taken from treated areas. NOTE: Blanks = Not sampled. depth to mid- TABLE 3. — Abundance of watermilfoil during 1969-1970 in four study areas and in an untreated area of Guntersville Reservoir Pre- treat- ment Posttreatment Area Weeks Months 2 4 2 3 4 5 6 12 Jagger Branch (Treated April 1, 1969, full rate) Station I, A+B+C— Top Station I, A+B+C— Mid Station I, A+B+C— Low Ab Ab Ab M M M 0 0 0 0 0 0 0 0 0 ~ - 0 0 0 0 0 0 Comer Bridge (Treated May 14, 1969, full rate) Station III, A+B+C— Top Station III, A+B+C— Mid Station in, A+B+C— Low 0 0 Ab 0 0 M T 0 T T 0 T T 0 T - — 0 0 T 0 0 T Ossa-Win-Tha (Treated April 15, 1969, fuU rate) Station 11, A+B+C— Top Station II, A+B+C— Mid Station II, A+B+C— Low T T Ab 0 0 M T 0 T T 0 T - T 0 T 0 0 T - T 0 0 North Sauty (Treated May 17, 1959, half rate) Station V, A+B+C— Top Station V, A+B+C— Mid Station V, A+B+C— Low T T Ab 0 0 M — 0 0 0 — 0 0 0 0 0 0 — 0 0 0 Control Area — Sublett Ferry Station IV, A+B+C— Top Station IV, A+B+C— Mid Station IV, A+B+C— Low ^ ^ idant at all t dant at all t idant at all M M Ab Ab Ab Abut imc ' imc * imc or floating fragment found following breakup (breakup 192 Ab = Abundant — Dense colonies of healthy milfoil. M = Moderate — Thinning colonies of milfoil in early stages of breakup T = Trace — Only an occasional living healthy-appearing milfoil root c practically complete at 4 weeks in treated plots). 0 = Living milfoil not found — All inspections negative for floating fragments and/or negative for root A+B+C represents three sub-areas within a study area. Application rates DMA 2,4-D at 40 lb/A a.e. for Jagger Branch, Comer Bridge, and Ossa-win-tha areas, 20 lb/A a.e. for North Sauty. Pesticides Monitoring Journal TABLE 4. — Concentration of total 2,4-D acid equivalent in posttreatment plankton tow samples collected from Jagger Branch, Guntersville Reservoir, 1969 Sampling Location Plankton Sample (Grams) Posttreatment Sampling Time Total 2,+D a.e. (MO/KO, WET WEIGHT) I, A+B+C 3.3 1 Hour 0.06 I, A+B+C 1.7 8 Hours 0.88 I, Mid Channel-Ref > 0.87 24 Hours 0.35 I, A+B+C 0.38 24 Hours 1.8 I, A+B+C 0.97 14 Days 2.6 I, A+B+C (center of sampling site) » 0.11 30 Days 3.6 I, A-f B+C 0.09 60 Days 2.2 I, A+B+C 0.16 4 Months 1.1 I, Plankton-Ref » 0.02 6 Months <0.10 I, A+B+C 0.71 6 Months 0.37 The plankton tow was conducted in open channel at mid-embayment of Jagger Branch. "Chemical mowing" had sufficiently progressed so that plankton tow could be conducted through the center of sampling sites without obstruc- tive effects from viable milfoil. Reference sample collected in untreated area near the confluence of Honeycomb Creek with Guntersville Lake. TABLE 5. — Algal identity, abundance, chlorophyll content, and productivity in Jagger Branch, Guntersville Reservoir, during 1969 sampling Identity and AainiDANCE (No. Cells/ml) Dominant Genera Hours Weeks Months 1 8 24 2 4 2 3 Cocconeis 41 198 113 85 34 0 0 Melosira 7 7 0 160 10 352 126 Navlcula 14 61 0 58 7 17 61 Chlorella 7 99 48 412 72 525 610 Cosmarium 7 7 0 55 113 239 72 Scenedesmus 3 0 0 34 7 68 41 Tetraspora 177 273 0 829 0 0 0 Raphidiopsis 38 78 116 160 280 130 222 Synedra 0 34 14 7 0 709 304 Biflagellate 0 3 68 720 0 477 597 Mertsmopedia 0 3 0 0 0 58 392 Bluegreen filament 0 3 17 0 0 3 351 Chlorophyll A Content as Standing Stock (mg/m') for Jagger Branch and Main Stream Reference Samples Pre- TREAT- ment Hours Weeks Months 1 8 24 2 4 2 3 6 Jagger Branch Surface 2.13 2.17 3.17 2.82 8.91 7.61 6.70 6.97 — 1 meter 1.04 2 5.13 3.78 10.05 7.61 8.06 14.45 8.83 2 meters 1.04 1.09 4.74 3.17 5.87 5.92 8.70 17.67 9.27 Reference' Surface 5.17 (1) 3.83 13.31 10.66 m 4.48 (J) 2.86 1 meter 6.31 (J) (2) 18.10 8.27 5.08 4.69 10.98 3.18 2 meters 8.26 <2. '" 18.10 9.31 "' 5.26 (2) 3.43 Producivity as Carbon 14 Fixation Rate (mg C/m^/hr) for Posttreatment Samples treatment 24 Hours Weeks Months 2 4 2 3 Jagger Branch Main stream reference 16.65 201.02 14.66 15.35 127.72 37.42 29.94 3.78 4.53 ~~ > Samples at untreated main chaimel. * Broken sample. NOTE: — = Sample lost during processing. Vol. 4, No. 4, March 1971 193 TABLE 6. — Zooplankters in surface samples for pretreatment and posttreatment periods, Jagger Branch, Guntersvitle Reservoir No. PER MILLILITER OF WATER Pre- treat- Posttreatment Hours Weeks Months ment 1 8 2 4 2 3 Rotifera Monogononta Ploima Asptanchna 1 Branchjonidae 62 5 7 Keratella 2 2 2 238 359 76 Polyanthra 2 7 Hexanthra 5 Arthropoda Crustacea Cladocera Bosmina 4 5 4 61 71 5 3 Daphnia 2 Diaphanosoma 2 1 3 1 Leptodora 2 1 Copepoda Cyclopoida 2 1 2 6 5 Nauplii 1 4 44 73 NOTE: Main channel zooplankton populations were high during pretreatment and early posttreatment, then declined through the sumrher. TABLE 7. — Levels of 2,4-D and volatile solids in pretreatment and posttreatment sediment samples, Jagger Branch, Gimtersville Reservoir, 1969 Water depth Total 2,4-D a.e. Percent Number (feet) Solids I-A-Pre Unknown <0.10 6.5 I-A-Pre 6 <0.10 6.0 I-A-Pre 5 <0.10 9.3 I-A-Pre 6 <0.10 2.3 I-A-24 hours 5 <0.10 4.4 I-B-24 hours 5 0.13 5.6 I-C-24 hours 5 0.10 6.7 I-A-2 weeks 6 0.33 6.0 I-B-2 weeks 6.5 0.25 4.6 I-C-2 weeks 5 0.18 7.6 I-A-4 weeks 5 <0.IO 3.3 I-B-4 weeks 6 <0.10 4.6 I-C-4 weeks 5 0.13 6.8 I-A-2 months 7 0.30 5.6 I-B-2 months 7 0.16 6.0 I-C-2 months 7 0.45 4.2 I-A-3 months 8 <0.10 5.8 I-B-3 months 7 0.37 5.0 I-C-3 months 7 0.28 5.0 I-A-6 months 6 <0.10 4.5 I-B-6 months 6 <0.10 6.0 I-C-6 months 6 <0.10 5.0 NOTE: It was anticipated that residual 2,4-D would accumulate in the sediments by sorption and by plant fragment accumulation bringing sorbed 2,4-D to the bottom as the plants broke up. Although leaf and plant fragments accumulated their organic and volatile solids, composition was not sufficient to register an increase in sediment samples. 2,4-D did, however, accumulate for a short period. TABLE 8. — Summary of 2,4-D concentrations in surface water samples, plankton tow samples at and near surface, plank- ton filtrate, and sediment, Jagger Branch, Guntersville Reservoir, 1969 Pre- ment Posttreatment Time Hours Weeks Months 1 g 24 2 4 2 3 6 H2O (mg/liter) Plankton (mg/kg) Filtrate (mg/liter) % in plankton % in H2O Soil (mg/kg) 0.023 <0.10 0.25 0.06 .22 24.0 4.8 0.88 .20 18.3 1.8 1.8 .23 100 0.11 0.63 2.6 24.2 .25 0.001 3.6 <0.001 0.003 0.11 0.028 2.2 0.013 0.012 0.30 0.042 1.1 0.038 0.25 <0.001 0.37 <0.0OI 0.002 <0.10 ^ Insufficient amount. NOTE: Blanks = Not sampled. — = Sample lost. 194 Pesticides Monitoring Journal TABLE 9. — Pretreatment gill net catches, Guntersville Reservoir, March 26, 1969 TRM 352 Honeycomb Creek TRM 364 OSSA-WIN-THA TRM 386 Comer Bridge Num- ber Length/Range (INCHES) Wt. (LB) Num- ber Length/Range (INCHES) Wt. (LB) Num- ber Length/Range (INCHES) Wt, (LB) Sauger 1 15.6 1.5 — — — 3 14.4-16.0 3,2 Largemouth bass 1 14.5 1.9 — — — — — — White bass — — — — — — 1 9.2 0,5 Yellow bass 1 9.4 0.5 5 8.7-9.7 2.2 4 8.5-9.1 1,6 Redear sunflsh — — — 2 7.4-9.7 1.4 2 7.8-8.1 0,8 Warmouth — — — — — — 1 9.0 0,7 Green sunfish 1 7.4 0.5 — — — — — — Channel catfish 36 11.2-18,6 30.4 14 12.5-20.0 17.4 3 11.3-16,7 2,6 Blue catfish 2 12.4-15.6 2.0 1 15.8 1.3 — — — YeUow bullhead — — — 2 10.6-10.7 1.0 — — — Drum _ — — 3 10.3-13.8 2.2 _ — — White sucker — — — — — — 1 15,2 1,3 Spotted sucker — — — — — — 3 12,2-14,9 3,6 Mooneye — — — — — — 1 10,9 0.6 Golden shiner 1 10.3 0.6 — — — — — — Skipjack herring Gizzard shad 10 31 12.4-17.8 9.6-13.8 13.1 14.5 2 41 14.5-14.7 9.8-12.2 2.1 15.3 4 11 13,7-16,3 9,5-13,7 4.3 5.5 Total 84 - 65.0 70 - 42.9 34 - 24.7 Catch per net-night 21 18.3 17.5 10.7 8.5 6.2 Game fish 4 4.4 7 3.6 11 9.4 Percent game fish 4.8 6.8 10.0 8.4 32.4 38.1 Dominant species by percent [Gizzard shad, 36.9 %) 1 Gizzard shad. 58.6%) 1 1 1 (Gizzard shad, 32.4% ) 1 1 TABLE 10. — 48-hour posttreatment gill net catches, Stations 2 and 3, Guntersville Reservoir, 1969 Station 2, TRM 364, 4/18/69 Station 3, TRM 386, 5/17/69 Species Length/Range Wt. Length/Ranoe Wt. (INCHES) (INCHES) Sauger 1 14.0 0.8 — — - Largemouth bass 1 12.0 0.9 1 19.6 4.0 White bass — — — 1 12.7 0.9 Yellow bass — — — 2 9.5-10.6 1.1 White crappie 2 9.2-9.7 0.7 1 10.2 0.6 Redear sunfish 1 9.2 0.5 4 7.5-9,1 1.3 Bluegill 4 7.0-7.7 1.1 12 6,8-7,5 3.1 Warmouth — — — 1 8,8 0.4 Channel catfish 3 12.0-16.7 2.5 3 12.3-18,6 3.6 Blue catfish 1 19.8 3.0 — — — Yellow bullhead 1 11.1 0.7 2 9,6-10.5 0.8 Smallmouth buffalo 1 20.4 4.8 5 12.6-21,0 13.4 Drum _ _ — 1 10.0 0.3 Carp 1 22.8 8.3 2 12.2-12.7 3.4 Spotted sucker — — — 3 12.6-18.6 5.2 Golden shiner — — — 5 9.3-10,2 1.9 Mooneye — — — 1 11.6 0.6 Spotted gar 1 20.0 1.2 7 21.4-32.9 21.9 Skipjack — — — 1 19.4 1.7 Gizzard shad 197 9.6-12.6 79.9 70 9.9-11.1 26.6 Total 214 - 104.4 122 - 90.8 Catch per net-night 53.5 — 26.1 30.5 — 22.7 Game fish 9 — 4.0 22 — 11.4 Percent game fish 4.2 — 3.8 18.0 — 12.6 Dominant species by percent (Gizzard shad, 92.1%) (Gizzard shad, 57.4% ) Vol. 4, No. 4, March 1971 195 TABLE 11. — Number of samples and 2,4-D acid equivalent content in fish collected from Honeycomb Creek below Jagger Branch treatment area during 1969 Pretreatment POSTTHEATMENT Species 4 Weeks 2 Months 3 Months 4 Months 6 Months No. mo/kg WET WT. No. mg/kg WET WT. No. mg/ko WET WT. No. MG/KG WET WT, No. mg/kg wet WT. No. mg/kc wet WT. Gizzard shad Bluegill White crappie Largemoutii bass Channel catfish White bass Sauger Redear sunfish 5 40 13 5 <0.10 <0.10 <0.10 0.15 20 12 4 11 1 1 0.34 <0.10 <0.10 <0.10 <0.10 <0.10 6 10 8 5 <0.10 <0.10 <0.10 <0.10 6 9 6 3 7 0.22 <0.10 <0.10 <0.10 <0.10 4 7 8 <0.10 <0.10 <0.10 6 6 6 3 6 <0.10 <0.10 <0.10 <0.10 <0.10 -Concentrations of 2,4-D acid equivalent in raw and treated water from water treatment plants on Guntersville Reservoir and at Huntsville, Ala. April 1-June 26, 1969 Location Date MC/LITEK 2,4-D A.E. Percent Operating Efficiency 1969 Background Raw Finished North Marshall 4-2 0.002 (3-31) 1.9 1.3 32 (First plant itsed as 4-3 5.4 3.4 37 test facility in relation 4-4 5.9 3.7 37 to spray operations) ^5 4-5 (short intake) 5.9 5.7 4-6 5.2 5.8 4-7 3.0 (8 a.m.) . 1.8 (11 a.m.) 40 4-7 1.6 (11 a.m.) 1.6 (4:30 p.m.) 0 4-8 1.5 1.4 7 4-9 1.1 0.92 16 4-10 0.42 0.54 4-11 0.23 0.14 39 4-14 0.048 n) 4-15 0.053 (1) 4-16 0.036 n) 4-17 0.014 4-21 0.009 4-24 0.016 4-28 0.020 5-1 0.003 Guntersville No. 1 4-1 0.006 0.006 0 (Spring Creek) 4-2 0.003 0.002 33 4-3 O.0O5 0.003 40 4< 0.015 0.017 4-7 0.005 0.002 60 4-8 a) 0.084 4-9 0.008 a) 4-10 0.008 (1) 4-13 0.089 (1) 4-14 0.043 0.026 40 4-15 — (1) 4-16 0.020 4-17 0.014 4-21 0.021 4-23 0.002 4-27 0.015 5-1 <0.0OI 5-8 0.081 5-22 0.001 5-29 0.008 6-5 0.006 6-12 0.039 6-19 0.003 6-26 0.012 196 Pesticides Monitoring Journal TABLE 12.- -Concentrations of 2,4-D acid equivalent in raw and treated water from water treatment plants on Guntersville Reservoir and at Huntsville. Ala. April 1-June 26, 1969 — Continued MG/LITEIt 2,4-D A.E. Background Raw 4-8 4-9 4-10 4-12 4-13 4-14 4-15 4-16 4-16 (short intake) 4-17 4-21 4-24 4-27 4-30 5-8 4-1 4-7 4-8 4-9 4-10 4-13 4-14 4-15 4-16 4-21 4-23 4-28 5-1 5-21 5-29 4-16 4-28 5-15 5-22 5-29 6-5 6-9 6-10 6-11 6-12 6-13 6-16 6-17 6-18 6-26 5-1 5-15 5-22 5-29 6-5 0.003 0.002 0.006 0.008 0.022 0.32 0.12 0.097 0.100 0.094 0.12 0.15 0.11 0.010 0.012 0.018 0.007 0.001 0.017 < 0.001 0.004 0.002 0.005 0.007 0.100 0.11 Oi 0.082 0.097 0.11 0.097 0.092 0.030 0.018 <0.001 0.005 <0.001 0.001 <0.001 < 0.001 0.002 < 0.001 <0.001 0.046 0.002 <0.001 0.23 O.0O5 0.002 0.009 0.002 <0.001 <0..001 0.002 0.002 0.003 < 0.001 0.002 <0.001 0.18 0.002 0.010 0.013 0.20 0.074 (i) 0.089 0.12 0.076 0.001 0.002 0.007 0.019 0.041 0.057 O) O) 0.064 0.062 0.090 0.013 0.002 0.002 Vol. 4, No. 4, March 1971 197 TABLE 12.- -Concentrations of 2,4-D acid equivalent in raw and treated water from water treatment plants on Guntersville Reservoir and at Huntsville, Ala. April 1-June 26, 1969 — Continued Location Date mg/liter 2,4-D a.e. Percent Operating Efficiency 1969 Background Raw Finished Section — Continued 6-9 = 0.014 & 0,050 0.005 6-11 0.002 0.015 6-12 0.002 (1) 6-13 < 0.001 (1) 6-16 0.001 0.002 6-17 0.001 < 0.001 6-18 <0.001 < 0.001 6-26 0.002 Widows Creek 6-11 0.002 Bridgeport 6-11 0.001 0.008 6-12 0.039 0.005 6-13 - 0.054 & 0.001 esticide concentration would be 33.8 ppm of DDT and 13.0 ppm of DDD. In a second operation, the mixture was surface- applied on each plot to deliver 17.0 lb/ acre of DDT and 6.4 lb/ acre of DDD. If incorporated and mixed as before, the resulting concentrations would be an additional 8.5 and 3.2 ppm of DDT and DDD. Although this second application was not incorporated, the total average "concentration" of DDT and DDD was calcu- lated to be equivalent to 42.3 and 16.2 ppm, respectively. The two modes of application (i.e., soil incorporation and surface application) were intended to simulate a condition in which a f>esticide was surface-applied to a soil where such chemicals had already accumulated to high levels. MOISTURE REGIMES Immediately after application of the insecticides, differ- ent soil moisture regimes were imposed on the three plots: (1) continuous flooding with water to a depth of 4 inches (2) alternate flooding and draining for 7 days each — i.e., a 14-day cycle (3) no water applied other than natural rainfall — here- after referred to as the nonflooded plot. AIR SAMPLING After application of the insecticides and flooding of the first two plots, a vapor-collection apparatus described earlier (7) was activated above all three plots for con- tinuous monitoring of the atmospheric concentrations of DDT and DDD during a 6-month period. In attempt- ing to characterize concentration gradients, two booms were positioned over each plot at heights of 10 and Mention of trade names or commercial materials is for the con- venience of the reader and does not constitute any preferential en- dorsement by the U.jS. Department of Agriculture over similar prod- ucts available. 30 cm above the water or soil surface depending on the particular moisture regime. The vapor traps (1 -liter Erlenmeyer flasks) contained 500 ml of technical grade ethylene glycol that had been washed with n-hexane to remove impurities that might interfere with gas chromatographic analysis of DDT and DDD. Air was drawn through the portholes on each boom and into the trapping solvent at a rate of 1 liter/ minute. EXTRACTION AND ANALYSIS Periodically, the vapor traps were replaced with identical units and removed for extraction and analysis. The ethylene glycol was extracted with 250 ml of n-hexane for 1 hour with a magnetic stirrer, after which the trapping solvent was discarded. The hexane layer was then washed with distilled water, dried with anhydrous NaoSO^, and concentrated by evaporation to an appro- priate volume for gas chromatographic analysis. Ex- tractant volumes were adjusted so that a 5-|u,l injection contained 1 to 5 ng, which was within the linear portion of the standard curve for both insecticides. Previous laboratory studies with DDT-spiked ethylene glycol (1000 ng in 100 ml) samples indicated that this extrac- tion procedure yielded 94 to 95% recovery. Soil samples were taken from the plots for pesticide residue analysis immediately after pesticide application and at different times during the study. Forty grams of soil was extracted for three 1-hour periods with 80 ml of 8:8:1 solution of n-hexane, acetone, and saturated sodium acetate, using a wrist action shaker. Studies with spiked samples indicated that this method yielded 93% recovery with DDT. Aliquots (5 ^1/ injection) of hexane extracts from air and soil samples were assayed for p,p'-DDT and p,p'-DDD with a Micro-Tek Model 220 gas chromatograph equipped with a ^^Ni electron capture detector and Infotronics Digital Integrator Model CRS-100. Operat- ing parameters were: Column: Glass, 180 cm x 6 mm, packed with 80/90 mesh Chromport XXX (acid and base washed, silanized) coated with 3% SE-30. Temperatures: Oven 195 C Detector 295 C Inlet 215 C Carrier Gas: No (prepurifled) at 130 cc/ minute. A 3% OV-1 on Chromosorb W column was used on some of the samples. No other confirmatory tests were used. High purity analytical standards of p,p'-DDT and p,p'-DDD were supplied by the Geigy Chemical Company and the Rohm and Haas Company, respec- tively. Vol. 4, No. 4, March 1971 205 Results and Discussion Results showed little difference in either atmospheric concentrations or cumulative recovery of DDT and DDD above the flooded plot compared with the one that was alternately flooded and drained on a 14-day cycle. The best explanation is that the latter plot re- mained essentially saturated for most of the 7 days allowed for draining, and behaved primarily as a flooded system. Thus, only the results for the flooded vs non- flooded plots will be reported. Table 1 reports the levels of DDT and DDD found in the upper 6-inch layer of soil on the nonflooded plot. Based on the application rate of the pesticide mixture, the theoretical concentrations of DDT and DDD were calculated as 42.3 and 16.2 ppm, respectively. Although these values compare favorably with the levels actually detected some 6 to 8 hours after application (47.3 ppm of DDT and 12.4 ppm of DDD), we agree with Van Middelem (6) that ". . . it is very difficult to incorporate a predetermined, homogeneous mixture of a pesticide to a 6-inch depth in soil plots of the size normally em- ployed in field experiments." TABLE \. —Levels of DDT and DDD in soil (0- to 6-inch depth) of the nonflooded plot after initial application and during the experimental period Residues in PPM i Date P.p'-DDT p.p'-DDD 42.3 2 16.2 = 10/2/68 10/17/68 1 1 /22/68 1/8/69 1/27/69 47.3 29.3 25.2 29.7 23.7 12.4 11.0 10.1 11.6 14.8 1 Based on oven-dry weight of soil. - Theoretical concentration assuming lb/acre in 0- to 6-inch depth. liform soil density of 2 x 10^ During the 4 months following application of the pesti- cides, the average concentration of DDT in soil was 27.0 ppm, considerably lower than the initial level de- tected on October 2, 1968. However, the average con- centration of DDD for this period was 1 1 .9 ppm, indi- cating little change from the initial value. The exact reason(s) for these results is not clear. Based on earlier work (1,2,4,5) a rapid rate of DDT degradation would not to be expected in the nonflooded, i.e., well-aerated, plot. Moreover, if extensive degradation of DDT had occurred, a concomitant increase in DDD would prob- ably have resulted since the major degradative pathway for DDT by the soil microflora involves the reductive dechlorination to DDD (1,4,5). However, degradation to less complex polar compounds not detectable with gas chromatography by electron capture cannot be ruled out (1,4). The extent to which surface-applied DDT would undergo photochemical decomposition yielding DDD and other products is not known (3). It Is also possible that these results (Table 1) could have been due to (a) adsorption of DDT by organic and inorganic soil colloids and (b) intracellular binding of the pesticide by microorganisms, both of which could lead to incomplete extraction. Direct volatilization losses to the atmosphere, the subject of this report, could also contribute significantly to these results. Climatological data for the experimental area from October 1968 through April 1969 are summarized in Table 2. October was an extremely dry month with only 0.11 inches of rainfall reported. Moreover, an unusually dry period began during the first week in January and continued through mid-February. With these exceptions, the data are considered to be near normal. TABLE 2. — Climatological data for the experimental area from October 1968 through April 1969 Temperature (F) Rainfall (inches) Evapora- tion (inches) Average Wind Average Maximum Average Minimum DAY) October November December January February March April 82.8 68.5 62.5 62.8 62.3 62.6 77.8 56.5 44.3 38.3 42.0 43.4 43.0 58.6 0.11 6.74 4.99 1.63 6.52 4.72 9.58 5.47 2.81 2.12 2.81 2.31 3.60 5.29 44 68 70 87' 73 72 59 ' Based on measurements for 19 days during the month. Atmospheric concentrations of DDT and DDD moni- tored at 10 and 30 cm above the water or soil surface of flooded and nonflooded plots during a 6-month period are presented in Table 3. Concentrations of both pesti- cides were highest for all treatments and sampling heights during the 24-hour period following application. The higher levels of DDT compared with DDD are indicative of the relative amounts applied in the original mixture (42% DDT and 16% DDD). By the second day, concentration of DDT at the 10-cm height had dropped markedly from 1977 to 58 ng/m'' above the flooded plot, and from 2041 to 100 ng/m^ above the nonflooded plot. The corresponding levels of DDD had decreased from 405 to 30 ng/m'' and from 575 to 92 ng/m'', respectively. Why the concentration of DDD (Table 3) above the nonflooded plot reached temporary minima of 44 and 17 ng/m-* at 10 and 30 cm, 4 days after application, is unknown. A similar phenomenon for DDT occurred 2 days later, with low values of 25 and 14 ng/m^, respectively, above the nonflooded plot. Concentrations of each pesticide then increased markedly, remaining rather high for several months. It is also noteworthy that the concentration of DDD above the nonflooded plot was actually higher than DDT from 4 to 34 days 206 Pesticides Monitoring Journal after application, suggesting differences in either the relative volatility of the two pesticides under these con- ditions, or in the relative amounts of each pesticide at or near the soil surface due to photochemical or bio- logical conversion of DDT to DDD. TABLE 3. — Atmospheric concentrations of DDT and DDD monitored at 10 and 30 cm above the water or soil surface of flooded and nonflooded plots from October 2, 1968, to March 21, 1969 < Atmospheric Concentration — ng/m^ DDT DDD Date Flooded NON- FLOODED Flooded Non- flooded 10 30 10 30 10 30 10 30 CM CM CM CM CM CM CM CM 10/2/68 1 1977 651 2041 1523 405 152 575 405 10/3/68 2 58 54 100 55 30 37 92 64 10/5/68 4 — — 36 25 — — 44 17 10/7/68 6 — — 25 14 — — 70 12 10/17/68 16 14 9 93 50 10 8 102 37 1 1/4/68 34 3 3 112 41 2 3 99 41 11/22/68 52 4 5 95 72 4 2 49 36 12/20/68 80 5 1 14 10 4 2 35 27 l/J/69 94 2 2 65 56 2 1 58 34 2/13/69 135 2 2 157 118 3 2 79 32 2/27/69 149 13 8 75 62 3 1 52 17 3/21/69 172 9 5 49 5 2 1 48 2 NOTE: — equals no sample. It is evident that flooding effectively retarded the vola- tilization of the pesticides. This was noted during the first several days of the study and became more pro- nounced with time. The presence of concentration gra- dients of each pesticide throughout the study, as well as retardation of the volatilization process due to flood- ing, are illustrated in Fig. 1 and 2. The cumulative recovery of DDT (Fig. 1), after 172 days, at 10 and 30 cm above the nonflooded plot was 20,335 and 13,520 ng. Corresponding values for the flooded plot were 4,960 and 2,639 ng, respectively. Cumulative recovery of DDD (Fig. 2) at this time. 10 and 30 cm above the nonflooded plot was 15,985 and 7,090 ng, while a total of 1,520 and 1,050 ng was re- covered above the water surface of the flooded plot. Several major changes in the atmospheric concentra- tions of both pesticides above the nonflooded plot (Table 3) are apparently related to certain climatological factors (Table 1 ). These relationships are suggested by separating Table 3 into three parts (shown by the broken lines). First, following the initial maxima during the first 24 hours after application, and the aforementioned minima 4 to 6 days later, the atmospheric concentration of each pesticide remained relatively high for approxi- mately 45 to 50 days, which corresponded to a period of extremely low rainfall and a rather high rate of evaporation. Increased precipitation during the next 30 days was related to the concentration minima reached after 80 days (12/20/68). FIGURE 1. — Cumulative recovery of DDT at 10 and 30 cm above the water or soil surface of uncropped, flooded and nonflooded plots from October 1968 to April 1969 RECOVERY OF DDT ^-. SOIL ENVIRONMENT / O NONFLOODED / _ A FLOODED / SAMPLING HEISHT / ^____^., OPEN SYMBOLS: 10 CM / --' ■ CLOSED SYMBOLS: 30 CM / _ /^ / ^ - " / /• . q/ «/ y^ • ^'''^ tL 4 riME-DArS, MONTHS FIGURE 2.— Cumulative recovery of DDD at 10 and 30 cm above the water or soil surface of uncropped, flooded and nonflooded plots from October 1968 to April 1969 RECOVERY OF DDD - SOIL ENVIRONMENT ^ O NONFLOODED A FLOODED SAMPLING HEIGHT OPEN SYMBOLS: 10 CM rj^ CLOSED SYMBOLS; 30 CM y^ /° y/O /° -•- 0/ ^m-'"^ 'irf^ — '^ — t~i — * — ^ -± — S 1 TIME--DAYS, MONTHS Vol. 4, No. 4, March 1971 207 A second relationship between pesticide concentration and certain climatological factors appears during the first 40 days of 1969. The marked increase in pesticide concentration, reaching a maxima after 135 days (2/13/69), was associated with a period of unusually low rainfall and the greatest amount of wind. Finally, a decrease in the atmospheric concentration of each pesticide is shown from mid-February to the end of the experiment and is related to a period of abundant precipitation. The plot size (12 x 12 feet) used in these studies may have been somewhat small to ensure a complete mixing equilibrium for a pesticide with air at the heights where concentrations were monitored. Such an equilibrium is essential to the application of an aerodynamic approach in calculating pesticide volatilization rates under field conditions. The use of larger plots (50 x 75 feet) with this particular objective in mind will be the subject of a future report. Nevertheless, these data indicate that different climatic variables, and interactions thereof, can significantly influence the extent of volatilization of two field-applied pesticides. See Appendix for chemical names of compounds mentioned in this paper. LITERATURE CITED (1) Guenzi, W. D. and W. E. Beard. 1968. Anaerobic con- version of DDT to DDD and aerobic stability of DDT in soil. Soil Sci. Soc. Amer. Proc. 32:522-524. (2) Hill, D. W. and P. L. McCarty. 1967. Anaerobic deg- radation of selected chlorinated hydrocarbon pesticides. J. Water Pollut. Contr. Fed. 39:1259-1277. (3) Mosier, A. R., W. D. Guenzi, and L. L. Miller. 1969. Photochemical decomposition of DDT by a free radical mechanism. Science 164:1083-1085. (4) Parr. J. F., G. H. Willis, and S. Smith. 1970. Soil anaero- biosis: II. Effect of selected environments and energy sources on the degradation of DDT. Soil Sci. (in press). (5) Spencer. D. A. 1967. Problems in monitoring DDT and its metabolites in the environment. Pesticides Monit. J. l(2):54-57. (6) Van Middelem, C. H. 1969. Cooperative study on up- take of DDT, dieldrin, and endrin by peanuts, soybeans, tobacco, turnip greens, and turnip roots. General sum- mary and conclusions. Pesticides Monit. J. 3(2):100-101. (7) Willis, G. H., 1. F. Parr, R. I. Papendick, S. Smith. 1969. A system for monitoring atmospheric concentrations of field-applied pesticides. Pesticides Monit. J. 3(3):172-176. 208 Pesticides Monitoring Journal PESTICIDES IN SOIL Soil Persistence of Fungicides — Experimental Design, Sampling, Chemical Analysis, and Statistical Evaluation' W. J. Polzin, I. F. Brown, Jr., J. A. Manthey, and G. W. Probst ABSTRACT Soil persistence of a candidate foliar fungicide, parinol a,tt- bis(p-chlorophenyl)-3-pyridinemethanol, was studied under practical field conditions. Included is an experimental de- sign, method of sampling, and sampling device devised to improve the reliability of soil persistence data. Variability present in chemical analysis is considered, and a technique for statistical analysis is presented for examining persistence when repetitive applications of chemicals are made during the growing season. Employing a linear additive model, levels of compound remaining in the soil at the end of a growing season were predicted, and small differences in fungicide persistence between cultivated and uncultivated plots were detected. Introduction With the general use of pesticides in agriculture, effec- tive techniques for determining soil persistence of these compounds is of interest. Predicting soil accumulations of chemicals applied to crops during a growing season may be uncertain for a number of reasons: difficulties may be encountered in recovering the test chemical from soil on a per-application basis; sampling at intervals may lack consistency: and chemical analyses performed at different times may vary considerably. Inadequate consideration of these factors and their interactions may result in heterogeneous data so erratic as to be incon- clusive or data that appear reasonable but are misleading. The purpose of this paper is to describe and evaluate a design and sampling technique devised to improve the reliability of data on the persistence of pesticides in soil by minimizing the effect of the above factors. From Agricultural Research, Eli Lilly and Company, Greenfield, Ind. 46140. The study relates to liquid foliar fungicides. Residues from these materials accumulate beneath foliage drip lines when orchards, vineyards, and row crops are sprayed to "run-off" with high-volume applications. With either low- or high-volume sprays, these soil areas receive accumulations from the washing action of con- densate processes and rainfall. Edwards {2,3) has shown that a number of factors can affect the rate of disappearance of pesticides from soil: also, that the larger the dose of the chemical applied to the soil the less disappears in terms of percentage of the original application in a given time. In the study reported here, the gradual buildup of fungicide was simulated by repeated applications during a growing season. Cultivation for weed control serves to mix surface chemicals in the soil. Since the distribution of a chemical through the soil profile may affect its persistence (4,6,7), this factor was examined. The chemical was applied to the soil surface and intermittently incorporated in the soil by cultivation. Because our laboratory experience showed that indi- vidual soil samples varied widely, numerous subsamples were used to provide composited samples that would reliably represent the overall average for field plots. In earlier studies variability due to gas chromatographic analysis was assessed; within-day and between-day com- ponents of variation each proved to be approximately 25% of the mean. Their combined effect was estimated to result in a coefficient of variation of 40%. This sug- gested that a sufficient number of composited samples be taken at each point in time to permit scheduling analyses in a manner that would reduce the effects of these sources of variation. Vol. 4, No. 4, March 1971 209 Materials and Methods P a r i n o 1 , a,a-bis (p-chlorophenyl)-3 -pyridinemethanol (coded as EL-241), a powdery mildew fungicide (8) was selected as the test compound since, in earlier trials (Unpublished greenhouse and field tests conducted at Eli Lilly and Company. Greenfield, Ind.), this compound appeared to resist leaching and degradation in the soil. DESIGN A randomized block design was employed in which both uncultivated and cultivated plots (A and B), were repli- cated six times (Fig. 1 ). A site on level ground was selected to minimize lateral washing of the compound by rainfall. The soil type was a Brookston silty clay loam. Each plot, 2 x 66 feet, was separated by 5-foot border strips to avoid cross-contamination. To increase uniformity of applications and subsequent sampling, the plots were not cropped. The uncultivated plots were treated with trifluralin at 1.0 lb/ acre to control weeds before the initial parinol application. Two shallow, 3-inch, cultivations were made to the B plots with a power-driven rotary hoe 1 and 2 months following the initial application to determine the effect of soil mixing on persistence. FIGURE 1. — Randomized block design: each block contain- ing 2 plots, 2 X 66 feet in size; each plot separated by strips 5 feel wide (A = uncultivated plot, B = cultivated plot) 2HI- 5H h y^^/ '/'///// X^//// W///i / /**■ y /yyy'//// ////// / / yy^ /y ''//// J / // 1 // yyy/ y / ////// // 1 // X yy /y / // // / / / / / / I ^ A B A 8 A B -^ -_- — — — . BLOCK 1 BLOCK 2 Through BLOCK 6 APPLICATIONS Soil plots were surface sprayed six times at biweekly intervals beginning June 30 and extending through September 14 with 35 ppm of parinol using 4% EC -|- 1% Sponto 206 and 6% Sponto 217 as emulsifying agents, diluted with water, and applied at a rate of 125 gal/acre (16.5 g/acre). The application was equivalent to a concentration of 0.350 ppm in the top 1 inch of soil. Applications were made with a tandem of three spray nozzles set at a height of approximately 6 inches. Low pressure (15 psi) Monarch Whirl Chamber type nozzles designated as 49 x 49, 120° angle (5), were employed to further in- crease uniformity of application. To minimize spray drift, plots were treated early in the morning when air movement was minimal. Samples were collected and analyzed immediately after the initial application to establish a percent recovery value. Three samplings were made at monthly intervals during the growing season on July 30, August 30, and September 30; two additional samplings were made, during the winter and the following spring (Table 1). At each sampling time, subsamples were taken from each plot and composited into two main samples. Each composite sample consisted of 44 subsamples, weighing approximately 25 g each, taken from a uniform soil surface area to a depth of approximately 1 inch. Sub- samples were taken with the aid of a 2 x 6 foot alumi- num frame which was divided into 108 4-inch squares, each square having an alpha numeric identity. The grid was laid over one end of the plot, then sampled from and moved to successive 6-foot areas until the entire plot was sampled. Stakes placed in the plot insured identical placement of the grid for each sampling. With each placement of the frame, samples were taken from TABLE 1. — Level of compound theoretically recoverable in the cultivated and uncultivated plots at different times during the study Soil Depth (INCHES) Time of Application/Sampling 6/30 7/14 7/30 8/14 8/30 9/14 9/30 12/30 3/30 Application Number Sampling Number 1 2 3 2 4 5 3 6 4 5 6 LEVELS IN PPM A Plots — uncultivated 1 .255 .510 .765 1.020 1.275 1.530 1.530 1.530 1.530 (Theory) B Plots— cultivated (Theory) 1 2 3 .255 0 0 .510 0 0 .255 .255 .255 .765 3 .510 .255 .255 .425 .425 .425 1.275 3 .680 .425 .425 .680 .425 .425 .680 .425 .425 .680 .425 .425 210 Pesticides Monitoring Journal eight specified squares. Four of these were pooled in composite sample No. 1 and the other four in composite sample No. 2. Eleven frame placements covered the plot length and provided the 88 subsamples making up the two composite samples. All plots were sampled with a common grid pattern at a given sampling time. By employing a different pattern at successive sampling times, previously sampled soil was avoided. A single plot was sampled in less than 5 minutes; the entire trial in less than 1 hour. FIGURE 2. — Metal grid for taking subsamples, 2x6 feel; the grid consists of 108 4-inch squares (o = subsamples for composite sample No. 1, x = subsamples for composite sample No. 2) CHEMICAL ANALYSIS The test compound, parinol, was extracted from the soil samples with methanol-acetone (1:4). Extracts were analyzed by gas-liquid chromatography (sensitivity 0.005 ppm) and monitored by thin layer chromatography fol- lowing procedures described in detail by Day et al, (1). SCHEDULING OF ANALYSES To avoid confusing day-of-analysis effects with level of persistence, sample sets representing blocks were con- founded with day-of-analysis at each sampling date. Confounding was accomplished by analyzing the eight portions representing the two plots of a given block during one day — so that sets representing different blocks were analyzed on separate days. This arrangement provided an estimation of the parinol residue of each plot type by a total of 24 analyses at each sampling time (two analyses on each of two sam- ples conducted on six different days). Randomizing the order of analyzing the eight samples in each test per- mitted valid comparisons of analyses within samples, samples within plots, and plot types within the blocks. The eight samples were a convenient number for the laboratory to process in a single day. Immediately after sampling, the individual bags each containing 44 pooled subsamples were shaken and kneaded to insure complete mixing. Each composite sample was then divided into two portions for analysis (Fig. 3). EFFECT OF STORAGE ON SAMPLES After each collection, samples awaiting analysis were kept refrigerated (0 C) to minimize the effect of storage. However, to account for any sample storage effects that might occur within the time [>eriod of 6 working days required to process each set of 48, a 6 x 6 latin square of day-of-analysis sequence was superimposed on the six blocks of samples over the six sampling times (Fig. 4). FIGURE 3. — Diagram of sampling procedures Block (Replicate) Composite Samples (44 subsamples each) Division of Composite Samples (For Analysis) 1 1 Plot A 2 1 2 2 1 1 PlotB 2 1 2 2 FIGURE 4. — Latin square design employed to examine the effect of storage on samples awaiting assay (Cells contain block identity) 1 July 2 Aug. I 3 Sept. Sampling ( Dates I 4 oct. 5 Dec. , 6 Mar. Period of Sample Storage Divided Into 6 Intervals Bi B- Ba B, B5 Be Be Bs B. Bi Bs Bi B, Bi Be Bs Be B2 Bb Bs Bi B2 Be B. B2 B. Bs Be Bi Bs Ba Bo B2 Bi B> Be Vol. 4, No. 4, March 1971 211 Results RECOVERABLE THEORY Statistical analysis of analytical results from samples taken immediately after the initial application of parinol indicated that the presence of the herbicide trifluralin did not interfere with the fungicide analysis (Table 2). This finding permitted all 48 assay values, the total from both plot types, to be averaged together to estimate an amount recoverable following each application. The overall mean amounted to 73% recovery of the quantity of compound applied (0.255 ppm vs 0.350 ppm) which agreed well with the recovery efficiency of 75% found routinely with standard laboratory reference controls. Exclusion of confounding between plot types and day- of-analysis increased the credibility of the conclusion that the herbicide did not interfere with analysis. Addi- tional findings from the statistical analysis (Table 2) indicated: (a) an absence of plot by block interaction, signifying that plot type differences were measured equally in each block, (b) a significant block effect caused by either plot location or day-of-analysis or a combina- tion of the two, and (c) a significant sampling effect suggesting the two main samples obtained from each plot were of value in estimating the true level of each plot. Table 3 presents the plot means from the first sampling in the order they were analyzed. A comparison of these plot means indicates that differences between cultivated and uncultivated plots within blocks were random in nature and not related to the presence of herbicide. A comparison of block means also appears random indicating no decrease in compound levels asso- ciated with length of sample storage. EFFECT OF SAMPLE STORAGE WITHIN DIFFERENT SAMPLING DATES After completing analyses for the six sampling dates, the overall effect of the sequence of analyses within dates was examined. To facilitate this, the six sequences of analytical mean values (day 1 through day 6) are summed over the six dates in Table 4. Inspection of the pooled means indicated no loss trend associated with length of sample storage. The effect of sample storage was not detectable with the test system employed; and it, therefore, was considered as one of the components of random error in the model. Analyses of variance of the latin squares confirmed this interpretation. TABLE 2. — Analysis of variance of soil analyses made on samples taken immediately after the initial applications Source of Variation Degrees of Freedom Mean Squares F Ratio Plot (Herbicide Presence) Block Plot/Block Sample/Plot/Block Assay/Sample/PIot/Block 1 5 5 12 24 .0027 .0418 .0085 .0034 .00146 12.31 2.5 2.3 = P <.05 P <.01 TABLE 3 . — Percent of compound (applied at 0.35 ppm) found in samples taken immediately aft pr the initial application No. OF Samples Blocks in order analyzed Percent Overall Mean 1 2 3 4 5 6 Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 A B A B A B A B A B A B Plot Means Block Means 4 8 75 84 79 61 59 60 61 50 55 50 72 61 121 102 111 57 85 71 73 73 NOTE: A = uncultivated plots; B = cultivated plots. TABLE 4. — Day-of-analysis mean values, summed over all six sampling dates (the columns of the latin squares) Sequence of analysis Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Overall A Plots (ppm) .306 .338 .249 .261 .387 .307 .308 Uncultivated (%) (99) (110) (81) (85) (126) (100) (100) B Plots (ppm) .149 .139 .151 .189 .177 .143 .158 Cultivated (%) (94) (88) (96) (120) (112) (91) (100) Both Plots (ppm) .228 .239 .200 .225 .282 .225 .233 Total (%) (98) (103) (86) (97) (121) (97) (100) 212 Pesticides Monitoring Journal COMPOUND PERSISTENCE Percentages of compound theoretically recoverable at each sampling time are presented in Table 5. The base- lines for these values were obtained by computing re- coverable theories at each sampling date using the recovery values obtained from the initial sampling (0.255 ppm) as a per application constant. For the uncultivated A plots, theoretical quantities expected at each point in time are the summed increments of the recoverable portions of the applications that were made. For the cultivated B plots, recoverable theory consisted of this same additivity, plus a 1:3 soil dilution factor adjustment to account for cultivations made after the third and fifth applications (Table 1). To provide an equation that would describe changes in the amount of compound remaining over time, indi- vidual data were adjusted for cumulative increases in theory by transforming to the proportion of recoverable theory remaining at each sampling date. A log trans- formation of these proportions was found to normalize the variation, as determined by half-normal plotting of the residual variations, and to provide a linear response over time. These properties of the transformed data permitted the exponential model v = e^' + ^ to describe persistence, where y = proportion of theory remaining; / =: time in months; B ■=^ slope of persistence curve (unknown parameter); and r = replicated plots. In the regression analysis, neither block effect nor the inter- action between the covariate sampling time and blocks was significant, so the degradation model became simply The persistence slopes of the uncultivated and cultivated plot curves were, — 0.3808 and — 0.3071, respectively. These values are significantly different (P<.01) indicat- ing greater loss when the compound was surface-applied than when surface-applied and intermittently soil-incor- porated (Fig. 5). Under the trial conditions (with the effects of location, weather, etc.) the prediction equation indicates that compound residues of 0.1% or less would remain in the undisturbed plots after 18 months and the same level or less after 21 months in cultivated plots (Table 6). Thus, each plot type showed sizable degradation in less than 1 year, but small differences in rate of degradation between the cultivated and uncultivated plots were detectable. FIGURE 5. -Persistence curves of compound in soil from cullivaled and uncultivated plots 170-3 160- 150- 140-1 ^eP' 130- ° Where t = time in months 120- 110- Percentage 100- Theory gg. A Plots (uncultivated) Slope (3) = - 3808 , • Plot Range Remaining ^^ (V « 100) 70; 60- Y* ^ B Plots (cultivated twice) V^ Slope (0) = -.3071 V\ o Plot Range V 50- \N^ 40- 30- 20- 10- 0- • \ \ 3 1 2 3 4 5 6 7 8 9 10 11 12 TIME (Months) TABLE 5. — Percent of compound, theoretically recoverable, found in cultivated and uncultivated plots of Blocks I through 6 Block Number Sampling Time ( MONTHS ) Date Sampled Cumu- lative Appl. Recov. Theory (PPM) Found/Recoverable Theory X 100 — Percent Block Number A PLOTS (UNCULTIVATED) 1 0 6/30 IX .255 103 85 83 68 168 78 97 2 1 7/30 3X .765 65 85 44 52 40 46 55 3 2 8/30 5X 1.275 80 51 59 52 42 32 53 4 3 9/30 6X 1.530 21 24 19 20 19 20 21 5 6 12/30 6X 1.530 8 8 5 8 4 4 6 6 9 3/30 6X 1.530 4 5 8 5 6 5 6 B PLOTS (CULTIVATED) ' 1 0 6/30 IX .255 115 80 69 99 140 116 103 2 1 7/30 13X .255 40 56 131 34 31 27 53 3 2 8/30 15X .425 40 36 48 50 57 53 47 4 3 9/30 6X .680 28 30 34 24 17 31 27 5 6 12/30 6X .680 10 14 9 11 8 6 10 6 9 3/30 6X .680 9 10 13 11 13 9 11 ' Cultivations on 7/30 and 8/30 immediately after applications; samples taken immediately after cultivating. NOTE: Samplings were made immediately after the applications on the dates indicated. Vol. 4, No. 4, March 1971 213 Discussion PERSISTENCE MODEL The linear additive model employed in this study to measure the persistence in cultivated and uncultivated plots could be used to evaluate the effects of any num- ber of factors (F), such as moisture, soil type, cropping, fertility, etc., where the exponential model: y ^- gRt+Fi+F2+ , , , +P„+biFi+h2F2+ +b„F„+ error becomes linear when log transformed values are ana- lyzed; log y = Bt+Fi+F.,+ +b„F„ + error +F„+b,F, + b.,Fr,+ Data in this form are amenable to conventional analysis of covariance procedures. With this technique slope by factor interactions (h-^F,, bnFo. etc.,) can be examined for detection of significant factor-associated changes in persistence slopes. SPECIFIC CONSIDERATIONS Assurance that leaching was not a factor was checked by a single set of 44 core samples taken during the eighth month from one of the uncultivated plots. One- inch sections to a depth of 9 inches were composited and analyzed. Over 92% of the total compound was found in the top 1-inch layer of soil, and none was detected below 4 inches (Table 7), confirming the original premise. Where leaching is a factor in the degradation model, cores should be taken. No attempt was made to evaluate the persistence of the compound at depths below 1 inch as occurred in the case of the cultivated plots. This could have been investigated by taking cores instead of surface samples. TEST SYSTEM VARIATIONS The plot and block values found at different sampling times (Tables 3 and 5) indicate the sizable amount of variation encountered in the study. Plot differences can be attributed to a combination of (1) lack of uniformity of applications, (2) inadequacy and inconsistency in sampling, and (3) analytical variation. Because of the care exercised in applying the compound, and the ex- tensive subsampling, the largest contributor to plot variation is postulated to have resulted from analytical variability. Components-of-variance estimates indicate only 18% of the total variability encountered within a given date of sampling could be assigned to within-day analytical variation. This together with the changing pattern of differences between blocks over the six sam- pling dates (Table 5) suggests that differences observed may he attributable to a day-of-analysis effect. If plot differences between blocks are largely due to day-of-analysis effects^ several approaches could diminish this effect: TABLE 6. — Percent of compound remaining estimated from curves Predicted Levels — Percent 15 18 A PLOTS (UNCULTIVATED) B PLOTS (CULTIVATED) 1 <.l TABLE 7. — Distribution of parinol in soil profile at eighth month in the uncultivated (A) plot of Block 1 Depth (INCHES) Chromatographic Observations iig/g Nonhydrolyzed Soil ^ Acid-Hydrolyzed Soil Gas Thin layer 2 Gas Thin layer - 1 .073 -l-l-l- .08 -I-+ 2 .006 ± .03 ± 3 .004 ±L NEG ± 4 NEC 0 NEG 0 5 NEG 0 NEG 0 6 NEG 0 NEG 0 7 NEG 0 NEG 0 8 and 9 NEG 0 NEG 0 Check NEG 0 NEG 0 No treatment of soil prior to extraction, ' Relative intensity of zones observed indicated by + signs. NOTE: NEG = negative. 214 Pesticides Monitoring Journal Refinement of analytical techniques. Historically, this has taken place many times with some major increases in precision occurring in recent years. However, despite these considerable increases in analytical precision, the day effect has remained relatively constant which some- what limits the value of this approach. Confounding blocks with day-of-analysis. Components- of-variance estimates from this study indicate that single composited plot samples from eight blocks instead of six, each analyzed once, on a different day, would have required one-third fewer samplings (96 vs 144) and two- thirds fewer analyses (96 vs 288) and would have pro- vided precision equivalent to that obtained. This paradox exists, because the major source of variation was asso- ciated with the confounded block and day-of-analysis component. In retrospect, the confounding employed was sound, and a more complete commitment to it would have lead to increased precision per analysis. Blocking directly on the day-of-analysis. This could be accomplished by collecting and storing sets of persistence samples taken over time. Accumulating complete sets of samples would permit employing an analytical sched- uling design that would more effectively remove the day-of-analysis effect. It would be necessary to analyze samples taken imme- diately after the initial application, over several days, to establish a recovery per application baseline. How- ever, samples taken subsequently for persistence deter- minations could be stored to await analysis until all were collected. If portions of the soil sampled initially were stored under similar conditions with those taken later, they would be available to be analyzed concurrently to permit the effect of sample storage to be assessed. The procedure would not require that the samples be completely stable under the conditions of storage, but only that the storage conditions for all samples be uniform. The design and techniques followed in this study were more than adequate to evaluate the persistence of the model compound which degraded to less than 10% of the total applied within less than 1 year. For com- pounds which exhibit properties of greater persistence, such as the chlorinated hydrocarbon insecticides, the techniques are appropriate, but the design modifications suggested for increasing efficiency would be of value. Acknowledgments The authors thank Jack R. Miller and Steven A. Barton, who conducted or supervised various phases of the application, sampling, and analytical work; and Lealon V. Tonkinson for his interest and consultation. LITERATURE CITED (1) Day, E. W., O. D. Decker, J. R. Koons, and J. F. Hotzer. 1970. Analytical methods for fungicide a, a- bis(p- chlorophenyl)-3-pyridinemethanol (parinol). J. Ass. Offic. Anal. Chem. 53(4):747-755. (2) Edwards, C. A. 1964. Factors affecting the persistence of insecticides in soil. Soil and Fert. XXVII p. 461. (3) Edwards. C. A. 1966. Insecticide residues in soils. Resi- due Rev. 13:113. (4) Klingman. G. C. 1961. Weed control: as a science. John Wiley and Sons, Inc., New York, N.Y. p. 65. (5} Klingham, G. C. 1964. Whirl chamber nozzles com- pared to other herbicide nozzles. Weeds 12:10. (6) Nash. R. G. and E. A. Woolson. 1965. Persistence of chlorinated hydrocarbon insecticides in soils. Science 157:924-927. (7) Sheets, J. J. and L. L. Danielson, 1960. Herbicides in soils. USDA ARS-20-9, p. 172. (8) Thayer, P. L., D. H. Ford, and H. R. Hall. 1967. Bis- (p-chlorophenyl)-3-pyridinemethanol (EL-241): a new fungicide for the control of powdery mildew. Phyto- pathology 57:833, abstract. Vol. 4, No. 4, March 1971 215 PESTICIDES IN WATER Chlorinated Hydrocarbon Pesticides in Iowa Rivers' Lauren G. Johnson and Robert L. Morris ABSTRACT The routine monitoring of a number of Iowa rivers for chlorinated hydrocarbon pesticides over a 3-year period has shown the presence of dieldrin, DDT, or DDE in the ma- jority of the samples taken. Dieldrin has occurred more frequently and in higher concentrations than either of the other residues, and this is attributed to the amount of agri- cultural activity in the watersheds involved and to the amount of surface water runoff. Introduction Chlorinated hydrocarbon insecticides have been widely used in Iowa for many years. The greatest use has been by the agricultural industry although municipalities have also applied significant amounts of these chemicals. DDT and chlordane were widely used in corn insect control, but they have been largely discontinued in the past 5 years. Heptachlor is still used to some extent; but aldrin has been used more than any of the other chlorinated hydrocarbon insecticides in Iowa for a number of years. Because of this widespread use, the State Hygienic Lab- oratory has maintained a pesticide residue monitoring program for several years. Residues of various chlori- nated hydrocarbon insecticides have been detected in soil, water, fish, and wildlife samples collected through- out the State (3.5). To assess the amount of various chlorinated hydrocar- bon pesticides carried into the rivers, a number of Iowa streams have been monitored for pesticides on a monthly basis. This paper reports the results of the 1968, 1969, and 1970 river water surveys. From the State Hygienic Laboratory, University of Iowa, Iowa City, Iowa 52240. Materials and Methods Six sampling sites were included in the 1968 survey; the 1969 and 1970 surveys included these six, plus an addi- tional four. The municipal water plants at the sampling sites were supplied with clean quart jars with teflon lid liners. Samples of the water plants' raw river water intake were collected in these jars and mailed to the State Hygienic Laboratory in expanded polystyrene mailing cartons during the first week of each month. The samples were extracted as received with petroleum ether and the extracts cleaned up on a I-g silica gel column following a procedure developed at the State Hygienic Laboratory (2). Identification and quantitation were done on an F & M model 400 gas chromatograph equipped with an electron capture detector. The tem- peratures of the injector, oven, and detector, respec- tively, were 200 C, 175 C, and 200 C. A mixed column (6% QF-1 and 4% OV-1 liquid phase) was used for quantitation, and a 3% OV-1 column was used for further confirmation of the identities of the individual pesticides. Glass columns 4 feet long and 3mm i.d. were used for the analysis. A 6% :4% mixed QF-1 :OV-l liquid phase column was used for quantitation and a 3% OV-1 column was used for further confirmation of the iden- tities of the individual pesticides. Recovery tests performed by spiking distilled water with dieldrin, DDT, and DDE at concentrations ranging from 0.025 to 0.250 ;u,g/liter gave average recoveries of 92%, 102%, and 105%, respectively. Extraction efficiencies decrease as the turbidity of river water samples creases, and the values reported on turbid samples may be somewhat less than the total dieldrin content. Tur- bidities were run on the 1970 samples, as received, us- ing a Hach DC-DR colorimeter. 216 Pesticides Monitoring Journal Results and Discussion Tables 1, 2, and 3 show the pesticide content of the rivers sampled in 1968, 1969, and 1970, respectively. Dieldrin was found in 40% of 179 samples analyzed; DDT was detected in 19%; and DDE, in 14.5%. Three distinct trends can be seen from these data. The overall pesticide concentration varies from year to year and from season to season, and the levels and particular pesticides present vary from river to river. These varia- tions are related to the amount of surface runoff water entering the river, the annual spring application of in- secticides by agricultural industry, and the location of the river with reference to nearby heavy row-crop agriculture. A study was conducted in Iowa in 1967 (4) on the pesti- cide content of runoff water from fields where aldrin had been applied in the spring for corn insect control. TABLE 1. — Pesticides in Iowa rivers, 1968 Pesticide Residues in PPB (hc /liter) April Mav June July August September October Cedar River Cedar Rapids Dieldrin DDT DDE 0.004 0.012 0.010 0.005 0.004 0.010 0.012 Iowa River Iowa City Dieldrin DDT DDE 0.006 0.004 0.008 0.012 0.006 0.007 Mississippi River Davenport Dieldrin DDT DDE 0.003 0.004 0.004 Mississippi River Dubuque Dieldrin DDT DDE 0.003 0.005 0.002 0.005 Missouri River Council Bluffs Dieldrin DDT DDE 0.002 0.004 Raccoon River Des Moines Dieldrin DDT DDE 0.003 0.006 0.005 0.004 0.0O4 0.005 0.004 0.005 TABLE 2. — Pesticides in Iowa rivers, 1969 Pesticide Residues in PPB (/io/liter) April May June July August September October November Cedar River Cedar Rapids Dieldrin DDT DDE 0.009 0.006 0.007 0.012 0.009 0.003 Iowa River Iowa City Dieldrin DDT DDE 0.007 0.004 0.051 0.047 0.022 0.014 0.015 0.005 Little Sioux River Cherokee Dieldrin DDT DDE 0.005 0.007 0.005 0.016 0.009 0.011 0.003 Mississippi River Davenport Dieldrin DDT DDE 0.006 0.004 0.011 0.006 Mississippi River Dubuque Dieldrin DDT DDE 0.004 0.005 0.010 0.004 Missouri River Council Bluffs Dieldrin DDT DDE 0.006 0.014 Nislmabotna River ^ Hamburg Dieldrin DDT DDE 0.004 0.014 0.021 0.063 0,054 0.039 0.040 Raccoon River Des Moines Dieldrin DDT DDE No Sample Received 0.007 0.009 0.007 0.015 0.041 0.006 0.010 Skunk River Oskaloosa Dieldrin DDT DDE No Sample Received 0.010 0.027 0.029 0.019 0.019 0.006 Upper Iowa River Decorah Dieldrin DDT DDE 0.007 0.006 Residues of heptachlor (0.600 ppb) found in samples taken in September. Vol. 4, No. 4, March 1971 217 TABLE 3. — Pesticides in Iowa rivers, 1970 ■■ turbidities in Jackson turbidity units determined on Hach DC-DR colorimeter] Pesticide Residues in PPB (ag/liter) March April May June JLILY August Cedar River Cedar Rapids Dieldrin DDT DDE 0.013 (35) (17) (50) 0.006 0.019 (80) 0.015 (55) (45) 0.009 Iowa River Iowa City Dieldrin DDT DDE 0.016 (60) 0.008 (70) 0.012 (90) 0,036 (75) 0.030 (75) 0.021 (60) Little Sioux River Cherokee Dieldrin DDT DDE (85) 0.009 (500) 0.020 (85) 0.009 (225) 0.013 (105) 0.011 (80) 0.006 Mississippi River Davenport Dieldrin DDT DDE (15) (65) 0.004 (100) 0.006 0.012 (300) 0.010 (75) (30) 0.009 Mississippi River Dubuque Dieldrin DDT DDE (50) (50) (90) (70) (30) Missouri River Council Bluffs Dieldrin DDT DDE (75) (125) 0.023 (70) 0.004 (90) 0.008 (65) 0.005 (40) Nishnabotna River Hamburg Dieldrin DDT DDE 0.065 (500) 0.014 0.012 (105) (30) 0.017 0.058 (150) No Sample 0.023 (60) 0.012 Raccoon River Des Moines Dieldrin DDT DDE (190) 0.023 (70) 0.012 (350) 0.019 0.011 (180) 0.013 (55) 0.020 (500) Skunk River Oskaloosa Dieldrin DDT DDE 0.005 (95) 0.015 (350) 0.008 (42) 0.021 (115) 0.019 (40) 0.024 (280) Upper Iowa River Decorah Dieldrin DDT DDE (15) (5) (12) 0.006 (300) (105) (25) 0.007 Samples of surface water draining from tiie fields were taken immediately after a heavy rain, which was 1 month after the application of aldrin on these fields. The principal residue found was dieldrin, and in these highly turbid samples about 50% of the total dieldrin was adsorbed on the soil particles which settled in the sample bottles during a 24-hour holding period. The dieldrin content in the water draining directly from the fields was 10 to 20 times that found in the Iowa River. Because a significant portion of the pesticide residues was carried by the suspended matter in the water, tur- bidity was also recorded for the 1970 samples. It is anticipated that as sufficient data are accumulated, a seasonal correlation will be developed between turbidity and the pesticide content of the individual rivers. Table 4 shows the flow in the Iowa River for 1968, 1969, and 1970. These are the discharge rates of the river at Marengo, Iowa, as measured by the U.S. Geo- logical Survey (7, 8 and 9). Marengo was chosen as the best location for an index of the variation of flow in the Iowa River, because it is only 40 miles upstream from the Iowa City sampling station. The table shows that in 1969 the flow in the Iowa River averaged 5 to 10 times more than in 1968, with the 1970 values aver- aging somewhat less than those of 1969. The concen- tration of dieldrin in Tables 1, 2, and 3 follow this same pattern indicating that as the surface runoff into the river increases significantly, so does the dieldrin level. The dieldrin levels also follow a strong seasonal pattern. Both 1969 and 1970 show large increases in the dieldrin concentration in June and July followed by decreasing concentration in later months. Aldrin is normally applied for corn insect control dur- ing May in Iowa and is rapidly converted to dieldrin in the environment. This seasonal usage coupled with high surface water runoff apparently accounts for the high concentration of dieldrin found in the Iowa River in June and July of 1969 and 1970. The levels found in the Mississippi River and the Mis- souri River are similar to those reported by Breidenbach et al. (1) in their survey of major river basins for 1957 to 1965. However, those smaller internal rivers in Iowa which drain highly cultivated portions of the State show significantly higher levels of dieldrin, especially during years of high surface water runoff. Those internal rivers in Iowa which do not drain highly cultivated areas, such as the Upper Iowa River in north- eastern Iowa, consistently have low pesticide concen- trations. Monitoring the chlorinated hydrocarbon pesticides in a selected group of Iowa rivers over a 3 -year period has shown several consistent trends which directly relate the dieldrin concentration in the rivers to the agricultural application of aldrin. 218 Pesticides Monitoring Journal TABLE 4. — Mean monthly discharge of the Iowa River (Gaged at Marengo, Iowa) Mean Discharge in C.F.S. 1968 1969 1970 March 358 5,217 2.917 April 920 4,643 1,563 May 588 2,725 4,238 June 460 4,469 1,180 July 1,286 11,340 August 859 2,453 September 295 783 The significant increase of dieldrin in the Iowa River from 1968 to 1969 and 1970 is a result of the in- creased surface water runoff noted in Table 4. The significant increase in the summer months follows the spring application of aldrin for corn insect control. These variations are not seen in rivers such as the Upper Iowa which drains an essentially nonagricultural part of the State. The Report of the National Technical Advisory Com- mittee on Water Quality Criteria (6) lists the permissible level for dieldrin in public water supplies as 0.017 mg/ liter. It also states in the criteria for fresh-water organisms that, since any addition of a persistent chlori- nated hydrocarbon insecticide is likely to result in per- manent damage to aquatic populations, their use should be avoided. The levels of pesticides in Iowa rivers are well below the listed permissible criteria for public water supplies. However, the agricultural use of aldrin has resulted in dieldrin concentration levels alleged to be significant to aquatic life in some of Iowa's rivers through the mechanism of surface runoff water from heavily cultivated areas. Research already completed and being prepared for pub- lication shows dieldrin concentrations several times greater than the Food and Drug Administration action guideline in the edible portion of catfish taken from rivers listed in this report. LITERATURE CITED (1) Breidenbach, A. W., C. G. Gunnerson, F. W. Kawahara, J. J. Lichtenberg, and R. S. Green. 1967. Chlorinated hydrocarbon pesticides in major river basins, 1957-65. Public Health Rep., U. S. Dep. Health. Educ, Welfare, 82(2):139-156. (2) Johnson. L. G. 1970. Analysis of pesticides in water using silica gel column clean-up. Bull. Environ. Con- tamination Toxicol, (in press). (3) Johnson, L. G., H. Harrison, and R. L. Morris. 1970. Preliminary study of pesticides levels in the eggs of Iowa pheasants, blue wing teal and coots. Bull. Environ. Con- tamination Toxicol, (in press). (4) Morris, R. L. and L. G. Johnson. Pollution problems in Iowa. In: Water resources of Iowa, eidted by P. J. Horick, 1970, available from the Iowa Academy of Science, Univ. of Northern Iowa, Cedar Falls, Iowa, p. 89-109. (5) Morris, R. L. and L. G. Johnson. 1969. Some aspects of pesticides in the Iowa environment. Rep. No. 70-10, State Hygienic Laboratory, Univ. of Iowa, Iowa City, Iowa. (6) National Technical Advisory Committee to the Secretary of the Interior. 1968. Water quality criteria. Fed. Water PoUut. Contr. Admin., Washington, D. C, p. 20-63. (7) U. S. Geological Survey. 1967. Water Resources Data for Iowa, p. 35-40. (S; Ibid. 1969. p. 48-53. {9) Ibid. 1970. (in press). Vol. 4, No. 4, March 1971 219 APPENDIX Chemical Names of Compounds Mentioned in This Issue ALDRIN BHC CARBARYL DDD (TDE) DDE DDT (including its isomers and dehydrochlorination products) DIELDRIN ENDRIN HEPTACHLOR HEPTACHLOR EPOXIDE LINDANE TOXAPHENE PROPANIL SILVEX 2.4-D BEE 2,4-D DMA 2,4-D 2,4.5-T TRIFLURALIN Not less than 95% of 1.2,3,4,I0,10-hexachloro-l,4,4a,5,8,8a-hexahydro-l,4-pnrfo-«o-5,8-dimethanonaphthaIcne 1,2,3,4,5,6-hexachlorocyclohexane, mixed isomers ]-naphthyl methylcarbamate l,l-dicliloro-2,2-bis(p-chlorophenyl)ethane; technical DDD contains some o.p'-isomer also. l,I-dichloro-2,2-bis(p-chlorophenyl) ethylene l,l,l-trichloro-2,2-bis(p-chlorophenyl)ethane; technical DDT consists of a mixture of the p,p'-isomer and thi o.p'-isomer (in a ratio of about 3 or 4 to 1 ) Not less than 85% of 1,2,3,4,10. 10-hexachloro-6.7-cpoxy-l,4,4a.5,6,7,8,8a-octahydro-l,4-eHdo-Mo-5,8-dimethano= naphthalene l,2,3,4,10,10-hexachloro-6,7-epoxy-I,4,4a,5,6,7,8,8a-octahydro-l,4-enrfo-en(/o-5.8-dimethanonaphthalene 1, 4,5,6,7,8, 8-heptachloro-3a,4,7,7a-tetrahydro-4,7-methanoindene l,4,5,6,7,8,8-heptachloro-2,3-epoxy-3a,4,7,7a-tetrahydro-4.7-methanoindan 1,2,3,4,5,6-hexachlorocyclohexane, 99% or more gamma isomer chlorinated camphene containing 67-69% chlorine A^-(3,4-dichlorophenyl)propionamide 2-( 2,4,5-trichlorophenoxy ) propionic acid 2,4-dichlorophenoxyacetic acid 1 -, butoxyethanol ester -, dimethylamine salt 2,4,5-trichlorophenoxyacetic acid a,a,Q:-trifluoro-2,6-dinitro-/V,N-dipropyl-p-toluidine 220 Pesticides Monitoring Journal Acknowledgment The Editorial Advisory Board wishes to acknowledge with sincere appreciation the efforts of the following persons who assisted in reviewing papers submitted for publication in Volume 4, Nos. 1-4, of the Pesticides Monitoring Journal: Judith A. Armour, Food and Drug Administration William F. Barthel, Public Health Service Jerry Burke, Food and Drug Administration Kenneth R. Hill, Agricultural Research Service Philip C. Kearney, Agricultural Research Service Alfred K. Klein. Food and Drug Administration Thair G. Lamont, Fish and Wildlife Service Granville Q. Lipscomb, Food and Drug Administration Bernadette Malone, Food and Drug Administration Mildred Porter, Food and Drug Administration William Reichel, Fish and Wildlife Service William T. Sayers, Environmental Protection Agency Sidney Williams, Food and Drug Administration A. J. Wilson, Jr., Environmental Protection Agency George Yip, Food and Drug Administration Vol. 4, No. 4, March 1971 221 Information for Contributors The Pesticides Monitoring Journal welcomes from all sources qualified data and interpretive information which contribute to the understanding and evaluation of pesticides and their residues in relation to man and his environment. The publication is distributed principally to scientists and technicians associated with pesticide monitoring, research, and other programs concerned with the fate of pesticides following their application. Additional circulation is maintained for persons with related in- terests, notably those in the agricultural, chemical manu- facturing, and food processing industries; medical and public health workers; and conservationists. Authors are responsible for the accuracy and validity of their data and interpretations, including tables, charts, and refer- ences. Accuracy, reliability, and limitations of the sampling and analytical methods employed must be clearly demonstrated through the use of appropriate procedures, such as recovery experiments at appropriate levels, confirmatory tests, internal standards, and inter- laboratory checks. The procedure employed should be referenced or outlined in brief form, and crucial points or modifications should be noted. Check or control samples should be employed where possible, and the sensitivity of the method should be given, particularly when very low levels of pesticides are being reported. Specific note should be made regarding correction of data for percent recoveries. Preparation of manuscripts should be in con- formance to the Style Manual for Biological Journals, American Institute of Biological Sciences, Washington, D. C, and/or the Style Manual of the United States Government Print- ing OflRce. An abstract (not to exceed 200 words) should accompany each manuscript submitted. All material should be submitted in duplicate (original and one carbon) and sent by first-class mail in flat form — not folded or rolled. Manuscripts should be typed on 8V2 x 11 inch paper with generous margins on all sides, and each page should end with a completed para- graph. All copy, including tables and references, should be double spaced, and all pages should be num- bered. The first page of the manuscript must contain authors' full names listed under the title, with affiliations, and addresses footnoted below. Charts, illustrations, and tables, properly titled, should be appended at the end of the article with a notation in text to show where they should be inserted. Charts should be drawn so the numbers and texts will be legible when considerably reduced for publication. All drawings should be done in black ink on plain white paper. Photographs should be made on glossy paper. Details should be clear, but size is not important. The "number system" should be used for litera- ture citations in the text. List references alpha- betically, giving name of author/s/, year, full title of article, exact name of periodical, volume, and inclusive pages. The Journal also welcomes "brief" papers reporting monitoring data of a preliminary nature or studies of limited scope. A section entitled Briefs will be included, as necessary, to provide space for papers of this type to present timely and informative data. These papers must be limited in length to two Journal pages (850 words) and should conform to the format for regular papers accepted by the Journal. Pesticides ordinarily should be identified by common or generic names approved by national scientific so- cieties. The first reference to a particular pesticide should be followed by the chemical or scientific name in parentheses — assigned in accordance with Chemical Abstracts nomenclature. Structural chemical formulas should be used when appropriate. Published data and information require prior approval by the Editorial Advisory Board; however, endorsement of published in- formation by any specific Federal agency is not intended or to be implied. Authors of accepted manuscripts will receive edited typescripts for approval before type is set. After publication, senior authors will be provided with 1 00 reprints. Manuscripts are received and reviewed with the under- standing that they previously have not been accepted for technical publication elsewhere. If a paper has been given or is intended for presentation at a meeting, or if a significant portion of its contents has been published or submitted for publication elsewhere, notation of such should be provided. Correspondence on editorial matters or circulation mat- ters relating to official subscriptions should be addressed to: Mrs. Sylvia P. O'Rear. Editorial Manager, Pesti- cides Monitoring Journal, Division of Pesticide Com- munity Studies, Pesticides OflRce, Environmental Protec- tion Agency, 4770 Buford Highway, BIdg. 29. Cham- blee, Ga, 30341. 222 Pesticides Monitoring Journal BOSTON PUBLIC LIBRARY 3 9999 05571 176 4