Document 2JqrxJB8GJmy4rqYRa4eojmja

CHLORINATED HYDROCARBONS*, THEIR DYNAMICS AND EGGSHELL EFFECTS ON HEREIN! GULLS AND OTHER SPECIES BY . DANIEL W. ANDERSON A thesis submitted as a portion of the requirements for the degree of DOCTOR OF PHILOSOPHY (Wildlife Ecology and Zoology) at the University of Wisconsin 1970 HARTOLDMONOQ29631 To Professors: Hickey McCabe Eralen This thesis having been approved in respect to form and mechanical execution is referred to you for Judgment upon its substantial merit. >?9. Dean Approved as satisfying in substance the dootoral thesis requirement of the University of Wisconsin, ( Major Professor Date of Examination, HARTOLDMON0029632 University of Wisconsin Library Manuscript Theses Unpublished theses submitted for the Master's and Doctor's degrees and deposited in the Univer sity of Wisconsin Library are open for inspection, but are to be used only with due regard to the rights of the authors. Bibliographical references may be noted, but passages may be copied only with the per mission of the authors, and proper credit must be given in subsequent written or published work. Ex tensive copying or publication of the thesis in whole or in part requires also the consent of the Dean of the Graduate School of the University of Wisconsin. DANIIL WILLIAM ANDERSON ' This thesis by------r----------- --------- ------ --------------------- --------------------------- ----------------- ------- -------- has been used by the following persons, whose signatures attest their acceptance of the above restrictions. A Library which borrows this thesis for use by its patrons is expected to secure the signature of each user. NAME AND ADDRESS DATE HARTOLDMON0029633 CHLORINATED HYDROCARBONS: THEIR DYNAMICS AND EGGSHELL EFFECTS IN HERRING GULLS AND OTHER SPECIES A thesis submitted to the Graduate School of the' University of Wisconsin in partial fulfillment of the requirements for the degree of Doctor of Philosophy. by Daniel William Anderson Degree to be awarded January 19-- June 19-- ( Ji HARTOLDMON0029634 AsAt J) /#V PREFACE VI The research reported in this thesis involved diverse* but related topics- The rapid changing nature of pesticide research---or now more accurately because of, the recent discovery of PCB's, chlorinated-hydro carbon research--has required that new hypotheses be developed and tested during the course of studies already in progress. My advisor, Joseph J. Hickey, has had a constantly fresh approach to research topics, providing new ideas as well as allowing me free reign to explore my own ideas. Necessarily, reference to the research reported here can only be honestly- reported in terms of plural authorships, I have therefore included within this thesis all coauthors in the belief that today's research must essentially be a teamwork effortj it would be naive to think otherwise. The first topic to be presented, kinetics of chlorinated hydrocarbons in Lake Michigan Herring Gulls, was developed at an informal meeting in Dr, Hickey's hotel room during the 1965 North American Wildlife and Natural Resources Conference. There, cofle&gnes from the U.S. Fish and Wildlife Service, Audubon Society, and University of Wisconsin discussed various aspects of wildlife-pesticide problems. J, J. Hickey and J. A. Keith had already demonstrated that Lake Michigan Herring Gulls were badly contaminated with metabolites of DDT, and Keith, under Hickey's supervision,, .demonstrated adverse effects of such residues on the reproduction of the Green Bay gulls. The basic hypothesis we tested was whether the very high residues observed in wild adults could be attained within one year in relatively lightly contaminated juveniles that had been fed Alewives, their main food-source in the wild. Residue concentrations in whole , Alewives were about 1000 times less than those found in the fat of wild adults. HARTOLDMON0029635 vii j We continued to monitor Lake Michigan Herring Gulls and. Alewives | in the years following our feeding experiment. A major research problem ' ,' developed after the completion of the feeding experiment: the interpret- ation of chromatograms for insecticide residues in the presence of PCB1s. ' This required nearly complete reevaluation of our residue data and it | delayed our final results for about two years. Satisfactory analytical | - ' -i | J | I i ' I?j ' . techniques for the precise quantification of PGB's are only now (1970) beginning to emerge. . . . A major discovery in the field of pesticide ecology, that of D. A. , Ratcliffe in 1967 in Great Britain, changed the direction of our research toward eggshells and calcium deposition. Ratcliffe, by examining museum ' specimens, found a significant change in the eggshell thicknesses of three declining species of British raptors, generally beginning in. I9I46 . ' or 191*7* J. A. Keith had observed a high frequency of broken eggs on . Green Bay gulleries.in 1961* and we saw the same phenomenon in later years. '` . ' ' 'Related phenomena--egg eating, egg breakage, and egg disappearance--had all been reported for the Peregrine Falcon, the species which initially received the most attention because it had exhibited the most alarming recent avian population crash in this century. Many of the significant ecological events since 19l*7, when DDT had achieved general, public use, began to come together in a Madison conference convened by J. J. Hickey in 1965'. I feel strongly that the "Peregrine Conference" has been a major stimulus to many of the significant findings in pesticide ecology for the past five years. , The feeding experiment completed in 1967, we followed up Ratcliffe's _ study in North America and examined avian eggshells in a large portion of ' HARTOLDMON0029636 viii Herring Gull, %e eggshell study was necessary to aid in determining some of the possible mechanisms of reproductive failures in Lake Michigan Herring Gulls as they might fit into other interactions observed in the wild. We had originally planned to study the eggshells of 11 raptorial species, 6 fish-eaters, and 2 species believed to be stationary in numbers (to serve as controls). The list was extended to 25 species to include other species already under study by colb ague in other parts of the country. We tested the hypothesis that eggshells had become thinner since the "era of the chlorinated hydrocarbons11, the decade of the 19h0s and later. Because egg collecting had died out in the late 1930s, our original plan was to describe normal geographic variation in selected species to give future workers bases of comparison with present-day' eggshells. Unexpectedly, we found many active collectors still in operation and our .eggshell study was expanded from an original proposal of four months and 2k museums to a year and a half and some 90 separate collections.- We needed more time to search out recent specimens once we knew they were available. Such specimens provided us with adequate materials to at least partially describe the presence or lack of eggshell changes in some of the 25 species we had planned to study. The greatest majority of egg collectors we contacted proved to be extremely helpful and cooperative. This thesis is, then, divided into three major parts along with an appendix of miscellaneous data and publications relating to our research. Each section is written in the appropriate journal style and manuscript pages are shown on the lower right-hand corners of each thesis page. . HARTOLDMON0029637 ix The parts are as follows: (l) basic pesticide dynamics in Lake Michigan Herring Gulls, to be submitted to The Journal of Applied Ecology, (2) studies of normal and abnormal variation in the eggshells of North American Herring Gulls, to be submitted to The Auk, and (3) studies of eggshell changes in certain North American birds, to be presented to the Proceedings of the XV International Ornithological Congress. The appendix includes: (l) geographical variation studies in the eggshells of Common Loons, accept ed, by The Canadian Field-Natural1st, (2) eggshell studies in Brown Pelicans, published by the Wilson Bulletin, .and (3) our initial results in Science. I cannot acknowledge enough the encouragement and. stimulus from my good friend and respected collegue, Joseph J. Hickey. Dr. %ckey provided me with unlimited opportunity to further myself, and for that, I. am sincerely / I. grateful. Dr. Hickey's youthful enthusiasm in research has maintained him as a leader in his field. K. W. Risebrough, J. A. Keith, and J. 0. Keith were occasional field companions as well as sources of intellectual . stimulation and aid. Many other coworkers and research aids are mentioned ' . in the acknowledgements of the separate papers. Lucille F. Stickel .and E. H. Dustman were our primary contacts and aids from our supporting ' ' agency, the Bureau of Sport Fisheries and Wildlife, Patuxent Wildlife Research Center, Laurel, Maryland. I am indeed grateful to the bureau. Had they not funded us, our'research, would not have been possible. I. must mention the names of individual persons who significantly aided us and helped us to expedite our eggshell data collection. I just Wish I could mention their names in each publications W. G. Abbott, G. D. Alcorn, D. Amadon, 0. A, Austin, Jr., J. L. Baillie, A.'M. Bailey, Laura B. Bailey, R. C. Banks, E. R, Blake, W. J. Breckenridge, D, B. Bull, T. J. Cade, J, M. Campbell, the late C. E. Carter, R. G. Chaffee, J. Cope, , HARTOLDMON0029638 X J, L. Diedrich, J. B. Dixon, H. H. Frost, A, Ganier, R. C. Hallman, W. G. Hanna, W. G. ,F. Harris, Ed N. Harrison, W. W. Hill, N. D. Hoy, . L. G, Izard, J. R. J'ehl, Jr., N. K. Johnson, P. A. Johnsgard, R. B. Lyle, G. H. Lowrey, Jr., T. G. Meitzen, R. M. Mengel, R. T. Orr, E. P. Odum, K. G. Parkes, K. E. Pauley, R, A. Paynter, Jr., R. S. Palmer, S, B. Peyton, R. J. Pickering, M. Pollock, R. Quigley, R. W. Quillan, A. L. Rafid, C. H. Richter, W. A. Squires, W, R. Spofford, E, A. Stoner, Ellie Stickney, H. 0, Todd, Jr., the late W. E. G. Todd, A, Wetmore, and L. R Wolfe. Many of my staff contacts at Wisconsin have influenced my thinking, though probably unknowingly. G. Cottam, J. T. Emlen, Jr., 0. L. Loucks, R. A. McCabe, and J. C. Neess were on various committees in my behalf. Finally, I wish to acknowledge my fellow graduate students, who were-more than beer-drinking companions. From them I have had many informative ' and professionally stimulating hours of discussion. 1 cannot say that pesticide, research has been overly encouraging, though far from uninteresting. Many wildlife problems are now deeply associated or potentially associated with industrial and agricultural pollution. Many of our problems in wildlife conservation have always been tied with, agricultural practices and in a sense, always at the mercy of industry and agriculture, whether it be mining, grazing, farming, land-clearing, drainage, etc. Now, with (environmental pollution, our field has received yet another aspect of competition. Despite the strong evidence for the potential or real hazards to-wildlife and natural systems, our society's leadership seems generally unresponsive to sound ecological 'S thinking and practice. I am fast coming to believe what Aldo Leopold saids "As for diversity, what remains of our native fauna and flora remains only because agriculture has not got around to destroying, it." I would say that-this applies as well to industry. Overexploitation of our environ- HARTOLDMON0029639 XI ment continues to be man's primary goal whether directly or via side-effects. Rather than making a conservative living* we still insist on maximum and selfish exploitation with short-term goals. I can see this in my experiences. Supposedly responsible scientists even deny the adverse ecological effects of chemical pollution,; or even worse, accept losses of diversity or potential losses as compatable with the "philosophy of progress." We have it in our own ranks. I am reminded of the pheasant biologist whose main goal was to "naturally farm" pheasants solely for the purpose of harvest. He failed to observe eggshell thinning in pheasants and on this basis concluded that wildlife was not suffering the adverse effects of pollution. His ecological world was limited to a "pheasant's viewpoint." Another of our ecologists counts blackbirds. Starlings, and White-tailed Deer and, finding that numbers are as high as ever or higher, concludes that, wildlife has not been adversely affected by chemical pollut ion. Some'even, conclude from such data that wildlife has even benefited from the use of chemicals. Perhaps a conservative approach to natural ecosystems and man's role in them is needed. Overexploitation and destruction of diversity are far from conservative. Daniel W. Anderson 13 August 1970 HARTOLDMON0029640 Ii $ xii i TABLE OF CONTENTS PART 1. SEASONAL VARIATION OF CHLORINATED HYDROCARBONS IN LAKE MICHIGAN HERRING GULLS. I' J | I. I. I I . '' imODIJCTION . 1 STUDY AREAS AND METHODS Collections and feeding trials Treatment of carcasses Chemical analyses . ' RESULTS AND DISCUSSION . General observations concerning wild adults Juvenile food consumption and general observations Annual weight and fat changes Relationships of sampled "tissues11 Seasonal residue changes GENERAL DISCUSSION AND DATA CONSOLIDATION ACKNONLEIGEMEIiTS ' SUMMARY REFERENCES . ' 1 2 2 66 1$. ' 15 1$ 22 30 33 38 ifj U9 5l PART 2. CHLORINATED HYDROCARBONS AND EGGSHELL VARIATION IN HaRRIKG GULLS.* INTRODUCTION . ' MATERIALS AMD METHODS ' Data sources Data treatment Chemical analyses ' RESULTS AND DISCUSSION ' Geographic variation' Inter- and intraspecific egg relationships Recent eggshell changes Chlorinated hydrocarbon residues, 196? Comparisons with other Herring Gulls Ability of Herring- Gulls to "tolerate11 residues So HARTOLDMON0029641 a"t 1 j j PART 3- EGGSHELL CHANGES IN CERTAIN NORTH AMERICAN BIRDS. xiii INTRODUCTION , 100 METHODS AND MATERIALS 101 RESULTS AND DISCUSSION '* 102 ' Pre-l<?U6 eggshell changes 102 Recent eggshell changes 103 Some general characteristics of recent eggshell changes' 109 Identification of pollutants ' ' 111 SUMMARY .. 113 REFERENCES FIGURES' AND TABLES .' . 21$ .120 PART it. . APPENDIX, A COLLECTION OF MISCELLANEOUS PAPERS RELATING TO 'THE GENERAL TOPIC OF THE THESIS. GEOGRAPHICAL .VARIATION IN THE EGGSHELLS OF COMMON LOONS Introduction Indices to geographical variation ' ' Materials and Methods Results and Discussion Eggshell measurements Clutch-siae General Discussion Acknowledgements References cited . 170 170 ` 171172' 172 177 179 180 182 OOEOGICAL DATA ON BCG AND BREEDING CHARACTERISTICS OF BROWN PELICANS 18$ CHLORINATED .HYDROCARBONS AND EGGSHELL CHANGES IN RAPTORIAL AND FISH-EATING BIRDS 9m HARTOLDMON0029642 2 captive herring gulls taken from the wild shortly before they attained the powers of flight. . Attention in this paper has been concentrated on p,p'-DDE, the now universally distributed metabolite of p,p*-DDT (l,l,l-trichl.oro~2,2-bis [p-chlorophenyl]ethane) in the world environment. This compound is known to be significantly involved in changes in eggshell thickness in North American herring gulls (Hickey and Anderson 1968) and has been found to produce these changes in mallards (Anas platyrhynchos) fed 3 . ppm (wet weight) under controlled conditions (Heath et al. 19o9). STUDY AREAS AND METHODS . Collections and feeding trials . We collected 1^6 prefledging juveniles in early July 1966 from the . breeding colonies in Green Bay and transported them to the Wisconsin State Game Farm, Poynette (Fig. l). Five newly fledged juveniles were shot on 3 August 1966 to obtain initial data for the feeding trials; and ten adults were shot on 31 May as a starting point for the field observations, which included collections of wild adults at. 2-month intervals. Approximately 39,700 kg of Lake Michigan alewives (Alosa W, ' pseudohanengus) were purchased from commercial fishermen in 2 batches .(batch 1, May 1966; batch -2, August 1966; Table l). They were glazed with water and stored at -20C. ,v Juveniles were individually color-banded and randomly placed in " 3.7-m by 3.7-m outdoor breeding pens, five birds per cage. Two water troughs (each 17 liters) were placed in each cage and cleaned and filled daily until freezup. During the winter months, we found that snow and moisture from food provided the birds with adequate water. The HARTOLDMON0029643 SEASONAL VARIATION OF CHLORINATED HYDROCARBONS IN LAKE MICHIGAN HERRING GULLS 1 BY .DANIEL W. ANDERSON AND JOSEPH J. HICKEY Department of Wildlife Ecology, University of Wisconsin, Madison, U.S.A. This paper reports on the seasonal variation of some insecticidal residues in wild, adult herring gulls (Larus argentatus), their build-up in a captive population of young birds over the course of 322 days and their relationships to changes in the lipid content of the avian body. Some approximations of changes in the levels of apparent polychlorinated biphenyls (PCB's) are also included. Wet-weight residue levels of p,p'~DDE (l,l~dichloro~2,2~bis[p= chlorophenyl]ethylene) have been reported as high as 2,800 ppm in the fat of a bald eagle (Haliaeetus leucocephalus) in California (Hunt' 19^9) and 5139 ppm in the fat of a herring gull collected in Wisconsin by J. A. Keith (personal communication 1964); but the dietary intake necessary to produce such levels remains to be characterized. On the basis of a small, known-aged sample df Lake Michigan herring gulls, Hickey et al. (1966) postulated that the major residue build-up in these birds occurred in their first year of life. However, the migratory habits of herring gulls at this age ruled out the possibility of relating this build-up to residue levels in the ecosystem of,a single region. In the present study, we have sought to test the possibility that high levels of DDE can be accumulated by birds exposed to a steady diet of about 5-6 ppm of this compound and related compounds. Ibis test was carried out by feeding Lake Michigan fish to a group of HARTOLDMON0029644 3 FIG. 1. Map of the general study area in which herring guils were sampled for chlorinated hydrocarbon residues in 1966 and 196?. THE GREAT LAKES KILOMETERS EKSESSEXa 0 20 40 60- N GULL COLLECTION SITE, NONBREEDING SEASON Q GULL COLLECTION SITE, BREEDING SEASON A FISH COLLECTION SITE LOCATION OF FEEDING EXPERIMENTS RING RECOVERY, RINGED IN GREEN BAY AS NESTLING . . HARTOLDMONOQ29645 k Table 1. Size of a.lewives and their pesticide residues (ppm fresh= weight), used in feeding trials with herring gulls , Measurement . Food item Alewife-batch 1* Alewife-batch 2* No. pools analyzed % water** % fat** Fresh-weight (g)/fish Mean length (mm)/fish ppm p,p'-DDE ppm p,p*~TDE + p,p'-DDT ppm apparent, PCB micrograms DDE/fish micrograms TDE + DDT/fish micrograms APCB/fish, ' A1254 equivalents 5 72.2 (70.5-75.3) 6.4 (4.8-7.8) 27.7 1 0.4 160.2 + 0.9 3.4l +0.17 2.55 + 0.20 1.84 + 0.34 94 4 70.6 3 67J* (66.3-69.0) 12.1 (11.0-13.2) 31.6 + 0.6 164.1 + 1.2 3.05 + 0.65 1.56 + 0.31 2.07+0.44 96.4 49.3 51.0 65.4 ' * Values given are means +_1 standard error, unless otherwise noted. ** Based on pool means, values in parentheses are ranges. HARTOLDMON0029646 5 '.! \ feeding trials ran from 3 August 1966 to 20 June 1967. Fresh, prethawed alewives were fed (up to ten fish/bird/day) as determined by the consumption for the previous feeding, occasionally leaving excess, Alewives left from the previous feeding were counted whole or in portion and then removed. This was done every one to four days, depending on weather conditions (cold weather), which in turn dictated the length of preservation of the fish and their acceptability by the gulls. The alewives we used were relatively standard-size (Table l) and, based on length criteria (Norden .1966), were all in age class III. We recorded number of fish consumed during each period between feedings and later converted the values to an estimate of biomass consumed. Residue analyses were conducted each month on pools of 25 randomly selected fish taken from the current food of the gulls (Table l) and combined with the biomass-consumed data for an estimate of residue intake. Five birds were sacrificed each month, and all birds were . weighed at one-month intervals. Only groups that appeared healthy and. were consuming amounts of food comparable to the mean for the entire colony were considered in this sacrifice. In adjusting the gulls to captivity and a diet of alewives, we initially encountered an outbreak of avian aspergillosus that killed 32 birds. Apparently triggered by the individual or emulative stresses of capture, transportation, confinement and malnutrition (Friend and Trainer 1969\ this epizootic Ras cleared up by adding 26 g of beef heart per week to each bird's diet for the duration of the experiment. This meat averaged<0.2 ppm (wet-weight) of total chlorinated hydrocarbons. . i HARTOLDMON0029647 6 Treatment of carcasses '' Both dead adults and juveniles were completely plucked after fresh body weights and plumage characteristics had been determined. Esophageal and stomach contents of wild-collected adults were removed and examined. The major and minor pectoralis muscles were( removed from the left side of each bird for a separate breast-muscle analysis. Equal amounts of mesenteric and subcutaneous fat (l to 2 g of each}' were pooled from each bird as a fat sample for pesticide analysis. Heads, tails, feet, and wings were removed from the carcasses after breast and fat sampling, and these carcass samples, corrected for removal of one breast and 2-4 g of fat, were the basis of our "whole- body" determinations of pesticides as well as our fat indices. All samples were wrapped in aluminum foil, sealed and frozen (-23C) until analysis 1-2 months following preparation. Just prior to analysis, each carcass and breast sample was ground and homogenized to a paste in a blender and then subsampled for fat and. water extraction as well' as for pesticide analysis. The fat samples and homogenized subsamples were ground in sodium sulfate prior to extraction. The five juveniles (sexes combined) from each month's sampling were pooled for single analyses except birds from the last sacrifice group, which was run on an individual basis. Our wild-collected samples were run on an individual basis in every case. . Chemical analysis . Since residue analysis for p,p'-TDE (l,l-dichloro-2,2-bis[p=- chlorophenyl]ethane) and p,p'-DDT is significantly interfered with by PCB's in standard, routine gas chromatography (Jensen 1966, Widmark HARTOLDMON0029648 7 1967, Holmes et al. 1967, Risebrough et al. 1969, Koeman et al, 1969, Reynolds 1969), and since our original analyses were performed before these problems became apparent, we chose to reevaluate our original findings concerning DDT, TDE and suspected, apparent PCB residues. All residue analyses and sample extractions were conducted by WARP Institute (hereafter referred to as WARF), Madison, Wisconsin. We have previously described the general laboratory procedures used for us by WARF (Hickey & Anderson 1968, Anderson et al. 1969), based on various combinations of the gas chromatography (GC) methods outlined by the U.S. Food & Drug Administration (1965-68), Risebrough et al. (1969) and Reynolds (1970). Dieldrin (not less than 85$ of 1,2,3,i|,10,10= hexaehloro~6,7-epoxy-l,h,iia,5,6,7,8,8a-octahydro-l,^-endo-exo-5,8= dimethanonapthalene) residues were removed by florisil cleanup. 1 The presence of p,p'-DDE was confirmed by WARF (D. L. Hughes, personal communication 1967) in seven randomly selected extracts representing brain tissue, carcass homogenates, breast muscle and alewives by thin-layer chromatography. It can be seen from the GC chromatograms in Fig. 2 that some interference between DDE and PCB is possible. Peak-height (PH) comparisons in all Lake Michigan samples showed DDE to be very high in relation to peak 5 and We do not believe that such interference significantly altered the DDE data. In addition, DDE quantifications were always performed at high dilutions (possible because of its high concentration) so that interference from.peak 5 was reduced to a minimum. Semiquantitative estimates of DDE, where possible, by thin-layer chromatography suggested that the levels were in general agreement with gas chromatography, our major source of HARTOLDMON0029649 8 residue data. However, DDT and TDE were either not detected by thin-layer chromatography, or suggested to be much lower than gas chromatography alone indicated, especially in herring |$ull tissues. Two small, unidentified spots appeared on the thin-layer plates which were suspected to be PCB's. Analyses of PCB's-by thin-layer chromatography have been shown to result in only a few spots, although most PCB's are mixtures of many different but related compounds (Lichtenstein et al. 1969). Our estimations of PCB-like compounds were based on standards of Aroclor 125k (Monsanto Chemical Co., trade name), expressed as "A125k equivalents". The selection of A125k (Pig. 2a) was based partially on the findings of Risebrough et al. (1969) and others (reviewed by Reynolds 1970) who found that of the commonly used Aroclors, the chromatographic pattern of A1254 (and that of A1260) most closely approximated those of field specimens on several columns- Likewise, we found a close resemblance to A1251* in the chromatographic patterns of our field specimens when they were compared to A12k2, AI2J48 and A125k. Bagley et al. (1970) have positively identified PCB's in bald eagles by mass spectral analysis and have demonstrated that most were identical to the PCB's found in Aroclor 125k. Veith and Lee (1970) showed that Lake Michigan waters and fish contained predominant and "apparent" Aroclor 125k mixtures, confirming the presence of PCB's by 3 methods: (l) comparative retention times of various peaks under different, analytical conditions, (2) infra-red spectrophotometry and (3) mass spectrometry. ' Our objectives in apparent-PCB quantification were not as much HARTOLDMON0029650 . .9 FIG. 2. Tracings of chromatograms of extracts of various tissues as seen on the DC-200. A. Aroclor 12$k with peaks numbered as discussed in the text. Arrows show the positions of the major DDT-family res idues. B. Saponified extract of a mixed garbage sample taken from a gull which was shot while feeding in a dump. Peaks 6, 7 and 8 did not change after saponification. C. Alewife extracts before and after saponification, showing interference of peaks 6 and 7 with p,p'-TDE. D, Juvenile herring gull fat extracts before and after saponification. E. Unsaponified and saponified extracts of an adult herring gull carcass homogenate, showing the nearly undetectable residues of p,p'-TDE and p,p'-DDT and the stability of the apparent PCB peaks 6, 7 and 8. Peak 11 suggests the presence of another' compound not prevalent as seen in A1251|, possibly representative of accelerated accumulation of a higher number, or mixture from A1260. In some instances, p,p'-DDE is off-scale and shown with dashed linesj in others, it is shown out of the linear range of response of the electron-capture detector. In all cases, DDE was quantified by a separate dilution. Slight discrepancies in retention times between the various chromatograms represent slight day-to-day variations in operating conditions. . HARTOLDMON0029651 10 HARTOLDMON0029652 SAPONfFfro SAPONIFIED 11 toward quantitative precision as they were toward internal constancy of estimates, so that comparisons with other residues within the system ve studied could be made and so that temporal comparisons of apparent-PCB's could be obtained (see Risebrough et al. 1969). They are identified in the same units as the other residues, but it should be understood that they may be less precise or inaccurate by some unknown constant. We did not,, attempt to identify any specific Aroclors by pattern resemblances and doubt that any single Aroclors compose the sole apparent PCB residues in any of our samples. ' Greater toxicity of the lower molecular-weight, more volatile, lower-chlorine-content compounds in PCB mixtures is known to occur to house flies (Musca domestica) and Drosophila melanogaster (Lichtenstein et al. 1969). One might expect a greater biological degradation rate of such compounds under field conditions. Jensen et al. (1969) suggested that the lower-number (lower chlorination) PCB components are probably excreted or metabolized at a greater rate than the higher numbers, thus resulting in food-web concentration of the higher numbers. Degradation of certain of the earlier retention-time compounds in A125b (QF-1 and DC-200 columns of gas chromatographs) has been reported foiv mallards (Anas platyrhynchos) by Risebrough et al. (1970). Our estimates of apparent PCB's based on A125k assumed a relationship between their total amounts and the.later-retention-time peaks 6, 7 and 8 (Fig. 2). . ; In the rereading of our earlier chromatograms from prior to WARF's ttse of saponification, p,p'-TDE, p,p'-DDT and apparent PCB's were estimated on the basis of various correction factors (CF's). Based on the data of Jensen et al. (1969) regarding degradation, we tested the '7 HARTOLDMON0029653 12 hypothesis that peaks 6, 7 and 8 were different in relation to one another between the various sample levels (Table 2, col. l) but not between the tissues in any level. Statistical analysis substantiated this (P<0,001) when randomly selected examples of the various classes of samples were reanalyzed with saponification to remove DDT and TDE (Table 2). All three corresponding peaks in A125k standards remained unchanged after saponification. Peak 7, according to the ratios given in Table 3 and compared to standards, in bur samples appeared to be relatively the most stable of the three in such biological materials. The differences, then, were used for the first correction factor determination, CPI (Table 3, col. 7). Total PH6 + PH7 + PH8 in Aroclor 125^ was considered as 100 percent of CF1. PH7 was found to be measurable in nearly all instances on the DC-200. Interference of peaks 6 and 7 with p,p'-TDE in some fish pools (Fig. 2) caused difficulty in measuring PH7 accurately; therefore, the apparent PCB determination in these samples was estimated only after saponification revealed its true height. The peak.height ratios of 6/7 and 8/7 were multiplied by the PH7 of standards (32.2 percent,of peaks 6, 7 and 8 as determined by actual height measurements in many standards). They were then added to .322 for each respective sample level to obtain CF1. Peaks 6 and 8 interfered with p,p'-TDE and p,p'-DDT, respectively (Fig. 2). Peak 7 did not interfere with any of the DDT-family residues, and its PH was used to determine the expected PH6 and PH8 in .chromatograms of unsaponified extracts. The ratio differences given in Table 3 (cols. 3 to 6) for each of the different sample levels were the bases used to determine the expected ratios. Differences between the peak heights in saponified and unsaponified extracts were assumed to '' 8 HARTOLDMON0029654 13 Table 2. Comparisons and statistical differences between the three major peaks of PCB-like compounds in saponified extracts of various Lake Michigan sample classes as compared to A125k standards Sample level Variable Standards Whole fish Juvenile gulls Gull eggs Adult gulls No. Peak ratio (6:7) Differences* Peak ratio (8:7) Differences* CPI CP3** ` lh 1.20 A 0.91 A 1.000 7.3^ 9 0.67 B 0.77 B 0.775 5.69 10 0.58 17 0.50 10 0.46 C DD 0. 48*** 0.73 0.71* CB . . BCD 0.7k* XX 0.7^6 0.715 5M 5.25 0.71 , BCD 0.715 5.25 * Differences not significant at the 99% level share a common letter, and were determined by the least-significant-difference-test and analysis of variance (Steel & Torrie 1960:106-107, 112-115). These relationships did not change at the 99% level. ' ** CF2 = (Cone. A125^injected/Conc. DDT injected) X (PH DDT/PH7), CP3 = CF2 X CPI. ' *** These ratios were combined on the basis of' nonsignificant differences. HARTOLDMON0029655 ' 14 represent DDT and TDE. The estimations based on rereading original chromatograms were in agreement with values based on later saponification of the same extracts (r 0.95, P<^0.001). The major, apparent PCB peaks were crudely estimated on the basis of PUT and on differences in sensitivity of equal volumes of A125** (peak 7) and DDT standards at a given attenuation on the gas chromatograph. Peak 7 was quantified as though it was p,p'-DDT, and small differences between the actual PH7 and PH DDT standard were adjusted (Table 2), The calculated CF2 values from nine chromatograms each of the two standards, injected at different times and at different volume-pairs in the linear range of response averaged 7.3** (S.E. = 0.17, CV, = 7$). A final CF, CF3, was calculated for,each of the four sample levels (Table 3, col. 8) and a PCB-index determined for each analysts: apparent PCB (Est. ppm, A125h equivalents) = CF3 X Peak 7 measured as DDT. , Dieldrin residues were detected in all the samples (10) specifically tested for that compound, but at low levels in comparison to the other chlorinated hydrocarbon residues. In herring gull fat samples in May 1966, dieldrin composed 0.19$ of the total chlorinated hydrocarbon load. Reinert (1970) found that low levels of dieldrin were present in all Lake Michigan fish analyzed. In further discussion here general chlorinated hydrocarbons of major importance because of their high levels ..are ;consi_4ei,ed.as J3DE,. J2DT,-- -TB1--and-F-GB-lik-e -resi-duesT--------- - Dieldrin and other chlorinated hydrocarbon residues not evaluated in this study remain a problem for future research. HARTOLDMON0029656 RESULTS' AND DISCUSSION 1$ General observations concerning wild adults The alewife most likely represented the dominant food, source'of Lake Michigan herring gulls during the general period in which this study was conducted. In 1965, this species was estimated, to comprise approximately '[h% of the total weight of Lake Michigan fish in exploratory trawl catches (Bureau of Commercial Fisheries 1966, hereafter called BCF). Ludwig (1966) has shown its importance as a food of herring gulls during the breeding season, but believes that insects were formerly more important in the birds' diet. He found that alewives comprised &3% of occurrence, smelt (Osmerus roordax) 10# and yellow perch (Perea flavescens) 3% of the diets of Lake Michigan and Lake Huron herring gulls, to attest to the importance of fish in the general diet in spring. The alewife has been abundant in Lake Michigan since the mi'd-1956's, is present in large numbers in all parts of the lake, and was believed to be still increasing in 1966 (BCF 1966), During the early winter of 1966-67, large numbers of herring gulls fed at various city dumps for about one-fourth of the year. By February no gulls could be found in any of the five dumps we visited, although large amounts of garbage were still available. Our hunting pressure did . not seem to affect the gulls' use of dumps, as local caretakers reported large numbers soon following our efforts< Fish remains discarded there by fish .processors.,even .then, .seemed. important -in- -the--birds--diet-. -..- - Later in the winter, we found herring gulls congregating about winter-fishing operations, in active harbors, along the ice-sheet edges 150 m or more offshore, and 1/2 to U km offshore over open water. At '' 10 HARTOLDMON0029657 . -t*v > 16 this time, gulls were feeding on fishing debris and unwanted species (mostly alewives) discarded by fishermen and turned up in the backwashes of fishing vessels. Gulls also appeared to be feeding on materials not associated with any human activities. Our small samples in stomach and esophageal analysis (Table 3) were evaluated on a qualitative basis only, but the data indicated general food-habits of the wild herring gulls we studied in 1966 and 1967. Fish included chubs, smelt, yellow perch and possibly other fish species in addition to the dominant' alewife. Plant materials included Polygonum seeds, Ulmus leaves. Thuja tips, grass blades and unidentifiable sterns and leaves (tree and lawn trimmings?), most of which was probably picked up incidentally to other food items. Actual identifiable garbage included meat trimmings, potato peels and pieces, noodles, lettuce, eggshells, and associated items such as toothpicks and miscellaneous wrappers. -Invertebrate materials included remains of Coleoptera, gastropods, crustaceans, and unidentifiable insect parts. Items classified as "miscellaneous" included feathers and gravel. All items but fish and garbage represented minor food items. It is apparent that Lake Michigan herring gulls are predominantly fish-eaters. The most important potential factor regarding residue levels in food which might have "diluted" the residue levels, of wild birds during the winter months was garbage. Miscellaneous,'random collections from .esophagi-did not bear this out, however, as the few samples suggested that residues in "garbage" were at least comparable or higher than those in alewives. Pork fat, representing meat trimmings, a common form of garbage, taken from a single bird in August 1966 averaged 3.85 ppm (wet-weight) of DDE, 0.28 ppm of apparent PCB, with DDT and TDE ' 11 HARTOLDMON0029658 17 HART OLDMON0029659 undetected, A pool of esophageal contents from gulls shot on 4 October 1966 averaged 3.94, none detected and 12.7 (Fig. 3b). This pool was composed of fish remains along with mixed garbage and much unidentifiable material. Plastic wrappers were commonly found in many of the gull stomachs. Chubs (Leuciehthys sp.) on which gulls were found feeding (discarded by fishermen) contained 4.31, 1,26 and 1.53 ppm (wet-weight) of the residues DDE, DDT + TDE and apparent PCB, respectively. Partly digested alewives taken from esophagi and stomachs of herring gulls shot near their breeding colonies on 3 August 1966 averaged 1.25 ppm DDE, 0.18 ppm DDT + TDE and 0.26 "ppm" apparent PCB. These values are the lowest we found for any alewives and suggest that samples taken directly from the alimentary tracts may somewhat underestimate the actual residues present. Alewives (2 pools of 25 fish each) picked up fresh near the breeding colonies on 23 June 1967 averaged 2.44, 0.97 and 1.39 of the same chlorinated hydrocarbons. About all that we can conclude from the above observation^, is that in these Wisconsin dumps, chlorinated hydrocarbon residues were present that were as high or higher than those found in natural food; DDT and ' TDE residues were not detectable in the few garbage representatives not definitely associated with discarded fish. - In early May, just prior to intensive nesting and breeding activity, flocks of 25 to 300 herring gulls could he found loafing on alfalfa and ' stubble fields near the shore. They were eating small insects at this time, but most activity appeared to be associated with reproductive behavior rather than feeding. Coleopiera taken by nesting gulls in this region ran 0.5 ppm DDE (wet-weight) in 1964 (Hickey et al. 1966). 12 HARTOLDMON0029660 19 Juvenile, food consumption and general observations We found that the weight of alewives, their corresponding water and lipid contents and their chlorinated hydrocarbon residues varied between late spring and early summer (Table l). Since p,p'-DDT is commonly converted to p,p'~TJDE in frozen specimens under anaerobic conditions (Jefferies & Walker 1966), we felt these had to be considered together in data analysis. The 1966 levels of DDT-complex (DDE + TDB + DDT) residues in alewives from our August batch 2 (Table l) were oply slightly higher (on a ppm, wet-weight basis) than those reported by Reinert (1969) for our general study area; however. May batch 1 alewives.were higher (by about 1 ppm) than any alewives he reported for Lake Michigan. Reinert (1970) reported a general DDT-complex of Lake Michigan alewives in 1967-68 as 3.89 ppm (96 analyses), a value which we believe to be closely comparable to the general levels we found. On a 100$ lipid basis, the residues reported by Reinert were 37.6 ppm, compared to 38,1 ppm in batch 2. The higher levels in batch 1 on a lipid basis (93.1 ppm) suggest a seasonal variation in alewife residues not strictly related to fat reservoirs (Table l). Actual weight of food consumed per captive herring gull per day peaked in December and dropped to a low in early May (Fig. 3a). Food consumption was probably again on the increase after that low. The -i April-May low in food consumption coincides with the normal breeding season of Green Bay herring gulIs,..although.theg e..juveniles.were..not. comparable in gonadal development to adults. The general period of peak egg-laying during the springs of 1966 to 1968 was iate-April to mid-May. We found that juvenile gonads were discernable by September HARTOLDMON0029661 20 FIG. 3- Seasonal variation in the amount of food and associated fat and residue components ingested by penned, Juvenile herring gulls. Since periods between samplings were not equal, points are placed at midpoints for each period and are expressed as units/bird/day. Sampling dates are shown with a small dot near the lower axis. Closed circles represent batch 1 alewives, and triangles represent batch 2. HARTOLDMON0029662 21 following fledging, but not readily so until December. Juvenile gonads were at their maximum, however, by mid-June 1967, but still not then comparable in size to adult gonads in their most regressed stage d\iring the winter (juvenile $ - 5 by 8 mm, juvenile = 1 by 3 mm; regressed adult $ = 6 by 10 mm, regressed adult <? 3 by 5 mm). Juyeniles were molting primaries and contour feathers in June 19^7 and beginning to show typical subadult plumage. One juvenile, however, a dark bird, was molting into new feathers as dark or darker than its original.plumage. We found that our captive herring gulls rarely failed to eat during each day, except during short periods of extreme heat or cold. Average weight of food consumed by the experimental colony and the groups we sacrificed were comparable (95# C. L.) in all instances but one (Fig. 3a), which we consider as a 1 in 20 sampling mischance. A noticeable increase in amount of lipid consumed occurred in February and March 1967, although actual weight of fish consumed was dropping (Fig. 3a), solely . as a function of the doubling of percent lipid in batch 2 alewives. It can be seen that resultant residue exposure solely, from the food viewpoint in our experimental group (Fig. 3b) was a possible function of at least three different variables: (l) variations in amount of food consumed throughout the year, (2) variations in lipid content of food and. thus in the potential residue reservoir or carrier, or both, and (3) variations in residue content of that food possibly related to. different seasonal exposures faced by alewives nr variable peRtfcfde______- kinetics in alewives at different times of the year. Robinson et al. (1967) have demonstrated distinct seasonal variation in two species of fish from the east-British coast to further support this final conclusion. , .14 HARTOLDMON0029663 Annual weight and fat changes 22 ' The affinity of most chlorinated hydrocarbon residues for fatty tissues leads to inquiry concerning variations in lipid weight (volume) as possib3.y related to total body burden. Herman et al. (1969) showed that TDE concentrations in various tissues of western greb.es (Aechmophorus occidentalis) were related to their fat content, but no information was available on total body burden. The consideration of adipose tissue in the compartmental model of pharmacokinetics.as generally applied to chlorinated hydrocarbons (see Mrak et al. 1969: 271-273 for a review) dictates the importance of understanding seasonal changes in the lipid deposition of birds. Sexual dimorphism in size is well known in larids but is not related to differential niche utilization of food-habits between the sexes (Ingolfsson 1969). Fully aware that males are largest, we chose to combine the sexes for our data analysis here, so that values should be ' considered as "average adult" rather than representative of any one sex. Sex ratios in our sample were not significantly different from a 50:50 ratio as determined by Chi (P<0,1). Weight and lipid deposition in both adults and juveniles followed a similar pattern over the year after early weight gains by juveniles (Table 4). Lipid deposition in juveniles, however, was greater in all , cases, even though fat-free carcass weights were consistently lower than those of admits. Adult.weights varied less than those...of .juveniles -.......... 1 (Fig. 4; CV, adults = 4$; CV, juveniles - 13$), with a small adult weight-increase associated with molt in early August 1966, and a large increase in early 1967 due to increased fat deposition (Table 5). 15 HARTOLDMON0029664 23 HARTOLDMON0029665 TABLE li c o n tin u e d TABLE li c o n tin u e d . * g (0 q bfl d a d a> CO CO q q CO +2 42 $ Js r~l . <id4 UCO O U co u o q *d S qo CO dp s fee p w "W. bp CA .g r-i tt *H ft H H o p d r, d o bp o H o -d if Vi ii rt CO 5p ft d q ft P Q> P P Vi o p m o n Q> X! CO c3 > P H a a) ,3 H s qH o *d x? CO * !s g d a O G> r* &$. H ci Hd &q c3 CO W Hh in p C-* tn d vO ON ii Po rH q ft t. CO 3p 0) b o XI bO X3 p *P Is O s H CO f\ CO 3 ft H H ? d o p q o p CO Vi o d o H 43 P .. ft d CD <D Op Xd H o CO d s *% o xf o sf CO d d o q P. 05 q 'd feV. H p p XA *rl V!. ri Vi Op II cd <n H - q q <3 p > s i* --<H- - iH O p O a. O CO ft p O dq q H o H ft n (D Vi w. CO G) ft 3 X3 O tH o jj5 5jC ` xf 0 bO o cd c~- 8 ft q o *0 d s P. p p d 0) CO q ft q !> p P p . .- - P u - ft CO d r~i d f> CO <!> CD W) d U o p(0 #%v VvU * to Hft , 1d5 q <H H CO P d Pq H ,,,_!a p CO `RH XJ d 43 43 d H 00 0 p tj 0 O g ft -H- 0 u 2h HART OLDMON0029666 25 Slightly smaller fat-free carcass weights in all juveniles' throughout the experimental period suggested that these birds had completed most, but not all, growth in their first year. Breast-muscle weights did not vary significantly over the year in either age class after initial increases in caged juveniles, and they were only slightly smaller in caged juveniles (Table k). Comparisons between fat weight and total body weight fell into three categories-(Fig. ^). Postfledging juveniles stored fat reserves heavily during their first k months. During the remainder of the year, the caged juveniles continued to store fat more heavily than adults. The increased lipid stores in juveniles may have been in part a function of their being captive, or a function of their immaturity. Adults would be expected to metabolize greater amounts of lipids in the m. wild where food would not be so constantly available. Greater fat deposition in first-year birds may be a general adaptation to their first year of necessarily independent life (see Tinbergen 1960:25-38), The two wild-collected juveniles in February 1967, weighed as much as caged juveniles (1323 g vs. 1328 g) but stored slightly less fat (222 g vs. 311* g). In any case, they were more like caged juveniles than adults, to support the hypothesis of increased fat deposition in young, wild birds. Increased lipid deposition in juveniles, where we had , data on food intake, was at least in part related to the amount of food consumed (Fife. 3. Fig, . 6).__________ :------------------------------------------------------------ :-------v Changes' in body weight in both wild adults and caged juveniles seemed most closely related to the amounts of fat deposited, except in wild adults in August, the period when they were molting heavily (Fig. - 16 HARTOLDMON0029667 26 FIG. I*. Relationships between weight of deposited lipids and total body weight in different-aged herring gulls. Line 1 represents post-fledging juveniles (all caged except the August sample), r 0.$>69, P<0.0f?. Line 2 represents caged juveniles after six months of age, r - 0,989, P< 0.01, Line 3 represents wild-shot adults, r = 0.963, P<0,01. The open circle represents molting adults and was not used in the calculations. HARTOLDMON0029668 l). This molt correlates with the general postnuptial molt described by Dwight (1925:93-96) for large larids in general. By the time we started collecting juveniles, they had already completed major feather growth. King et al. (1965) showed a decrease in fat-free body weights correlated with postnuptial molt in white-crowned sparrows (Zdnotrichla leucophrys). The fat-free body weights of August herring gulls increased only slightly. We therefore conclude that the total weight increase was associated with the parts we removed, probably the wings and feathers. Perhaps the increased fat-free carcass weights in adults (Table 5) reflect increased synthesis of feather proteins and increased water, etc. in the parts we removed. Nonfat dry weight of adult carcasses was relatively constant throughout the year, although slight increases appear to have occurred during the winter (Fig. 5a). Child (1969) has described the year-round, constancy of water to nonfat dry weight in Swainson's thrushes (Hylocichla ustulata), and similar relationships were described by Child and Marshall (1970). This relationship, however, was even more distorted in our-caged juveniles (Fig, 5b), and cannot be explained on the basis of sampling error or unbalanced sex ratios. Two hypotheses are offered: (1) increased protein synthesis occurred during the winter months in juveniles, and (2) increased glycogen storage or synthesis occurred. during the winter. Hanson (1961), studying energy reserves in Canada geese (Branta'canadensis interior), showed that glycogen reserves were utilized first for energy, but he suggested than an interrelationship existed between carbohydrate, fat and protein metabolism. It may be that increased glycogen synthesis is needed to maintain the herring gull HARTOLDMON0029669 28 FIG. Variation in nonfat dry weight and weight of water in juvenile (A) and adult (B) herring gulls over a period of one year. WEIGHT HARTOLDMON0029670  2? occurred in both age classes during the coldest period of the year (January-February). By April, fat levels were generally comparable with those of the previous May. ' Young herring gulls are known to be much more migratory--rather, they show a greater tendency for dispersal---than adults (Gross 19^*0, Smith 1959). The idea of lipid deposits as a "fuel" for migratory flights-has been .proposed in such studies as Odum & Connell (1956), Raveling & LeFebvre (1967), and others. Weise (1063) suggested that extensive vernal fat deposition in several species 'of overland-migrating passerines was primarily an adaptation to unfavorable weather conditions, Baldwin & Kendeigh (1938), studying gross weights in 2k species of birds, showed a general inverse relationship between weight and, temperature*, and, differences in "weight-temperature" curves between different species were suggested as being related to migratory status and distribution. The potential importance of stored energy and physical condition (both tied in part to lipid.reserves) to winter survival in such sedentary species as house sparrows (Passer domesticus) and ring-necked pheasants (Phasianus eolchicus) has been demonstrated (Davis 1955 Rabat et al. 1956). Lipid reserves might function not only as potential energy reserves but also as additional insulation to protect against heat loss. Brenner (1967) suggested that the adaptive value of reserve energy supplies in red-winged blackbirds (Agelaius phoeniceus) was not as importantly tied with migration per se as it was with reproduction and inclement-weather needs. The general role of internal rhythm in regulating physiological changes in birds, including i- . lipid changes, in relation to migration and breeding is reviewed by ,, Marshall (1961:307-339). In general, the degree and condition of 18 HARTOLDMON0029671 . -30 migratory state (viz. the distance, route and climatic circumstances surrounding movements and distribution) as related to the internal rhythm, seem to dictate the chronology, degree and adaptive function of fat deposition. In herring gulls from Lake Michigan, fat deposition seems to have a major adaptive function in conditioning the birds to environmental changes- associated with winter conditions. Possible roles of large amounts of fat at that time may relate to both insulation and labile energy stores. In juvenile herring gulls, even greater amounts' of stored fat (other than those we observed as possibly related to captivity) might serve the additional function of.having adapted the birds to an independent life in competition with more experienced adults and to`a more nomadic habit than adults. . Relationships of sampled "tissues11 . We feel that the most reliable index to seasonal changes in body burden is mg/kg (ppm) in whole carcass homogenates because these values comparatively represent the body burdens of different-sized birds. It also seems most suitable in calculating the total micrograms present (an absolute measurement of body burden). Dindal (1970) has shown that fat samples (ppm wet-weight) vary considerably from site to site on the carcasses of mallards (Anas platyrhynchos) and lesser scaup (Aythya affinis). He warned against the use of such in extrapolating to body -burden.--We--found...that- micrograms ^m- careass' set.based- on the small fat biopsies from individual, wild adults failed to represent total carcass ^micrograms as calculated from mg/kg values$ but, in juveniles where fat was more easily obtained and where samples were pooled and large, the 19 HARTOLDMON0029672 31 relationship was more satisfactory (Table 5). Breast-muscle extractions (whole breasts) more closely represented total-carcass homogenates in both relative lipid content and ppm lipid-basis, thus amount of chlorinated hydrocarbons present (Table 5)- The carcass (X) and breast (Y) lipids were related as expressed by the regression equation, Y 2.86 + 0.13 X, with both juveniles and adults combined. Neither the slopes nor the intercepts of adult and juvenile regressions were significantly different from one another (non-overlap of 95$ Confidence Limits as described in Simpson et al. 1960:224-229)-. The intercepts were significantly different from zero and suggested that lipid levels below 2.9 minus 1,3 percent in breast (lower C.L.) were not associated with seasonal fat storage but -rather other physiological functions. Percent lipid in carcasses was likely dominated quantitatively by seasonal fat changes and most representative of that value. "Fat" biopsies from our adult herring gulls varied seasonally in extractable^ lipid content as follows (means + 95$ C.L.'s in sequence of collection): 58+3-0$> 58+13$, 50+18$, 82+_4$, 81+11$, 79ji5$. Juveniles varied in a . similar pattern from 51 to 99 percent but were more consistent at the higher percentages after October 1966. The seasonal differences in the above-described lipid contents of "fat" more likely represented variations associated with differences in lipid storage, resulting in variations in the relative amounts of other tissues such as water and possibly connective tissue. The data suggest that muscle biopsies might be more satisfactory in monitoring wild birds where whole carcasses are not available or in remote field situations where whole specimens, cannot be saved. . Levels of DDE and apparent PCB in eggs represented adult-body. 20 HARTOLDMONOQ29673 32 TABLE 5* Comparisons between measurements of residues by different methods and in different tissues ' Variables .Independent Dependent r, adults (n6 periods) r , juvenile s (n<Ll periods) % lipids, carcass % lipids, breast 0.94** MICROGRAMS DDE IN STRIPPED CARCASS BASED ON mgAg (ppm) in fat biopsy and carcass wt. extracted fat -0.10' MICROGRAMS APPARENT PCB IN STRIPPED CARCASS BASED ON mg/kg (ppm) in fat biopsy and carcass ' wt, extracted fat -0.22 PPM DDE, LIPID-BASIS IN stripped carcass breast muscle PPM APPARENT PCB, LIPID-BASIS IN . 0.93** stripped carcass breast muscle 0.88** % approaching 9%% significance. * significant, at 9$% level. ** significant at 99$ level.' 0.90** 0.59* 0.77** 0.99** 0.87** HARTOLDMON0029674 33 burdens (mg/kg) multiplied by an average factor of 0,75. When comparisons were made on a lipid-basis, the factor increased to 0.95. DDT + TDK was present in small amounts (6% of DDE + TDB + DDT in adults and 2% in eggs) and the factors varied from 0.4 to 0.5. We have generally found very low levels of DDT + TDE (relative to(DDE) in herring gull eggs from Lake Michigan since our 1966 sample. Enderson & Berger (1970) reported a positive correlation between ppm lipid-basis dieldrin in eggs and fat of prairie falcons (Falco mexicanus)., but the factors calculated by us for high (60-100 ppm) and low ( 25 ppm) levels were only 0.18 to 0.32. The correlation between levels (ppm wet-weight) in breast muscle and levels in eggs from oviducts of western grebes reported by Herman et al. (1969), though no conversion factor was given, further support the relationship between adult female residues and ' residues in her eggs. We regard egg residues as an index to female condition, thus an indirect, but valid measure. Residue levels in egg contents representing field data have been related to eggshell changes in several species (Hickey & Anderson 1968, Anderson et al. 1969, Fyfe et al. 1969, and others). Seasonal residue changes The overall picture of seasonal residue changes in adult and juvenile herring gulls is represented in Fig. 6. Breast and carcass residues followed similar patterns, as expected; thus, only carcass= based..residues...are expressed here. Temporal variation in pesticide residues has been demonstrated by Robinson et al. (1967) with- dieldrin in shags (Phalacroeorax aristotelis), by Herman et al. (1969) with DDT/TDE-derived compounds in western grebes, by W. L. Anderson et al. HARTOLDMON0029675 FIG. 6. Seasonal variations in the body burdens of chlorinated hydrocarbons in adult (A) and juvenile (B) herring gulls from 1966 to 196?. HARTOLDMON0029676 35 HARTOLDMON0029677 LIPIDS AS PERCENT TOTAL BODY WT. 36 (1970) with DDE, dieldrin and heptachlor epoxide (1,14,5,6,7,8,8= heptachloro-2,3-epoxy~3a, 1+,7,7a-tetrahydro-4, 7~methoniondan) in ring= necked pheasants, and others. Various explanations have teen given for such changes. The decrease in relative "bodyburden" (mg/kg) from, eggs to fledglings (Pig. 6B) is similar to that described for shags by Robinson et al. (1967). It is presumably a "dilution" in relative body burden by growth. Total micrograms of DDE and apparent PCB, respectively, in eggs versus fledglings, increased or decreased depending 4h L on the residue: from 1,5 x 10 to 2.3 x 10 and from 1.2 x 10 to 5-9 x 10^, All residues in adults (Fig. 6A.) seemed to follow a cyclic pattern, varying somewhat in relation to the fat cycle. DDE ran , opposite the fat cycle in December 1966 and between February-May. The only explanation for the December decrease in DDE that we can offer is the possible dilution by human garbage at that time. It did not occur with the other residues,. nor did it occur in juveniles being fed alewives at the same time (Fig, 6B). Apparent PCB in adults seemed stable at around 160 mg/kg (the average of the four middle points and considered here as "equilibrium"), except in May and June, the breeding season. Our best estimate of a theoretical "equilibrium" for DDE was considered as the mean of months 2 through 4, 175 mg/kg. These means . are estimated and theoretical in equilibrium concept, and they are proposed here for discussion purposes. They seem to represent one phase of.two in the adult residue-picture and one In three for caged juveniles. DDT + TDE in both adults and juveniles was closely related in direction of change to the lipid cycle (Fig. 6), The three apparent phases in juveniles for DDE and apparent PCB are (l) dilution of /relative, amounts from egg to fledging by probable growth, then general 22 HARTOLDMON0029678 . 37 stabilization of weight, (2) post fledging buildups to "equilibrium" levels until the following early spring and (3) sudden increases in body burden associated with rapid decreases in lipid reserves, followed by a return to "equilibrium." The theoretical "equilibrium" levels in phase 2 for juveniles were estimated by eye for both DDE and apparent ' PCB (we assumed the curves would follow an asymptotic form such as that described by Laug et al. i95o),ard thsy are given in Table 6, Phase 3 implies that ppm lipid-basis in fat samples had to increase, since body burden (mg/kg) increased when lipid reserves decreased. We found that this supposition held in juveniles where fat biopsies were generally satisfactory, but it could not be observed in adults. The values (ppm lipid-basis in fat biopsies) for juveniles in sequence were as follows: DDE = kh2, 306, 189, 252, 231, 192, 2U, 298, 968, 1927, 915; Apparent PCB = lb6, 67, 82, 9^, 203, 229, 223, 219, 507, 776, 629. Phase 3 residues represented the maximum levels attained in both juveniles and adults, when the two age classes were closely comparable. Juvenile levels were 290 mg/kg DDE, 19 rag/kg DDT + TDB. amd 200, mg/kg apparent PCB. This compared to wild adults as follows: 300 mg/kg DDE, 4 mg/kg DDT + TDE, and 219 mg/kg apparent PCB. In general, it seems safe to conclude that caged juveniles fed Lake Michigan fish assumed a residue pattern similar to adults in their first year, acquiring phases 2 and 3, and then probably continuing as p; such. Residue buildups of DDE in juveniles similar to the residue levels in wild, spring adults occurred in the same time periods on a fish diet of generally 3 ppm DDE, 2 ppm DDT + TDE and 2 ppm apparent PCB (Table l). The contribution of dietary DDT or TDE, or both (see |- Abpu-Donia & Menzel 1968, Ecobichon & Saschenbrecker 1968 and Bailey 23 HARTOLDMONOQ29679 et al. 3.969a, b) as veil as dietary DDE to total body burdens of DDE in herring gulls seems clear, and the rapid and extraordinary accumulation of DDE compared to DDT + TDE vas most likely related to both (Table 6). Carcass residues on a ppm lipid-basis, however, were much lower in juveniles which had more fat: 1305 ppm compared to 2558 ppm and 10.5$ (whole-bird basis) compared to 5.h% lipid in adults. The smaller relative amounts of fat in adults at this time were presumably related to additional stresses associated with breeding; nonetheless, resulting DDE body burdens were about the seme. It may be, however, that the larger concentrations of residues in adults may be of more physiological importance. Baaed on the assumption that peak tissue storages of the chlorinated hydrocarbons discussed here behave as DDT and dieldrin in phase 2 of the annual cycle, that is, that they are directly related by some factor to daily dose (reviewed by Hayes 1959:58-70 and Mrak et al. 1969:263-265), poor adult-juvenile comparisons of theoretical "equilibrium" levels for apparent PCB and DDT + TDE (Fig. 6) suggest that additional food sources with less DDT + TDE and more apparent PCB than found in our alewives partly contributed to the adult residues. . GENERAL DISCUSSION AND DATA CONSOLIDATION The concept of "equilibrium" levels (see Laug et al. 1950, L. F. Stickel et al. 1966, Robinson 1967, Robinson and Roberts 1968, Mrak et al. 1969 and others) seems to apply only in part to the annual residue cycle in herring gulls, the initial buildups of phase 2. Unfortunately, as we did not measure liver residues and weight (volume) changes, we cannot apply the compartmental model to our juvenile data generally as . 2k HARTOLDMON0029680 TABLE 6 . Change in r e la tiv e p ro p o rtio n s o f DDE, DDT+TDE and apparent PCB from d ie t to body burden o f caged 39 HARTOLDMON0029681 . . . UO described by Mrak et al (1969) and Robinson et al. (1969). Our initial calculations concerning the total, yearly dynamics of chlorinated hydrocarbon residues were based on monthly changes, since so many potentially influential variables were in a constant state of flux : (Table 7)* Of course, percent of total intake lost exceeded 100$ when the gulls were in the "return to equilibrium" portion (phase 3) of the annual cycle (Table J, periods 9 and 10, between 8 and 9, and 9 and 10} when stored residues were presumably being lost as well. We suspect that greater turnover rates in the central compartment were most important in these changes. During phase 2 (periods 1 to 7 months 1 to 8) we applied drug absorption formulae (Nelson 1961) as previously applied to heptachlor. in woodcocks (Philohela minor) by W. H. Stickel et al. (1965a). The basic assumption of this general model is that each daily dose yields a given concentration, but that the loss from the animal is proportional to the amount present (also Robinson 1967). Based on these calculations (Table 8), nearly all of the DDT-complex residues were taken into the body. Residues of p,p'~DDE most likely represented, as already suggested (Table 6), partial conversion from other DDT-related residues and partial buildup of dietary DDE. The large gain rates calculated for apparent PCB (Table 8) seem explainable on one of two bases: (l) the basic assumption that loss rate is constantly proportional to amount __ present does not hold and these residues tend to be more accumulative than would be predictable by normally used models, or (2) our estimates of apparent PCB's were either too low for fish or too, high for birds. The suggested increase of a higher number PCB in adults (Fig. 2e) points to both possibilities, as higher number (higher chlorine content) PCB's . ?5 ! .? S 4: HARTOLDMON0029682 hi HARTOLDMONOQ29683 Table 7 continued orH 0\ CO r~ 'O (XU0S)i u\ -=t C''* CNJ H rSH> 0) a r9l sw0) fc&> 04 1*2 43 TABUS 8. Residue dynamics for the gain period in juvenile herring gulls fed Lake Michigan alewives for a period of 219 days . ,' Residue Loss rate (5/day)* Percent of Daily dose Mean daily daily < absorbed (Ag) intake (Ag) absorl p,p'-DDE p,p'~DDT + p,p> -TBE DDT-complex res idues Apparent PGB 0.6? 1-61*. 0.74 1.09 916 320 1155 842 ' 686 481 , 116? 389 1345 675 99% 2165 Calculated as described by . H. Stickel et al. (1965a). HART OLDMON0029685 uu are known to be more accumulative than the lower ones in food-chains (Jensen et al. 1969)}and thus overall PCB loss rates would be expected to decrease proportionately as PCB levels increased in the body, resulting in disproportionately high estimated percentages of daily dose absorbed. Further research is needed concerning the physiological behavior of PCB's in avian tissues. The overall percentages (means of monthly rates) of total doses retained (overall means from. Table 7) were: DDE = 50%, DDT + TDE = l6$, DDT-complex = h6%, and apparent PCB = 6k%. The difference between DDT-complex and apparent PCB is presumably related to differences in accumulative behaviors of these residues. M: W. H. Stickel et al. (1965a) reported 16-20$ absorption (daily Mi'- HARTOLDMON0029686 \. . u$ . feeding on plant material. Dindal explains such delayed absorption or nonabsorption on the possible basis of adsorption of DDT molecules to nondigestable wax and cellulose materials of plants and possible phytosterol inhibition of normal intestinal absorption of cholesterol (citing Peterson 1951). These observations may partially explain one mechanism of many in the phenomenon of trophic concentration of chlorinated hydrocarbon residues in nature. Evidently, herring gulls have a high digestive efficiency, with few interferences such.as might be expected in herbivorous species. Perhaps a key observation in explaining the increased body burden associated with a decreased volume of part of the peripheral compartment in both adults and juveniles during phase 2 of the residue cycle is that of Dale et al. (1961). They observed that during starvation, increased concentrations of DDT~derived materials occurred in the fatty tissues of laboratory rats. Although augmented excretion of DDT-derived ' materials occurred at this time, residues were probably not lost at the same rate as fat, suggesting a change in residue metabolism at the time of lipid mobilization. Booth & Gillette (1962) demonstrated that increases in microsomal enzyme activity of rats administered certain - androgens were more closely related to the anabolic activity than to the androgenic activity of the steroids studied. In mammals, several halogenated hydrocarbons are known to enhance metabolism of steroids and drugs by hepatic microsomal..kvdrQXvl-ase-ind.uc.tion...although..separate- hydroxylase enzyme systems are probably involved for various reactions (Conney et al. 1967). Both DDE and DDT are known to cause large increases in liver weights of pigeons (Columba livia), to connect those 27 HARTOLDMON0029687 ' . 1*6 residues with hepatic function (Bailey et al. 1969a, 1969b), and Risebrough et al. (1968) have shown that'PCB's are stronger enzyme inducers than even DDT. The liver is assumed to be the major site of lipid metabolism in birds (Couch & Saloma 1968), but the connection between lipid-metabolizing enzyme systems and those associated with halogenated hydrocarbons is purely speculative. The general relationship between steroid metabolism and chlorinated hydrocarbon .metabolism has been established (see Conney 1967 and Kupfer 1967). It seems that several possible mechanisms are involved in the large body burden increases in spring herring gulls, probably relating chlorinated hydrocarbon metabolism to either lipid metabolism or steroid metabolism, or both. Steroid hormone levels are certainly highest during this period in adults as suggested by gonad sizes. Decreased activities of several foreign compounds, are known to occur as a result of competitive metabolism between certain steroids.and drugs by oxidative, liver microsomal enzymes (Tephly & Mannering 196b), with pregnant mammals and when certain steroid hormones are at high levels (Crawford & Rudolfsky 1966, Juehau & Fouts 1966). It is conceivable that during the period of increased gonadal activity in birds, the central compartment (of which liver is assumed a major component [Robinson et al. 1969]) becomes less efficient at chlorinated hydrocarbon metabolism and exchange, or in a sense "overloaded." The occurrence of increased body burdens in juveniles, however, which showed only very small gonadal increases, these Ts that other factors were also involved. Changes in lipid metabolism in both age classes which was closely related indrcnology to changes in residue dynamics suggest an intimate relationship between the two, as well. Any . . 28 HARTOLDMON0029688  . hi changes in enzyme induction and liver function -would likely change rate constants in the central compartment for various compounds as suggested by Robinson et al. (1969). Ibis then would change the kinetic relationships between the compartments, in the case of herring gulls, resulting in an increase in retention of already highly absorbed daily ,doses of residues. . Crawford & Rudolfsky (1966) also suggested that human subjects would be more susceptible to drugs during such periods of greater retention and decreased, metabolism. It follows, then, that herring gulls might be more sensitive to the toxic effects of the various chlorinated hydrocarbons at this time, as greater quantities of toxic materials would be expected to reach the central nervous system. L Extensive data on brain residues are badly needed from dead birds found on the breeding colonies in spring. Hickey et al. (1966) concluded that 2 of 3 dead adults picked up on the breeding colonies, and tested in 196b, had died of pesticide poisoning. The role of chlorinated hydrocarbons in possibly increasing normal adult mortality on Lake Michigan herring gulls needs to be elucidated. In summary, it can be seen that, under natural ecological conditions, the residue load in Lake Michigan herring gulls can . conceivably be a function of various potential factors, these factors being related to both the ecology and physiology of the gulls. The variables we have discussed above might be generally summarized_an_f_ollows.:---------_------ (l) seasonal changes in food-habits or residue levels in a given food source, or both; (2) possible seasonal changes in the amotints of food consumed, or 29 HARTOLDMON0029689 *gp . variations in daily intake-, U8 (3) changes in the amount and function of stored body lipids, or changes in compartmental volumes, thus potential reservoir space for organochlorine residues; (4) changes in the absorption, metabolism and excretion of chlorinated hydrocarbons as a result of changing lipid and steroid-hormone metabolism requirements placed on the liver (part of the central compartment); (5) different physiological behaviors' of the various chlorinated hydrocarbons present (as veil as possibly additive or synergistic . I' effects at varying concentrations not discussed) and (6) possible changes in intestinal absorption of CH residues (although a potential factor, we'found no suggestion of this). Additional ecological factors vhich might be related to one or more of the variables above might be seasonal variation in chlorinated hydrocarbon usage or fallout, or even potential increases in input of past-used chlorinated hydrocarbons due to spring runoff each year. Such ecological variables were suggested as one of several potential factors influencing residue levels in ring-necked pheasants (W. L. Anderson et al. 1970)* The variety of potential variables and their seemingly constant state of flux seem to point up the need for a systems-analysis approach as an aid in understanding the kinetics of pesticides. -------One point seems clear. THe~RerrIngr"gull population in Green Bay is severely burdened with various chlorinated hydrocarbon residues. Several sublethal effects of such residues are discussed by Keith (1966), Ludwig & Tomoff (1966) and Anderson et al. (1970). If one 30 HARTOLDMON0029690 U9 considers the average adult weight of Green Bay herring gulls as around 1100 g and the maximum body burden as 300 ppm (wet-weight) DDE and 200 ppm apparent PCB, the micrograms per bird come to around 3.3 -X 105 and 5 2.2 X 10' of DDE and apparent PCB, respectively. If the total breeding and non breeding population in Green Bay is 20,000 birds, then 6.6 X 10^ and 4i .4i X 109 are the yig's of DDE end apparent PCB present in the entire biomass of resident. Green Bay herring gulls. This amounts to 6.6 kg of DDE and 4.4 kg of apparent PCB, seemingly enough to have ecological consequences. ACKNOWLEDGMENTS Total financial support was provided for us as contractual research by the Bureau of Sport Fisheries and Wildlife, Fish and Wildlife Service, U.S. Department of the Interior, Patuxent Wildlife Research Center, Laurel, Maryland, U.S.A. E. H. Dustman, L. F. Stickel, J. A. Keith, and especially D. L. Hughes and R. W. Risebrough aided us in various aspects of the project. The Wisconsin Department of Natural Resources provided complete pen facilities of the State Game Farm. We are especially grateful to N. E. Damanske., J. 0. EVrard, P. J. Vinje, 0. A. Torgerson, H. C. Wilson, R. A. McCabe, Nita Hewins and Pearl Davis also assisted us importantly. . SUMMARY ' The dynamics of certain chlorinated hydrocarbons were studied over a period of one unnuar cycle lii" caged Juvenile and wild adult herring gulls from Lake Michigan. Fish, mostly alewives, comprised the major | year-round food items in the wild. Alewives were fed to the caged juveniles, and their residues averaged around 3 ppm (wet-weight) DDE, 31 HARTOLDMON0029691 . . 5o 2 ppm DDT + TDE and 2 ppm apparent PCB, In late fall and early winter, human refuse (including fish) became an additional food source. Lipid deposition, chlorinated hydrocarbon residues ajarcl body weight varied seasonally, as well as food consumption in juveniles, where data were available. Fat deposition in caged juveniles followed a pattern similar to that of adults except that amounts were greater. Lipid deposition was generally related to amount of food consumed in juveniles but residues were not. Juvenile residues showed a continual buildup and eventual, temporary stabilization..- Total body-burdens in both age classes were similar after these buildups by juveniles in their 9th month of life. The seasonal variations of residues of DDE and apparent PCB were characterized by two phases in adults and three in juveniles. Juveniles apparently assumed the adult pattern. The maximum body-burdens in both juveniles and adults were attained when lipids were being lost in spring and when adult gonadal activity was presumably at its maximum. This was followed by a return to "equilibrium." The maximum body burdens attained by caged juveniles on a diet of Lake Michigan alewives were 290 ag/kg DDE, 19 mg/kg DDT + TDE and 200 mg/kg apparent PCB. Residues in wild adults at the same time were 300,. k, and 200, respectively. Apparent PCB's. seemed to show a greater propensity for accumulation than DDE, although both were highly accumulative, DDE levels apparently resulted from dietary DDE as well as probable conversion from DDT. The extraordinary spring increases in body burden xrere attributed tochanges in the ability of the central compartment (of which liver is a part) to maintain "normal", "equilibrium"-type excretion rates of residues already in the body. ' 32 HARTOLDMON0029692 REFERENCES $1 Abou-Donia, M. B. & Menzel, D. B. (1968). The metabolism in vivo of l,l,l~trichloro-2,2~bis(p-chlorophenyl)ethane (DDT) and 1,1= dichloro-2,2~bis(p-chlorophenyl)etha,ne (DDD) and l,l~diehloro= 2,2-bis(p-chlorophenyl)ethylene (DDE), in the chick by embryonic infection and dietary ingestion. Biochem. Pharmac. IT, 2lk3~6l. Anderson, D. W,, Hickey, J. J., Risebrough, R. W., Hughes, D. F. [=L.] & Christensen, R. E. (1969). Significance of chlorinated hydrocarbon residues to breeding pelicans and'cormorants. Can. Field-Bat. 83, 91-112. Anderson, D. ., Hickey, J. J. & Keith, J. A. (1970), Chlorinated hydrocarbons and eggshell variation in herring gulls. Unpublished report to U.S. Fish & Wildl. Serv., 2k July 1970. Submitted to Auk. . ' Anderson, W. L,, Greenberg, R. E., Duzan, R. E. & Johnson, M. A. (1970). Concentrations and distributions-of p,p'-DDE, dielrin, and heptachlor epoxide in pheasants in east-central Illinois. Trans. 111. State Acad. Sci. 63, in press. Bagley, G. E., Reichel, . Li & Cromartie, E. (1970). Identification of polychlorinated biphenyls in two bald eagles by combined gas= liquid chromatography-mass spectrometry. J. Ass. off. analyt. Chem, 53, 251-61. '. Bailey, S., Bunyan, P. J., Rennison, B. D. & Taylor, A. (1969a). The metabolism of l,l~Di(p-chlbrophenyl)~2,2,2~trichloroethane and 1,1-Di(p-chlorophenyl)-2,2-dichloro~ethane in the pigeon. Toxic, appl. Pharmac. lk, 13-22. Bailey, S., Bunyan, P. J., Rennison, B. D. & Taylor, A. (1969b). The metabolism of l,l-Di(p-chlorophenyl)-2,2 dichloroethylene and 1, l-Di(p-chlorophenyl)~2-chloroethylene in the pigeon. Toxic, appl. Pharmac, Ik, 23-32. . Baldwin, S. P, & Kendeigh, S. C. (1938). Variations in the weight of birds. Auk, 55, kl6-6j. , Booth, J. & Gillette, J. R. (1962). The effect of anabolic steroids on drug metabolism by microsomal enzymes in rat liver. J. Pharmac. exptl. Therap. 137, 37^-9. Brenner, F. J. (1967). Seasonal correlations of reserve energy of the red-winged black-bird. Bird-Banding,. 38, 195-211. Bureau of Commercial Fisheries. (1966). Lake Michigan alewife situation: 1966, Ann Arbor. Child, G. I. (1969). A study of nonfat weights in migrating Swainson's thrushes (Hylocichla ustulata). Auk, 86, 327-38. Child, G. I. & Marshall, S. G. (1970). A method of estimating carcass fat and fat-free weights in migrant birds.from water content of specimens. Condor, 72, 116-9. Conney, A. H. (1967). Pharmacological implications of microsomal enzyme induction. Pharmac. Rev. 19, 317-66. Conney, A. H., Welch, R. M., Kuntzman, R. & Burns, J. J. (1967). Effects of pesticides on drug and steroid metabolism. Clinical Pharmac. Therap. 8, 2-10. ' Couch, J. R, & Saloma, A, E. (1968). Fat metabolism in the__la.ying-hon,-- . Texas Nutr. Conf. 23, 17*t~89. . Crawfordj J. S. & Rudofsky, S. (1966). Some alterations in the pattern of drug metabolism associated with pregnancy, oral contraceptives, HARTOLDMON0029694 r"'- and the newly-born. Br. J. Anaesth. 38, 446-54. 53 ' Dale, W. E., Gaines, T. B. & Hayes, W. J., Jr. (1962). Storage and excretion of DDT in starved rats. Toxic, appl. Pharmac.. 4, 89-106, Davis, E. A., Jr. (1955). Seasonal changes in the energy balance of the English sparrow. Auk, 72, 385-411. Dindal, D. L, (1970). Accumulation and excretion of Cl^ DDT in mallard and lesser scaup ducks. J. Wildl. Mgmt, 34, 74-92. Dwight, J. (1925). The gulls (Laridae) of the world; their plumages, moults, variations, relationships and distributions. Bull. Am. Mus. Nat. Hist. 52, 63-401. Ecobichon, D. J, & Saschenbrecker., P.(1968). Pharmacodynamic study . of DDT in cockrels. Can. J. Physiol. Pharmac. 46, 785-94. . Enderson, J. H. & Berger,, D. D. (1970). Pesticides: Eggshell thinning and lowered production of young in prairie falcons. BioScience,20, 355-6. Friend, M. & Trainer, D. 0. (1969). Aspergillosis in captive herring gulls. Bull. Wildl. Disease Ass. 5, 271-5. Fyfe, R. W., Campbell, J., Hayson, B. & Hodson, K. (3.969). Regional population declines.and organochlorine insecticides in Canadian prairie falcons. Can. Field-Nat. 83, 191-200. Gross, A. 0. (1940). The migration of Kent Island herring gulls. Bird-Banding,11, 129-55. . Hanson, H. C. (1962), The dynamics of condition -factnrr, in:Cn.r>nxin ---------- geese and their relation to seasonal stresses. Arctic Inst. M, Am. Tech. Paper 12. Hayes, W. J., Jr, (1959). Pharmacology and toxicology of DDT. Human . and Veterinary Medicine (Ed. by Sv W. Simmons), pp. 11-247, Basel HART OLDMON0029695 and Stuttgart. $h . Heath, R. G., Spann, J. W. & Kreitzer, J. F. (l969). Marked DDE' impairment of Mallard reproduction in controlled studies. Nature, Lond, 22l<, 1*7-8, Herman, S, G., Garrett, R, L. & Rudd, R. L. (1969). Pesticides and the western grebe: A study of pesticide survival and trophic concentration at Clear Lake, Lake County, California, Chemical Fallout: Current Research on Persistent Pesticides (Ed. by M. W. Miller & G. G. Berg), pp. 2^-53. Springfield, 111. Hickey, J. J., Keith, J. A. & Coon, F..B, (1966). An exploration of pesticides in a Lake Michigan ecosystem. Pesticides in the environment and their effects on wildlife. J. appl. Ecol. 3 (Suppl.), 11*1-54. Hickey, J. J. & Anderson, D. W. (1968), Chlorinated hydrocarbons and eggshell changes in raptorial and fish-eating birds. Science, N.Y. 162, 271-3. . Holmes, D, C., Simmons, J. H. & Tatton, J. O'G. (196?). Chlorinated hydrocarbons in British wildlife. Nature, Lond. 216, 227-9. Hunt, E. G, (1969), Pesticide residues in fish and wildlife of California. Peregrine Falcon Populations: Their Biology and Decline (Ed. by J. J. Hickey), pp. 455-60. Madison, Milwaukee and London. Ingolfs son, A. (1969). Sexual dimorphism of large gulls (Larns F.rm-1__ Auk, 86, 732-7. Jefferies, D. L. & Walker, C. H. (1966). Uptake of pp'-DDT and its post mortem breakdown in the avian liver. Nature, Lond. 212, ,533--^. HARTOLDMON0029696 IT''- 55 Jensen, S. (1966). Report of a new chemical hazard. New'Scient. 32, 612. Jensen, S,, Johnels, A. G., Olsson, M. & Otterlind, G. (1969). DDT and PCB in marine animals from Swedish waters. Nature, Lond. 22k, 2i7-50. .. Juchau, M. R. & Pouts, J. R. (1966). Effects of norethynodrel and progesterone on hepatic microsomal drug-metabolizing enzyme systems. Biochem. Pharmac. 15, 89I-8. , Kabat, C., Meyer, R. K., Flakas, K. G. & Iline, R. L. (1956). Seasonal variation in stress resistance and survival in the hen pheasant. Wisconsin Conserv. Dept. Tech. Wildl. Bull. 13. Keith, J, A. (l966). Reproduction in a population of herring gulls (Larus argentatus) contaminated by DDT. Pesticides in the environment and their effects on wildlife. J. appl. Ecol. 3 (Suppl.), 57-70. King, J, R,, Farner, D. S. & Morton, M. L, (1965). The lipid reserves of white-crowned sparrows on the breeding ground in central Alaska. Auk, 82, 236-52. Koeman, J. H., Ten Noever de Brauw, M. C. & De Vos, R. H. (l969). . Chlorinated.biphenyls in fish, mussels and birds from the River Rhine and the Netherlands Coastal area. Nature, Lond. 221, 1126-8. Kupfer, D. {1967). Effects of some pesticides and related compounds on steroid function and metabolism. Residue Rev. 19, 11-30. Eaug, E. P., Nelson, A. A., Fitzhugh, 0. G, & Kunze, F. M. (1950), Liver cell alteration and DDT storage, in the fat of the rat induced by dietary levels of 1 and 50 ppm DDT. J. Pharmac. exptl. Therap. HARTOLDMON0029697 98, 268-73. 56 Lichtenstein, E. P., Schulz, K. ft., Fuhremann, T. W. & Liang, T, T. (1969). J. econ. Ent. 62, 761-5. Ludwig, J. P. (1966). Herring and ring-hilled gull populations of the Great Lakes I.96O-I965. Great Lakes Res. Div. Publ. 15', 80-9. Ludwig, J. P. & Toraoff, C. S. (1966). Reproductive success and insecticide residues in Lake Michigan herring gulls. Jack-Pine Warbler, 1*4, 7T-8L - Kelson, E. (1961). Kinetics of dr\ig absorption, distribution, metabolism, and excretion. J. Pharmaceut. Sci. 50, 181-92, Horden, C, ft. (1966). Age, growth, and fecundity of the alewife in Lake Michigan. Midwest Fish & Wildl. Conf. 28, 34-5. Marshall, A. J. (1961). Breeding seasons and migration. Biology and Comparative Physiology of Birds. Vol. 2. (Ed. by A. J. Marshall), pp. 307-339. New York and London, - [Mrak, E. M., et al.] (1969). Report of the Secretary's Commission on Pesticides and Their Relationship to Environmental ffealth. Parts 1 & 2. U.S. Government Printing Office, Wash. Oduin, E. P. & Connell, C. E. (1956). Lipid levels in migrating birds. Science, N.Y. 123, 892-4.' .' . Peterson, D. W. (1951)* Effect of soybean sterols in the diet on the plasma and liver cholesterol in chicks. Proe. SOe. Exptl. Biol. Med. 78, 143-7. Raveling, D. G. & LeFebvre, E. A. (1967). Energy metabolism and theoretical flight range of birds. Bird-Banding, 38, 97-113. Reinert, R. E. (1970). Pesticide concentrations in Great Lakes fish. HARTOLDMON0029698 $1 Pest. Monit, J. 3, 233-40. . Reynolds, L. M. (1969). Folychlorobiphenyls (PCB's) and their interference with pesticide residue analysis. Bull, envir. Contain. Toxicol. (U.S.), It, 128--43. ' Reynolds, L. M, (1970). Pesticide residue analysis in the presence of polychlorobiphenyls (PCB's). Residue Rev. 34, in press. Risebrough, R. W., Reiche, P., Peakall, P. B., Herman, S. G. & Kirven, M. N. (1968). Polychlorinated biphenyls in the global ecosystem. Nature, Lond. 220, 1098-102. Risebrough, R. W,, Reiche, P. & Olcott, H. S. (1969). Current progress in determination of the polychlorinated biphenyls. Bull, envir. Contam. Toxicol. (U.S.), 4, 192-201. Risebrough, R. W., Davis, J. D. & Anderson, D. W. (1970). Effects of various chlorinated hydrocarbons. The biological impact of pesticides in the environment. 1969 Oregon State Univ. Symposium, in press. . Robinson, J, (1967). Dynamics of organochlorine insecticides in vertebrates and ecosystems. Nature, Bond..215, 33-5. Robinson, J., Richardson, A., Crabtree, A. N., Coulson, J. C. & Potts, G. R. (1987). Organochlorine residues in marine organisms. Nature, Lond. 214, 1307-11. . Robinson, J. & Roberts, Mi (1968). Accumulation, distribution and elimination of organochlprine insecticides by vertebrates. Soc. . Chem. Ind. Monogr. 29, 106-19. Robinson, J., Roberts, M., Baldwin, M. & Walker, A. X. T. (1969). The pharmacokinetics of HEOD (dieldrin) * in the rat. Pd Cosmet. Toxicol. 7, 317-32. ' HARTOLDMON0029699 58 Simpson, G. G., Roe, A. & Lewontin, R. C. (i960). Quantitative Zoology. New York, Chicago, San Franciso and Atlanta. Smith, W. J. (1959)- Movements of Michigan herring gulls. Bird= Banding, 30, 69-104. Steel, R. G. D, & Torrie, J. H. (i960). Principles and Procedures of Statistics: with Special Reference to the Biological Sciences. New York, Toronto and London. . . Stickel, L. F,, Chura, N. J., Stewart, P. A., Menzie, C. M., Prouty, R. M. & Reichel, W. L. (1966). Bald eagle pesticide relations. Trans. N. Am. Wildl. Nat. Res. Cohf. 31, 190-200. . Stickel, W. H., Hayne, D. W. & Stickel, L. F. (1965a). Effects of heptachlor-contaminated earthworms on woodcocks. J. Wildl. Mgmt, 29, 132-t6. . Stickel, W. H., Dodge, W. E., Sheldon, W. G., DeWitt, J. B. & Stickel, L. F. (I965h). Body condition and response to pesticides in woodcocks. J. Wildl, Mgmt, 29, 147-55. Tephly, T. R. & Mannering, G. J, (1964). Inhibition of microsomal drug metabolism by steroid hormones. Pharmacologist, 6, 186, Tinbergen, N. (i960). The Herring Gull's World: A Study of the Social Behavior of Birds, New York. U.S. Food & Drug Admin. (1965-68), Pesticide Analytical Manual, Vol.. . 1. U.S. Dept. Health, Education & Welfare, n.p. Veith, G. D. & G. F. Lee. (1970). Chlorinated organic contaminants .in the Milwaukee River. Prog. Rep. to Wise. Dept. Nat. Res., June 1970. Weise, C. M. (1963). Annual physiological cycles in captive birds of ' differing migratory habits. Proc. Int. Ornith. Cong. 13, 983-93. HARTOLDMON0029700 HARTOLDMON0029701 6o CHLORINATED HYDROCARBONS AND EGGSHELL VARIATION IN HERRING GULLS DANIEL W. ANDERSON, JOSEPH J. HICKEY, AND J. A. KEITH Our objectives here are to describe normal geographic variation in several eggshell measurements of North American Herring Gulls (Larus argentatus), to report recent eggshell changes in various parts of their breeding range, and to relate these changes to residues of 'Chlorinated hydrocarbons resulting from environmental pollution. It was. in 1946 that eggshell changes began in Peregrine Falcons (Falco peregrinus) ' and Eurasian Sparrow Hawks (Accipiter nisus) in Great Britain (Ratcliffe, 1970). We found that these changes occurred in North America as early as 1947 (Hickey and Andersen, 1968)J we lacked the adequate sample of 1946 eggs with which Ratcliffe worked. Ratcliffe (1970) has convincingly discussed the causal link between organochlorine residues in general and recent eggshell thinning for a substantial list: 'of British birds; and experimental verification of the link has been supplied by Heath et al. (1969) and Porter and Wiemcyer (1969). The problem of noratal variation in large Larus gulls is by no feeans simple, as all reviewers of this group have already brought out, Tngolfsson (1969) has discussed sexual dimorphism in Herring and other I. / ' gulls; and because of this, all measurements discussed below are those of females. Dwight (1925: 182) described intersubspecific variation Ln Herring Gulls as follows (wing, tarsus, and culmen measurements, respectively, for each subspecies in mm): argentatus 384, 58, and . ` ' *7: smithsonianus 411, 62, and 50; vegae - 422, 66, and 51; "thayeri" = ' ?95, 60, and 48, Thayer's Gull is.considered as a subspecies of Larus. 1 HARTOLDMON0029702 ' 6i argentatus by the A.O.U, Chock-list (1957: 222), but some authors (Rand, 1942; Salomonson, 1950: 319-321; Macpherson, 1961; and Smith, 1966: 5-7) treat it as other than L. argentatus. Snyder (1957: 214) and Moynihan (1959) suggested that thayeri might be a distinct species. In the present paper, we have included the few specimens we examined and have followed the terminology used by Macpherson (1961) and Smith (1966), Dwight (1925:194) considered vegae as the largest of the Herring Gulls.Perhaps a significant observation was that of Snyder (1957: 213), who reported that thayeri averaged slightly smaller than argentatus in the arctic. Smith (1966: 12, 17, 22) gives the above measurements for arctic specimens of Herring Gull from three locations as 419, 63, and 51-52 mm, suggesting that North American Herring Gulls may be slightly larger in the north (compared to Dwight's specimens cited above, wbicli judging from his range description came from south of Hudson Bay), Voous (1959) established that Herring Gulls from Great Britain- were smaller in winglength (sexes combined) than North American specimens (10) of smithsonianus, but his specimens from Fennoscandia were as largei or larger. Harris |, (1964) reported egg volumes of Herring Gulls from Wales to average ground 76 cc, but volumes calculated from the data of Paludan (1951) |;1 averaged around 86 cc. Paludan studied Herring Gulls near Bornholm, ci,I. on the Baltic Sea. These data support Voous (1959), who demonstrated that geographical variation in the European Herring Gull was clinal in nature* The breeding ranges of the North American gulls described pabove are given in Figure 1 to show the geographical relationships our various test groups. This map was derived from the A.O.U. Iheck-list (1957: 221-222), Macpherson (1961), mostly Smith (1966: HARTOLDMON0029703 62 7, 91) and only generally from Voous (1960; 147), who considers all North American forms (save possibly vegae, if it breeds in North America) as' argentatus. Bent (1921; 124) doubts the validity of Alaskan records of vegae, although such records do not seem unlikely. We have accepted two sets of eggs from St, Lawrence Island because of their proximity to the range of vegae as described by Dement'ev and Gladkov (1951: 531), Some range descriptions of vegae (Dwight, 1925: 181; and A.O.U. ' Check-list 1957: 222) mention only Siberia, , DATA SOURCES MATERIALS AND METHODS , The measurements described herein for Herring Gulls comprise those of scientifically collected eggshells from about 65 major museums and private egg collections representing a large proportion of North American collections. The methods we used to measure eggshells and most of our data sources have already been outlined by Anderson and Hickey (1970). Additional details concerning our methods are in Anderson et al, (1970), Data concerning other species of larids were all obtained from the oological collection in the Field Museum of Natural History, Chicago, Illinois; these represent samples from small geographical ,areas. Other larid eggshells examined included those of Sabine's |Gull (Xema sabini), Franklin's Gull (L, pipixean), Mew Gull- (L, eanus), pleermann's Gull (L. heermanni), Ring-billed Gull (L, delawarensis), plifornia Gull (L. californicus), Western Gull (L. occidentalis), Shd Glaucous Gull (L. hyperboreus). Eggshells from two species of |eruvian larids (Peruvian Gull, L, belcheri and Dominican Gull, JL, 3 HARTOLDMON0029704 63 Figure 1.. Breeding range of North American Herring Gulls as determined from various authors. The general geographical areas used to test for geographic variation in eggshells are numbered in accordance with Table 1. Closed circles represent those areas where samples were obtained for chemical analysis. HARTOLDMON0029705 '6h domlnlcanus) wore collected by Anderson in 1969. These were plotted but not included in any statistical analyses except for Figure 3. ' Our recent Herring Gull egg specimens, besides being found in museums and private collections, represented materials collected by us in the course of our studies on Lake Michigan or by cooper.ators in other areas (Figure 1), Our specimens from 1967 from which we obtained eggshell and residue data are representative of gulleries ' from Knife River, Minnesota (Lake Superior) and Sister Bay, Wisconsin (Lake Michigan) (collected by us), Roger's City, Michigan (Lake Huron) (collected by J. T. Kmlen, Jr. and D. H. Thompson), Block Island, Rhode Island (collected by M. E. Slate), and. Penobscot Bay, Maine (collected by W. H. Drury), T. C. Grubb collected our 1969 Herring Gull eggs from Kent Island, New Brunswick. Residues of ,j>'"DDE and ' their relationships to shell thickness, on a colony-to^colony comparison, have already been reported by Hickey and Anderson (1968) but will be 5" further and in more detail related to overall contamination profiles below. DATA TREATMENT ' Statistical methods generally followed those outlined by Steel ,and Torrie (1960), Our general procedure was to review the literature on geographic variation in Herring Gulls and then to test trends in Eggshell volume, empty shell weight, eggshell thickness (including |y ^membranes), egg shape, and length/breadth ratios generally against u:,. _ already described trends, if available, in body size or some measurement fo-S ' . relating to it. Most of our measurements in other species have been .4 HARTOLDMON0029706 65 shown to be a satisfactory general index to body size (discussed by Anderson and Hickey, 1970, in relation to Brown Pelicans, Pelecanus occldentalis; and Anderson et al., 1970, in relation to Common Loons, Gavla immer). In Great Lakes and Atlantic Coast data series we first tested various groups for thinning as they fell into decades (1900 09, 1910-19, etc.). This test showed only highly significant differences in the decades after 1940. However, this did not indicate precisely when changes occurred. Post-1946 data were then broken down into as small groups as possible depending solely on sample distributions. They were not selected except that smaller groups were set up wherever feasible. Consistent nonsignificant differences in pre-1946, decade samples were considered as justification for their eventual combination and assumed biological consistency. . CHEMICAL ANALYSES The Wisconsin Alumni Research Foundation (WARS' Institute, Inc., Madison, Wisconsin) conducted our chemical analyses, using electroncapture gas chromatography (GG). The chromatographic interpretations concerning polychlorinated biphenyls (PCBs) are ours, but are generally based on methods described by Risebrough et al. (1969) and Reynolds (1970), WAKF chemists employed the methods outlined by the U.S, Food and Drug Adininistration (1965-68) for analysis of chlorinated-hydrocarbon |esticides. Further modifications used for us by WARF to confirm -DDT and 2.*2?-TDK are described by Anderson et al. (1969), and pP interpretations regarding PCBs are described by Anderson and i-ekey (in preparation). The compounds in our analyses included HARTOLDMON0029707 66 .*,,~'W)^> apparent PCB, jj^'-TDE, ,jd'~DDT, heptachlor epoxide (HE), and dieldrin. All but PCBs are known to be insecticides or insecticide" derived materials. Further data on PCB sources and uses as well as analytical techniques are reviewed by Reynolds (1970). At least one PCB peak is known to potentially interfere with ,*DDE on the DC--200 column of the GC in Lake Michigan Herring Gull samples, but interference is reduced to a minimum by the methods WARE used for quantification (dilution) and by the very high proportions of DDE present. Interference is potentially more critical, however, in samples such as those, we analyzed from Maine and Rhode Island, where i DDE was present in much less amounts in relation to apparent PCB, compared to our Great Lakes samples. We found that standard, peakheight quantifications of ,'-DDE had to be multiplied by' a factor of 0.80 for Rhode Island specimens and by 0.86 for Maine specimens, assuming that the "hidden" peak was constant in relation to the PCB peak that we could see just preceding DDE in retention time. Because of the potential variability on a ppm wet-weight basis due to possible desication in addled eggs and water losses in incubated eggs, we express' here all residues on a ppm lipid basis unless otherwise specified. We consider such values as representative of concentrations in, and possibly affecting, the female at the time of eggshell deposition (Enderson and Berger, 1970; Vermeer and Reynolds, 1970; and others). RESULTS AND DISCUSSION . GEOGRAPHIC VARIATION The intersubspecific variations in several eggshell measurements HARTOLDMON0029708 F . . . 67 \ (Table 1 and Figure 2) seemed generally to follow other variations in Herring Gulls, as described above. These variations are most likely clinal in nature, as discussed by Voous (1959) and Mayr (1965: 361-364), The geographical subdivisions suggest a cline, but also evident in all areas is the high degree of variance which tends to mask differences over the entire range (Figure 2). Where we had larger samples, however, differences became more apparent. Because of the apparently high variance in Herring Gull eggshells, especially in relation to those in other species we have studied, it seems that larger samples might be needed to describe differences for this species more precisely. It may be that variations within individual females (Harris, 1964; Coulson et al,, 1969; Vermeer, 1969) tend to mask or reduce significance of differences between local populations. An analysis of variance to test whether differences did or did not occur in the North American "populations" (or geographical test units) gave highly significant F-values: - 12.21, F^ 7.51, and fch - 7.37, when (0.005) = 2.90, A Duncan's new multiple-range test for differences in eggshell volumes of the North American test units of Herring Gulls showed clinal overlap but suggested the distinctness of northern, Atlantic-Interior, and Great Lakes populations; the data were not uniform enough nor sample sizes large enough for any further distinctions (means not significantly different from one another are underscored, reading from left to right): Arctic Interior St. Lawrence Atlantic Great Lakes HARTOLDMON0029709 'W HART OLDMON0029710 r~i H rH Oo 69 m p0 W f3t 0 f0t) c3 ft v> ag> s f0t P 0 -P 0 * .w fftt CO * & <D f0t) P H S 0 fftt w >0 03 *73 O +> (4 O fi 0 'Q d <U > id ,<U (4 H Ui O *u fat> w ti 0 R H H M H KS O k 43 fdt +J O O `H -P f0t) <L> H 0) ft <D <U to vftj H Ci O w H U XS O O #\ <10/5 0 H g 3 TABLE 1 c o n tin u e d fd3t fwt 03 O<M-- <i> uis cO \A p 0 d 03 u 03 J* P fttd P 0.3 0 fpOt p 0 fftt d 0 0 ' >> ft R 03 d03 bo *d 03 0 fdt P d 0 0 ft wft Eh HARTOLDMON0029711 h HARTOLDMON0029712 r-- 71 Figure 2. Frequency distributions of eggshell weights of North American Herring Gulls from various geographical regions. A = arctic-taiga, B = interior, C = Great Lakes, D - St. Lawrence River system, E = Atlantic Coast. ' HARTOLDMON0029713 72 Because of the likelihood of dines, we included intermediate .areas in our presentation of the data (Table 1), although they were not significantly different from adjacent areas. Banding data, reported by Gross (1940), suggest that Kent Island Herring Gulls tended to "cling" to the Atlantic Coast, whereas Great Lakes birds tended to follow lake shores and water courses. These data support the idea of at least limited population isolation. Smith (1959) further reported that Great Lakes birds followed somewhat distinct dispersal patterns and that, after their second year, most birds tended to remain on the 'Great Lakes year-round within 300 miles of. their breeding colony. Olsson (1958; 151-155) reported similar findings for Herring Gulls from Fennoscandia. Data are very much needed on interior and arctic Herring Gulls from North America. INTER- AND INIRASPECIFIC EGG RELATIONSHIPS Lack (1966; 293) believes that relative egg size is a specific characteristic and mainly a matter of heredity on an interspecific basis. Eggshell volume in the twelve species of larids that we examined .-relates closely to our index of body size (tarsus X culmen, in mm) ;on inter- and very likely on intraspecific bases (Figure 3). Amadon .(1943) pointed out that egg size, was generally a nonlinear function Jof body size in small birds, as expressed by the equation, Y bx2t f"where a is a constant expressing the ratio of egg size versus body size ^increase in a aeries. He found to be constant at the "intermediate" |taxonomic category (subfamily) in small birds. If one assumes a similar i" ' '' ' - jjponstancy of a in the larids we examined, the straight-line relationship n a nonlog scale in Figure 3 seems reasonable. Since the 95% confidence 8 HARTOLDMON0029714 73 Figure 3. Eggshell volume as a general function of index to body size (in mm), in 12 species of larids and 3.subspecies of Herring Gulls (VEG = vegae, SMI smithsonianus. and AEG = argentatus). The correlation coefficient (r) was significant at P<0.001. Additional abbreviations are as follows: SAB Sabine's Cull, FRA Franklin's Gull, MEW = Mew Gull, HEE = Heermann's Gull, R~B Ring-billed Gull, CAL California Gull, PER Peruvian Gull, HER* Herring Gull, THA = Thayer's Gull, DOM Dominican Gull, WES Western Gull, GLA * Glaucous Gull. i i TARSUS X CULMEN HARTOLDMONOQ29715 7h limits (C L) of the intercept overlapped zero considerably, it was not significantly different from zero (Figure 3), and the simple equation, Y_ ~ bX, seems applicable, Jb- being the general constant relating eggshell volume to body size in larids. Additional factors, such as clutch-size variation deviating from the familial mean trend, might be expected to slightly change the value of a for a given species. In this instance, our small sample of Thayer's Gull fell farthest from the fitted line. The general relationship and the closeness of fit for the remainder of the species (Figure 3) further support the view that eggshell volume (and thus likely egg weight) is fairly constantly related to body size in closely related forms'(Araadon, 1943). Eggshell thicknesses and weights were closely related to their eggshell volumes in all the species we examined (, comparing volume and thickness 0,968, P< 0,001, $ 0,0025X_+ 0,183; _, comparing volume and eggshell weight = 0,995, 1?^0.001, Y - 0.079X + 0.387). RECENT EGGSHELL CHANGES Ratcliffe (1970) has documented that eggshell changes have occurred since 1945-46 in certain species of British birds. He validly divided test periods into "late" and "early" (Mountford, 1970), presenting pre- and post-1946 eggshells to demonstrate this phenomenon. We obtained data on 295 post-1946 eggshells. Demonstrable eggshell changes occurred in most of our more recent i- - samples, although many represented what we considered to be minor changes, that is, decreases of not over 10-15% (Table 2), Our best histories came from Lake Michigan and Lake Huron, On the latter, 1947-4$ eggs showed no changes, but by 1951-58 changes were detectable. They HARTOLDMON0029716 CHANGES IN INTACT EGGSHELLS OF NORTH AMERICAN HERRING GULLS SINCE 19^6 'ABLE 2 c o n tin u e d 75 CM /WIt t>0 V. g Ct1O va. tI VC V. o I rH H CO1 vs. O1 n 1 1 &1 O rH -'"'N id %wfc-? O 0 H j to f + CO rH + + o 00 o -4 f+ ++ On -5f On * 4- 1 t ot -4* O l , H -4 o -4" -4" CO eg m +>l 1 M o H & +> q M o fc-- On CO On -4 CO ir\ O CO NO t- ir\ CO .4 co CO <o ro * * s; . o o o o o o CO CO * o rH H 1 M C/3 c3 pce3 td o CO .4 On fr-- o CO *ti "d 7% H rH rH rH oO CM b0O> MVOCI tI- VI Vb VC CMI MOI CiO tI- Vs. CMI VS. T VcC~I 6 bD HJ * ** o CO 4 -4 eg CM * ** ON ON CO IfN 4 O H o CM rH CM' 4* 1" t-- CO LA fc- NO o CO H eg ~=t rH NO CO O o OH ON c~ On NO * V0 LA IA IA IfN LTN eg 4 -4* ON On o H rH rH rH rH H t-- o CO On IfN VO * * LA IA IA On On VO r-i eg CO CO MO VO VO i 4 1 IA ii MO11 IA 1 4 H o 17" cn o CM LA MO 4 IA MO VO On On On ON ON ON LA MO MO ON On ON H H H rH rH tH H H rH ou H <-oOuPH . . <D CO ^ o oq (O4 qo s 1 wS. S w # >0 q iH 9bO 4Orf 4OH *H sH; * 4 iH' HART OLDMON0029717 76 Vi V*. Vi Is. V. V. LA CO H O O CO H rH H rH H 1 4- + H* f 4- + on On VO LA VO ON O On LA o o c~ a LA LA H vo H c~ b-- LA On CO A H O CO CO (O CO co * CO 9 CO CO CO o o o O oo on 00 CU ro On rH CM A Oi Vi vs. CO Ov VL ON CO VI VS. On t*--- r~t H 1 1 11 1* + T4- + t-- o OJ 0\ On CO CO OJ + f 4 H C~ CO HrH O CM A CO O cO VO o On A 9 IA A vo A IA LA A IA o\ a CM H CO ON CM A CM -zt A VO t-- CO On VO VO VO VO VO MO On ON On ON ON ON rH H rH H H H CM Vi CM H* Vi HI VCOIi OLA O i rH II <D rPH -f oS * rH rH CO t*- EH CO rH rH IA t-- > 4 o rH pH o. cn rH rH &O <D O CO ON CO t-- VO CO r* ^f rH CO CO CO CO <o tsO cd o o o o (0 K$ od CD -P tO cd do <d *rl h d Q co On ON cd rH rH H CD O H <d XZ Q* CD do Ud Vi Vi vs. bQ CD rH o rH CM O k 1 rH rH CD 1 iw o iH p rd d H t H- O MO 1" + ON H* VO & 0 to CM * o * ro * vo P CD H O o CO CO -p cd H H 3 H o -t o OJ vo eu cd CM CO * vo * A * *+H rg o NO NO IA A cd b P Ai cd *d 5! o CO On On XZ rH rH sr *p ON rH 1 <H O d M P- CO H O xz !> t*~ t-- (O On p o \o MO VD VO H rH ON On ON On > H H H rH a> CD o r> g #-h o H (0 Po Mo m sS rH W rtHo HI * H <D ro a> P m P pq O B S! B d rd a> Pi Ui 8m o <rHw-t t/7 Si WH & oO P o07 H ra B "S oo CM 9 TABLE 2 c o n tin u e d HARTOLDMON0029718 77 HARTOLDMON0029719 TABLE 2 c o n tin u e d * .. . , 78 seem to have remained at generally less than a 10% change. The changes on Lake Michigan prove even more interesting (Table 2), They seem to have paralleled Lake Huron into the early 1960s although we have no data for 1961 or 1963, In 1964 and 1965, eggshell changes were 15% or greater. It is perhaps significant that this period was one of frequent and readily observed egg-breakage on Lake Michigan study areas (Keith [1966] for 1964 and Ludwig and Tomoff [1966] for 1965, each on opposite shores of Lake Michigan). Eggshell fragments from three broken eggs incidentally picked up by Keith in 1965 measured 0.16, 0.28, and 0.34 mm, averaging 0.26 mm, 31% thinner than pre-1946 eggshells. Nine broken or cracked eggs collected by Anderson in 1968 averaged 0.34 mm, not different from our random sample of the colony that year (Table 2.), Data are critically needed on frequencies of egg breakage at different eggshell -thicknesses. No precise quantitative data of this type were obtained in our studies, but in 1969, when egg breakage xis the least seen by us in 4 years of visits to the colonies, about 5-10% of the nests contained obviously broken eggs (other than eggshells possibly broken open by predation), and about 20% of the nests had loose, uncrushed eggs nearby. Keith (1966) estimated readily observable egg breakage at about 11-24% in 1964. In general, it seems that eggshell thinning on the order of and greater than 15-20% is associated with readily observable breakage. Eggshells of California Brown Pelicans (including the relatively constant membranes) averaged 53% thinner than "normal" in 1969, and such changes were associated with obvious, virtually total failure in reproduction (Risebrough et al., 1970 and J. 0. Keith et al., 1970). Ratcliffe (1970) reported that eggshell changes roughly on the order of 10-20% were associated with declines in populations of some British raptors, 10 ` HARTOLDMON0029720 F"' 19 but such eggshells represented those that had "survived" to be taken by egg collectors, Hickey and Anderson (1968) showed changes of roughly 18-25% circumstantially associated with declining species of American raptors. Both studies associated eggshell thinning with reported frequent egg breakage, lowered reproductive success, and high residues of pesticides. The 1964 levels of ,*-IJDE reported by Keith (1966) for Lake Michigan Herring Gull eggs are the highest reported in 7 years of monitoring the same population. Levels were still high, however, in relation to other North American Herring Gulls in 1967 (Hickey and Anderson, 1968), Eggshell breakage still continues on Green Bay, but reproductive success is apparently high enough to produce reduced but bandable numbers of young in recent years (II. C, Wilson, personal communication). Whether or not this population is presently reproducing enough to maintain stationary numbers is a matter open to question; it was not in 1964 (Keith,, 1966). - Keith reported a mean clutch size of 2,11 as compared to our best estimate of 2.72 in 1968 and 2.81 from southern, Atlantic, and interior egg data. Although there is a tendency for Herring Gulls to return to their natal colony (Ludwig, 1963), there is no reason to doubt, if one segment of the population is declining, colonization by gulls from other, more successful, areas of the Great Lakes. Herring Gull control in Massachusetts colonies was generally unsuccessful from 1942 to 1952, presumably because of such immigrations from more successful colonies outside the state (Drury, 1963). Great Lakes Herring Gulls have generally undergone a recent population increase, probably due to a recently developed superabundant food source, the alewife (Alosa pseudohaneugus)(Ludwig. 1966), which would tend to lessen the normal 11 HARTOLDMON0029721 80 regulating effects of limited food supply (Lack, 1954: 141-153). Our Atlantic samples suggested only local eggshell changes on the East Coast (Table 2). Since these populations have been under control measures since the early 1940s--they are also the beneficiaries of a superabundant food resource, human or human-associated refuse--and have increased in recent years (Gross, 1951; Drury, 1963; Kadlec and Drury, 1968), the interpretations of shell-thinning become nearly impossible. Eggshell data suggest, nonetheless, the presence of only low amounts of thin-eggshell-inducing compounds such as j^jj'-DDE (see Heath et al., 1969) in most of the samples. Decreased eggshell weights or thicknesses from Kent Island (Table 2) may even reflect a higher freqxiency of second clutches in these birds, due to the control measures. Residue analyses are needed from Kent Island before any conclusions can be made regarding eggshell changes in that area. CHLORINATED HYDROCARBON RESIDUES, 196? ' Highly significant (P<0.005) differences (analysis of variance to test whether or not geographical residue-differences were present) were found for DDE, DDT + TDE, apparent PCB, HE, and dieldrin as well as. eggshell weights and thicknesses. DDE/PCB ratios showed the same highly significant differences,which suggest varying contamination patterns, especially when one compares Atlantic and Great Lakes populations (Table 3, Figure 4). Apparent PCB increased in relation to DDE in the East? although levels of both decreased considerably- outside the Great Lakes (Table 3). The calculated mean egg volume (+ S.E.) for .the three colonies from the Great Lakes was 85.4 +_ 1.3 em^. That for three Atlantic colonies (1969 Kent Id. sample added) was 87.5 + 1.3 em^. ' 12 HARTOLDMON0029722 81 \ HARTOLDMON0029723 was ; TABLE 3 c o n tin u e d . \ H o. 4-1 -4* o ON H o. 4*1 CO lf\ o CM on o. +1 o CO MO CO o +1 -=r t- * H OV o o o + Ov C" cn 4 o o vo* H o -- r--1 ~=t cy V0 VO CO oo O *4 O o, o, o. Of 4* i + 4co + 1 VO H COi VO VO CO * o i/\ rH o ?i r4 LA rH O i-p d H 3 ---... ...... - - -......... r4 CO LA ON O H CO O O, O, O, O. 4* 4* 4*1 4-1 ON O 4-^f ON O VO LA CO t~* 1 H ; IA r4 O ; . " . H. <D . O Cl <L> nd H Cl 0 0 vs. IA ' Ov nX *H '" # !> <"*s H 0 xS cl ON H r4 xax> #v -p VO ON Sh H O g <D 3 + 1 a am r4 a> O V~N 3 xi P feG col CM | o H p 41 tiO u .0 fd ---- . g i O bD P< P cd H a> a H wM* j jg to Jh 41 w CO to U 0 H to CO PQ H <D a> P 0 <l> d a u d< 41 M 3 O H N* to bO *rO*i 0 H O 41 P Q to 41 <u n P3 EH p4 <u u gJ tli 2 g rH flj > h[ p n 3 ti a> p g & 0 0 0 ***, cm! Cd PI r*> P K P M a? p co| TABLE 3 continued 82 HARTOLDMON0029724 83 Figure h. Variations in chromatographic patterns of extracts from Herring Gull eggs representing different colonies. These chromatograms represent injections of different dilutions and are meant only to show qualitative differences between DDE and the numbered apparent PCB peaks in general linear range of response of the electron-capture detector. A *= Rhode Island, B end C = Maine (B suggests the possible presence of aldrin), D = Lake Huron, E = Lake Michigan, and F = Lake Superior. HARTOLDMON0029725 . . 8h These are not unlike the pre~1946 eggshell volumes (Table 1) and suggest (1) that eggshell changes did not importantly involve changes in volume and (2) that geographical variation in volume can therefore probably still be detected with adequate sample sizes. When correlation matrices were run with data on individual eggs, we found significant negative ' relationships between DDE, DDT + TDE, and apparent PCB and eggshell thickness and other measurements relating to thickness (Table 4). This, of course, casts doubt as to which residue might be most importantly involved, or if all are not involved' to some degree. As DDT + TDE was related to DDE (r -- 0.537), their relationship to eggshell thickness may reflect a cause-effect relationship or a general association with "DDT-family" residues. Heath et al. (1969) have experimentally produced eggshell thinning in the Mallard (Anas platyrhynehos) with a low dietary Intake of DDE, and Porter and Wiemeyer (1969) have obtained similar effects in the American Sparrow Hawk (Falco sparvarius) with p,.o'-DDT' combined with dieldrin. Bitman (1970) and Bitman et al. (1969) have also shown a similar effect with both o^'-DDT and jg^p'-DDT in Japanese Quail (Coturnix coturnix), and Lehner and Egbert (1969) have demonstrated minor and levelling eggshell changes in Mallards with dieldrin alone, On the other hand, Dahlgren and Linder (1970) were unable to induce eggshell thinning in Ring-necked Pheasants (Phasianus colchicus) with dieldrin,.nor did injections of dieldrin succeed with Ringdoves (Streptopelia risoria)(Peakall, 1970a), Peakall did, however, induce significant eggshell thinning in Ringdoves injected with j> ,*-DDE, In attempting to formulate a hypothesis from our field data, and to eliminate individual variability, we arbitrarily set up six class intervals of shell thickness and calculated weighted means for each; 13 ' HARTOLDMON0029726 8$ TABLE It CORRELATIONS BETWEEN VARIOUS EGGSHELL VARIABLES AND RESIDUES OF DDT-CQMPLEX AND APPARENT PCB Residue (ppm lipid basis )-*- Variable R,E'~M>E E,p_'-DDT + ,e--tde Apparent PCB Eggshell wt, (g) Thickness (mm) 2 Thickness index -0.484**# -0.646*#* -0.505*** -0.302** -0.427*** -0.309#* -0.555*** -0.700*** -0.534*** Significance levels are marked as follows: . .0 001 ** = PcO.Ol, ##* = P < Devised by Ratcliffe (1967, 1970): (weight in grams X 10 )/'(length X breath ih cm). . HARTOLDMON0029727 86 likewise, weighted means of residues were calculated. In this way, we felt that we were basically looking at the variations relating to thickness as a result of the residues, although our degrees of freedom were considerably reduced. Simple correlations (r) of the means were as follows: DDE = -0.97, P<0.01; apparent PCB * -0.91, PC0.02. Ftests with the regressions gave values of 56.7. for DDE and' 20.3 for apparent PCB, significance levels P<0.005 and P<T0.01, respectively. Multiple-regression analysis (Steel and Torrie, I960: 277-304) with DDE and apparent PCB gave an F-value of 22.9, P<0.025; R^ = 0.94, P<0.05. Simple exponential (curvilinear) functions in both types of regressions failed to improve any of the relationships, although the higher significance levels in the F-tests suggested that the relationships were not perfectly linear in the ranges we tested. Because the significance levels were decreased with apparent PCBs alone and with the two residues considered together, it seems that more variation in eggshell thickness can be accounted for by DDE than apparent PCB, but that both seem potentially important. Although DDE/PCB ratios were significantly different between the Great Lakes and Atlantic Coast Herring Gulls (Table 4), the actual levels were different enough between the colonies so that such ratios shed little light concerning eggshell changes. Vermeer and Reynolds (1970) found a significant negative relationship between DDE and eggshell thickness in Great Blue Herons (Ardea herodias), but not PCB. DDE + TDE alone were related in field data to eggshell thinning in White Pelicans (Pelecanus erythrorhynchos), but both DDE and apparent PCB showed significant negative correlations in Double-crested Cormorants (Phalacrocorax auritus), both species coming from identical breeding grounds (Anderson et al., 1969). Egg residues l4 HARTOLDMON0029728 IRT-' '. 8? were probably a result of dissimilar nonbreeding-area exposures (potentially geographical-area differences as well as food-habits differences), however. It may be that DDE and PCB interact either positively or negatively, when both are present. For two Insect species, Lichtenstein et al. (1969) have predicted biological interactions of PCBs and other synthetic chemicals in nature on the basis of interactions in tests with PCBs, dieldrin, and DDT, About all that we can conclude from our data here and the experimental evidence already cited is that DDE residues are Important in eggshell thinning and that apparent PCBs are suspect, COMPARISONS WITH OTHER HERRING GULLS It seems that Great Lakes Herring Gulls contain the highest chlorinated hydrocarbon residue levels in eggs reported for that ' species. Atlantic Coast samples are certainly lower (Table 3). Ludwig and. Tomoff (1966) reported DDE levels in eggs (ppm wet-weight), from several colonies northeastern on Lake Michigan as averaging from 103 to 138 ppm (n * 32, 4 pools); no estimations were given for PCBs, but ''DDT" levels (assuming the same DDE/PCB ratio as found in the Lake Michigan samples reported here, Table 4) suggest residues of apparent PCB at around 100 "ppm". These convert to around 1,500 ppm on a lipid basis if one assumes a general lipid level of 8% (Table 4, and Vermeer and Reynolds, 1970). Keith's (1966) wet-weight mean of 202 + 34 ppm DDR for 1964 in Green Bay eggs converts to 2,525 ppm lipid basis. Chlorinated hydrocarbons in British Herring Gulls reported by More and Tattoo (1965) ranged from 0,3 to 0.9 ppm, or about 4 to 11 ppm lipid basis, Vermeer and Reynolds (1970) reported that levels of DDR.in Herring Gulls from interior Canada varied ' .' 15 HARTOLDMON0029729 88 from 2,7 (n 1 colony) to 14.6 ppm wet-weight (mean of 6 colonies), or roughly 34 to 183 ppm lipid basis. These levels are generally intermediate between Atlantic and Great Lakes birds, , ABILITY OP HERRING GULLS TO "TOLERATE" RESIDUES Why do Lake Michigan Herring Gulls, despite their high egg residues of chemical pollutants, show levels of eggshell thinning generally less than those of many other species which have similar or lower residues in their eggs (see Risebrough et al,, 1970; Keith et al., 1970; Fyfe et al,, 1969; Enderson and Berger, 1970, Berger et al,, 1970; and Vermeer and Reynolds, 1970)? Partial explanation when data are comparable at one end of the curve which compares chlorinated hydrocarbons and thickness may come from the nonlinearity of effect of DDE on eggshell thickness toward the higher levels, as proposed by Risebrough et al, (1970), Peakall (1970b), and as suggested by Vermeer and Reynolds (1970). Otherwise, poor general comparisons between species may suggest interspecific differences in susceptibility. The results of D, K. Wetherbee (mentioned in Drury, 1963) suggest that Herring Gulls have an exceptional ability to detoxify many chemosterilants, If dietary intake of chlorinated hydrocarbons is more closely related to eggshell effects than to body or egg-burden at time of laying as suggested by Bitman (1970) in short-term tests comparing o.'"DDT and ,*-DDT, it is conceivable that physiological peculiarities in the annual cycles of birds might result in nonproportional (but diet-related) deposition of high or low residues in the eggs in relation to eggshell thinning. The population effects of egg breakage in wild populations may also possibly and secondarily Involve compensatory adjustments in the survival rates of both juveniles and adults; and they almost surely will ' 16 HARTOLDMON0029730 89 be found to vary in an interspecific fashion. The critical level of thinning seems logically associated with the behavioral traits of nest and incubation behavior characteristic of various groups of birds. Normal thickness-volume ratios for eggshells which are characteristic of taxonomic groups of birds may be important in interspecific differences of susceptibility to eggshell thinning. For example,- this ratio in Great Lakes Herring Gulls (Table 1) was 225.9; that for grassland Prairie Falcons (Falco mexicanus)(D. W. Anderson and J. J. Hickey, unpublished) was 119.9. These ratios' seem relatively constant between various groups of birds and may be related both to egg placement and nesting substrate. Murphy (1936: 1068, citing Bennett, 1920) comments on the general fragility of the eggshells of Dominican Gulls (see Figure 3 for its relationship to other species of gulls we' studied) and its ability to hatch young, even from cracked and dented eggshells, J. A, Keith in 1964-65 and D, W. Anderson in subsequent years have observed that Lake Michigan Herring Gulls seem to have a remarkable ability for at least some embryonic development in their eggs, despite flaking away on largeportions of many of the shells (Figure 5), Many eggs are also found with longitudinal cracks, their fresh contents drained away onto the ground. Such eggshells often appear perfectly intact until handled (Figure 5h). Eggshells in domestic poultry are known to be weakest immediately after laying and when wetted (Tyler and Geake, 1964), and it may be that such eggshells are broken at this critical stage. Many various "mitfro-circumstances" undoubtedly characterize variations in eggshell breakage on the Green Bay gulleries, normal breakage probably being compounded by the presence of environmental pollutants, . . 17 HARTOLDMON0029731 90 Figure 5. Examples of Lake Michigan Herring Gull nests. A. Well "built nest with 2 young; May 1970* B. Well built nest with 3 apparently intact eggs; however, all three were cracked and partly drained from underneath; May 1970. C. Poorly built, late-season nest with an . unincubated egg, broken as female flushed from nest; May 1970. D. Flaked egg with partly developed, dead embryo within; nearly half the %' eggshell had been chipped away; May 1969. . S'. . HARTOLDMON0029732 HARTOLDMONOQ29733 ACKNOWLEDGMENTS 92 This study was supported under various contracts with the Bureau of Sport Fisheries and Wildlife, Fish and Wildlife Service, U.S. Department of the Interior, Patuxent Wildlife Research Center, Laurel, Maryland. We are especially grateful to the many museum curators and private egg collectors, too numerous to mention here by name, for complete cooperation and helpful suggestions regarding egg collections in general, II. C. Wilson kept us informed on his banding operations and relayed his general impressions on the status of Green Bay gulleries over the years. The persons mentioned in the section entitled "data sources" were kind enough to obtain some current specimens for us. WARF chemists, especially D. L, Hughes, were extremely helpful and prompt. Finally, we wish to sincerely thank J. T. Emlen, Jr., J. P, Hallman, and J. R. Jehl, Jr,, for critical advice. Acknowledgment is also due Nita Hewins, who assisted us in the preparation of the manuscript, . SUMMARY Measurements of eggshells of Herring Gulls collected prior to the mid-1940s were found to vary geographically, this variation probably being best described as clinal in nature. These variations were most closely related to indices of body size, as described by other authors. Nearly distinct geographical groupings of eggshell volume and other measurements involved arctic and northern, interior-Atlantic, and Great Lakes populations of Herring Gulls in North America, Eggshell volume was also shown to relate generally to body size in ten other species of larids. Eggshell weight and thickness were generally related to eggshell volume. , . . 18 HARTOLDMON0029734 93 Recent eggshell changes (dated back to the early 1950s in our samples from lakes Huron and Michigan), mainly in eggshell thickness and empty weight, were demonstrated up through 1969 in some populations. The 1967 eggshell thinning was evident in.Great Lakes specimens, but not in many eggs collected on the Atlantic Coast. Levels of chlorinated hydrocarbons were significantly less in the Atlantic Herring Gull eggs. DDE/apparent polychlorinated biphenyl (PCB) ratios varied as well, with apparent PCB Increasing in relation to DDE as one proceeds out from Lake Michigan, Actual levels, however, of both DDE and apparent PCB decreased considerably outside the Great Lakes, The eggshell changes showed a highly significant negative relationship with residues of a metabolite of j>,j>?-DDT, as well as apparent PCBs. The role of PCBs in eggshell thinning and embryonic mortality remains to be determined. The most extreme eggshell thinning recorded in Green Bay Herring Gulls occurred in 1964 and 1965 when obvious reproductive failures occurred, shell thinning in intact eggs approached 20%, and residues of DDE in the eggs were highest (around 2,500 ppm lipid basis). There is limited evidence that Herring Gulls may not be as susceptible to eggshell-thinning as reported for other species. 19 HARTOLDMON0029735 LITERATURE CITED 9k ' AMADQN, D, 1943, Bird weights and egg weights. Auk, 60: 221-234, AMERICAN ORNITHOLOGISTS' UNION. 1957. Check-list of North American birds. Fifth Ed, Baltimore, Amer. Ornithol. Union. ANDERSON, D. W., J. J. HICKEY, R. W. RISEBROUGH, D. F. [=L.] HUGHES, AND R, E, CHRISTENSEN. 1969. Significance of chlorinated hydrocarbon residues to breeding pelicans and cormorants. Canadian Field-Nat,, 83: 91-112. . . ANDERSON, D. W., AND JOSEPH J, HICKEY. 1970. Oological data on egg and breeding characteristics of Brown 'Pelicans, Wilson Bull., 82: 14 28. . ' ANDERSON, D. W., H, G. LUMSDEN, AND J. J. HICKEY. 1970. Geographical 1 variation in the eggshells of Common Loons. Unpublished report submitted to U.S, Fish and Wildl, Serv., Laurel, Maryland, 25 May 1970. 'BENNETT, A, G. 1920, Breves notas sobre las aves antarcticas. Hornero, 2: 255-258. -' BENT, A.C. 1921, Life histories of North American gulls and terns. U.S. Natl. Mus,, Bull. 113. BERGER, D. D., D. W. ANDERSON, AND J. A. KEITH. 1970. Shell thinning in eggs of Ungava peregrines. Canadian Field-Nat., in press. BITMAN, J, 1970. Hormonal and enzymatic activity of DDT. Agr. Sci. Rev., 4: 6-12. BITMAN, J., II, C. CECIL, S, J, HARRIS, AND G. F. FRIES. 1969,. DDT , induces a decrease in eggshell calcium. Nature (London), 224: 44 46. . ' 20 HARTOLDMON0029736 95 COULSON, J. C., G. R, POTTS, AND J. HOROBIN. 1963. Variation in the eggs of the Shag (Phalacrocorax aristotelis). Auk, 86: 232-245, DAHLGREN,.R. B., AND R. L. LINDER. 1970. Eggshell thickness in pheasants given dieldrin. J, Wildl. Mgmt,, 34: 226-228, DEMENT'EV, G. P., AND N. A. GLADKOV, [eds.]. 1951. Birds of.the Soviet Union, vol. 3. Jerusalem, Israel Prog. Sci. Translations, DRURY, W, H., JR. 1963, Results of a study of Herring Gull populations and movements in southeastern New England, Paris, Institut Nat, Recherche Agronomique, Colloque le Probleme des Oiseaux sur les Aerodromes, 25-27 Novembre, 1963: 207-219. .DWIGHT, J. 1925. The gulls (Laridae) of the world; thejLr plumages, moults, variations, relationships and distributions. Bull. Amer. Mus. Nat, Hist., 52: 63-401. ENDERSON, J. H., AND D. D. BERGER. 1970. Pesticides: Eggshell thinning and lowered production of young in Prairie Falcons, BioScience, 20: 355-356. FYFE, R. W., J. CAMPBELL, B. HAYSON, AND K, HODSON. 1969. Regional population declines and organoehlorine insecticides in Canadian Prairie Falcons, Canadian Field-Nat,, 83: 191-200, GROSS, A. 0, 1940. The migration of Kent Island Herring Gulls, Bird Banding, 11: 129-155. GROSS, A. 0. 1951, The Herring Gull-cormorant control project. Proc, 10th Intern. Ornithol, Congr., 532-536. HARRIS, M. P. 1964, Aspects of the breeding biology of the gulls . 21 HARTOLDMON0029737 Larus ardentatus, ]L. fuscus and L. marinas. Ibis, 106: 432-456. HEATH, R. G., J. W. SPAM, AND J. F. KREITZER. 19&9. Marked DDE impairment of Mallard reproduction in controlled studies. Nature (London), 224: 47-48. . HICKEY, J. J., AND D. W. ANDERSON. 1968. 'Chlorinated hydrocarbons and eggshell changes in raptorial and fish-eating birds. Science, 162: 271-273. ` INGOLFSSON, A. 1969. Sexual dimorphism of large gulls (Larus spp.). Auk, 86: 732-737. KADLEC, J. A., AND W. H. DRURY. 1968. Structure of the New England Herring Gull population. Ecology, 49: 644-676, KEITH, J. A. 1966. Reproduction in a population of Herring Gulls (Larus argentatus) contaminated by DDT. J. Appl. Ecol., 3 (suppl.): 57-70. KEITH, J.O., L. A. WOODS, JR.', AND E. G. HUNT. 1970. Reproductive failure in Brown Pelicans on the Pacific Coast. Trans. North Amer. Wildl. Nat. Res. Conf., 35: in press. LACK, D. 1954. The natural regulation of animal numbers. London, Oxford Univ. Press. LACK, D. 1966. Population studies of birds, London, Oxford Univ. Press. LEHNER, P.N., AND A. EGBERT. 1969. . Dieldrin and eggshell thickness in ducks. Nature (London), 224: 1218-1219. LICHTENSTEIN, E. P., K. R. SCHULZ, T. W. FUHREMANN, AND T. T. LIANG. 1969. Biological interaction between plasticizers and . insecticides. J. Econ. Entomol., 62: 761-765. '' ' '. , . 2002 . i ; I j I < If' I>' HARTOLDMON0029738 . 97 LUDWIG, J, P, 1963. Return of Heirring Gulls to natal colony,- Bird Banding, 34: 68-72. . LUDWIG, J. P, 1966, Herring and Ring-billed Gull populations of the Great Lakes 1960-65. Great Lakes Res. Div. Publ,, 15: 80-89, LUDWIG, J, P,, AND C. S, T0M0FF, 1966. Reproductive success and insecticide residues in Lake Michigan Herring Gulls, Jack->Pine Warbler, 44: 77-84. MACPHERSON, A. II. 1961. Observation on Canadian arctic Larus gulls, and on the taxonomy of L, thayeri Brooks. Arctic Inst, North America, Tech. Paper 7, MAYR, E. 1965. Animal species and evolution. Cambridge, Mass., Harvard Univ. Press. MOORE, N, W., AND J. O'G. TATTON. 1965, Organochlorine Insecticide residues in the eggs of seabirds. Nature (London), 207: 42-43. MOUHTFORD, M. D. 1970. Test of change in egg-shell index. J. Appl. Ecol,., 7: 113-115. , MOYNTHAN, M. 1959. A revision of the Family Laridae (Aves). Amer, Mus, Novitates, no. 1928. ' MURPHY, R. C. 1936, Oceanic birds of South America, vol, 2. New York, Amer. Mus, Nat, Hist. OLSSON, V. 1958. Dispersal, migration, longevity and death causes of Strix aluco, Buteo buteo, Ardea cinerea and Larus argentatus. Acta Vertebratica, 1: 91-189, . PALMER, R, S. 1962. Handbook of North American Birds, vol. 1, New Haven, Conn., Yale Univ, Press. PALUDAN, K 1951, Contributions to the breeding biology of Larus argentatus and Larus fuscus. Vidensk. Medd. Dansk naturh, Foren, 114: ,1-128. ' PEAKALL, D. B, 1970a. p,p'-DDT; Effect on calcium metabolism and 23 HARTOLDMON0029739 98 concentration of estradiol in the blood. Science, 168: 592-59**. PEAKALL, D. B. 1970b. Pesticides and the reproduction of birds. Sci. Amer., 222: 72-78. PORTER, R. D., AND S. N, WIEMEYER. 1969. Dieldrin and DDT: Effects on Sparrow Hawk eggshells and reproduction. Science, 165:-199-200. RAND, A. L. 19**2. Larus kumlieni and its allies. Canadian Field-Nat., 56: 123-126. RATCLIFFE, D. A. 1967. Decrease in eggshell weight in certain birds of prey. Nature (London), 215: 208-210. RATCLIFFE, D. A. 1970. Changes attributable to pesticides in egg breakage frequency and eggshell thickness in some British birds. J. Appl. Ecol., 7: 67-113. . REYNOLDS, L. M. 1970. Pesticide residue analysis in ihe presence of polychlorobiphenyls (PCB's). Residue Reviews, 3*4: in press. RISERROUGH, R. ., P. REICHE, H. S. OLCOTT. 1969. .. Current progress in determination of the polychlorinated biphenyls. Bull. Environ. Contamination Toxicol., **: 192-201. RISEBROUGH, R. ., J. D. DAVIS, AND D. W. ANDERSON. 1970. Effects of various chlorinated hydrocarbons. In: The biological impact of pesticides in the environment. Oregon State Univ. Symposium, 1969, in press. SALOM0NS0N, F. 1950. The birds of Greenland, part I. Copenhagen, Ejnar Munksgaard. . SMITH, N. G. 1966. Evolution of some arctic gulls (Larus): An experimental study of isolating mechanisms. Ornithol. Monogr., **: 1-99. . . . ': ' ' 24 ' ;, . ' : . ;; j' l i HARTOLDMON0029740 . 99 SMITH, W. J. 1959. Movements of Michigan Herring Gulls. Bird-Banding, 30: 69-104. . . SNYDER, L, L, 1957. Arctic birds of Canada. Toronto, Univ. Toronto Press, . STEEL, R. G, D., AND J. II. TORRIE. 1960. Principles and procedures of statistics: With special reference to the biological sciences. Hew York, McGraw-Hill. TYLER, C,, AND F. H, GEAKE. 1964. The effect of water on egg shell strength including a study of the translucent areas of the shell. British Poultry Sci., 5: 277-284. U.S. FOOD AND DRUG ADMINISTRATION. 1965-68. Pesticide analytical tnanuel, vol. 1. Food and Drug Admin. Publ., July 1968: n.p. VAURIE, C. 1965. The birds of the palearctic fauna, non-Passeriformes, London, H, F, and G. Witherby. VERMEER, K. 1969. Egg measurements of California and Ring-billed Gull eggs at Miquelon Lake, Alberta, in 1965, Wilson Bull., 81:' 102 103. VERMEER, K., AND L. M. REYNOLDS. 1970, Organochlorine residues in aquatic birds in the Canadian Prairie Provinces, Canadian FieldNat,, 84: in press. HARTOLDMON0029741 100 Eggshell Changes in Certain North American Birds DANIEL . ANDERSON AND'JOSEPH J, HICKEY# Department of Wildlife Ecology, University of Wisconsin, Madison, Wisconsin, 53706, U.S.A, #Our research was supported by the U,S. Department of the Interior, Bureau of Sport Fisheries and Wildlife, Patuxent Wildlife Research Center, Laurel, Maryland. D. A. Ratcliffe aided us in the initial planning along with L. R. Wolfe. Lucille F. Stickel and E. H. Dustman advised throughout the study, and J. H. Torrie aided us statistically. At least 8? private egg collectors and museum people extended us many courtesies. Ed N. Harrison and W. C. Hanna were especially helpful. J. A. Keith, J.O. Keith, A, Sprunt IV, J C. Seidensticker IV, T. J. Cade, H, G. Lumsden and Sergej Postupalsky aided us on recent eggshell data. . ~ ""r'r' ' '' r ' ''''' ' ' j:: j- INTRODUCTION In this paper we will attempt to add to the current knowledge of chemical pollution by testing the hypothesis first described by Ratcliff (1567, 1970) that eggshell changes discovered in nine species of British birds have also occurred in North America. We plan to do this by (l) reviewing studies already published or in press and (2) by reporting our own study of the eggshells of 25 species in virtually all the major museum and private egg collections in Canada and the United States. The 25 species selected by us and our collegues are a biased sample. Some were selected because of their known or suspected high residue levels of chlorinated hydrocarbons, others because of their known or suspected : ? ! : ' ; m i l ls i n f i l M i sm HARTOLDMON0029742 101 population decreases and still others because their populations were known or suspected to be stationary or possibly increasing. It became necessary to work with large-sized species because of the fragility of small eggshells, and we in no way wish to present data here to be considered as a representative cross-section of North American birdlife. METHODS AND MATERIALS Our own studies have involved 35,017 eggshells. We propose here to summarize our data on 3,005 of these--collected after 1956--and to compare them to 20,655 collected prior to 1957* Our laboratory methods have generally followed those of Ratcliffe (1967) except that we also used a micrometer to measure eggshell thickness (which included the firmly attached membranes) in a fraction of the eggs which had holes as large or larger than 2 mm (Anderson & Hickey, 1970). We compared Ratcliffe's (1967) thickness-index measurements to actual, thickness in nine species representing ten geographical regions (Table l). These tests were run on 100 eggshells from each group and were randomly selected from a larger pool of data. Correlation coefficients were all highly significant (P-40.001), and therefore we consider thickness index as a reliable and H: II independent estimate of thickness. Ratcliffe defines thickness index as weight of shell in mg divided by the product of length and breadth in mm. As expected from a continent-wide study, many of the species we examined showed clinal variation in their eggshell measurements (see Anderson & Hickey, 1970; Anderson et al., 1970a, 1970b). We have grouped state and provincial samples together only when they failed to have statistically significant differences at the 95% level. We did, however, initially divide Texas, California, British Columbia, Quebec, Northwest Territories and Alaska into smaller geographical units before carrying 2 . HARTOLDMON0029743 102 out such tests. In addition, specific collection sites were sometimes used to divide geographical units further on an ecological basis. It is. possible that the groupings by region shown in our tables should be further broken down when larger samples become available for statistical analysis. It is also probable that certain geographical groupings might be further combined. Potential biases exist in egg collections, as Storer (1930) has pointed out. We regard improperly cleaned eggshells as unlikely or easily detected (see Anderson & Hickey, 1970) in the well-curated collections we examined] their occurrence, if undetected, would of course affect any decreases in shell weight or thickness that we encountered. Major eggshell changes leading to' an early or almost immediate breakage by the have parent birds appear to/at least occasionally taken place in at least the last decade. Risebrough et al, (1970), Keith et al. (1970) and Jehl . (1969) report eggshells of Brown Pelican (Pelecanus occidentails) so thin in 1969 that many were broken nearly immediately after laying, so that pesticide residue analyses had to be done on lipids extracted from crushed reminants of some of these fresh eggs. Such eggs would tend to be destroyed before the arrival of egg collectors] their absence in. collections would potentially bias reported decreases as too'low. This bias is possibly present in our data through 1966] it is present to a lesser extent after about 1966 when eggs were collected more often by ecologists than oologists. Since raptorial birds tend to eat their own broken eggs (reviewed, by Ratcliffe, 1970), the bias still tends to be a real one in such species. RESULTS AND DISCUSSION ' Pre"19ll6 Eggshell Changes . A critically important hypothesis at the start centers on the ' HARTOLDMON0029744 103 have in some way been occurring before 191*6 or 19lj.7 on a large scale. That this has not taken place in Eurasian Sparrow Hawks (Acclpiter nisus) and Peregrine Falcons (Falco peregrinus) in Great Britain for the period 1900 to 19U5 was shown by Ratcliffe (1967, 1970) and. for Peregrine Falcons in southern California from 1890 to 1915 by Hickey & Anderson (1968) (see Figure 1), Pre-19l7 changes were also not evident in Florida Bald Eagles (Haliaeetus leucocephalus), where a 19l7 change is suggested from eggshell-weight data (Figure 2), We further tested this hypothesis by examining the shell-thickness indices on a decade by decade basis for 11 species from II4 geographical areas (Table 2). In order to make all observations in these tests independent of one another, we randomly selected single eggs from all clutches. Where recent data were used, these represented both clutches collected by egg collectors and single eggs taken by us or our cooperators in a random fashion from individual nests. We used Ratcliffe's thickness' index in this test because of its suitability as a measure of thickness (Table l). These decade compar isons gave only 1 significant, pre-19U0 difference in'91 'Seeade-compar- isons (Table 2)j Golden Eagle (Aquila chrysaetos) eggs in 1890-99 (n 32) were riot significantly different from post-1950 eggs of this species. Due to the uraritytt of pre-1939 eggshell changes (the one we report here might even theoretically be due to a sampling error), we believe that our total, pre~19h7 samples--all decades combined--are our best representatives for descriptive purposes as well as for comparisons within a given geographical unit. Thus, we feel justified in the com-' parisons presented below. - Recent Eggshell Changes . Where we made 166 po5t-19^6 eggshell comparisons (Tables 3 to 7, further scientific names will be given there), wa found statistically .h HARTOLDMON0029745 significant decreases in 103 (62$), nonsignificant changes or no changes in 6l (37$) and significant increases in 2 (1$). ' Fish-eating birds.--We found variable, but generally what we considered significant changes in most of the fish-eating species of nonraptorial birds we examined (Table 3)* White Pelicans and Common Loons seem to show only more recent changes (1960s) when compared to Brown Pelicans (early 1950s) and Double-crested Cormorants (late 19b0s and early 1950s in most cases). Geographical as well as temporal variability associated with post~19b7 eggshells is evident from the data on Great Blue Herons and possibly Black-crowned Night Herons. Brown Pelicans, Double-crested Cormorants and Black-crowned Night Herons seem to have suffered the greatest effects of the eggshell thinning syndrome. The northern population segment of the Brown Pelican along our West Coast is now believed to be uon the brink of disaster11 (AOU, 1969). The Gulf Coast population was also considered as lt:endangeredtl by the AOU (1968). Risebrough et al.' (1970) and Keith et al. (1970) have demonstrated | that high levels of p,p'-DDE are present in California Brown Pelicans and : that extreme eggshell thinning (-3b to -53$) is associated with such : residues. PCBs (polychlorinated biphenyls) are also present. Eggshell '' thinning in eastern colonies was not as extensive in 1969 (-7.5$ in j .I ' Florida and -16.9$ in South Carolina) (3lus, 1970). White Pelicans from .' interior North America showed DDT-related eggshell thinning of a seemingly minor nature (4w5$) (Anderson et al., 1969), and this species does not f presently appear to be seriously declining in recent decades except for possibly slow, long-term declines associated with habitat encroachment j and disturbances (Lie's & Behle, 1966). The high level of recent thinning . -' . in British Columbia White Pelicans (Table 3) and tbq presence of highly *\ - significant decreases in this species (Table 2), suggest potentially . ' f j t f. HARTOLDMON0029746 105 serious problems with this species in some areas, and we, conclude that the White Pelican population deserves a close future watch. Double-crested Cormorants from interior Wisconsin showed eggshell thinning on the order of ~2Q% and the highest egg levels of DDE compared to other colonies of this species from interior North America in 1965 (Anderson et al., 1969)3 this population has decreased considerably in recent years (Anderson & Hamerstrom, 1967). Eggshell changes in this species were observed on Lake Michigan as early as 1955 (Table 3). Cormorants are now nearly gone from Lake Superior (Ontario and upper Michigan) (S. Postupalsky, personal communication), where substantial eggshell thinning was observed as early as 1959 (Table 3). These birds likely winter on the Gulf Coast (Lewis, 1929: 20-22) where Brown Pelicans once bred in large numbers. The substantial decreases in eggshell measurements of Black-crowned Night Herons in New Jersey in 1952 (Table 3) have been accompanied by reports of a 90$-decline in nearby eastern Long Island, New fork, as of 1965 (Peterson, 1969). . ' Accipiters and Buteos.--Accipitrine hawks tend to have had eggshell -' changes about twice as great as the buteos we examined (Table-1|), un- weighted means of the two groups being about 11 and J4. percent, respectively. These phenomena tend to parallel the reported population declines of Cooper's Hawks and Sharp-shinned Hawks in eastern North America (Spofford, 1969), the better numbers of Cooper's Hawks in the West (Peterson, 1969) and the generally stationary numbers of buteos at least in the East (Spofford, 1969)--to which the Red-shouldered Hawk was a puzzling exception. The changes for Goshawks in California (-12$), Red-tailed Hawks in Montana (-15$) and Red-shouldered Hawks in south , Texas (-15 to -18$) come as some surprise, and their population significance is unknown at this time. The general conclusions of .. 6 :j j y ,? .j J ! ,, | i : '* l! ;' '. j Ij J - .j Ij |l I1 i v' ' I' HARTOLDMON0029747 . 106 Peterson (1969) that bird-eaters are more affected than mammal- or . reptile-eaters are sustained by these data, although the eggshell change in Red-shoulder Hawks remains to be explained, . Eagles, Osprey and Marsh Hawk.--Golden Eagles (primarily mammal- eaters), in contrast to Bald Eagles (fish-eaters), showed few changes exceeding 10$ throughout the spectrum of our data, excepting a small sample from Alaska in the early 1960s (Table 5)* As already noted (Table 2), one pre-1939 decade sample of this species showed a thickness index similiar to and not significantly different from our post-1939 samples. The recent Alaskan sample represented eggshells from two females, and we have reservations on that eggshell change. Our pre-19l47 sample from Alaska was small and not significantly different from its temporal equivalent. Golden Eagles from further to the southj nonethe- ! less, we kept it separate on the. basis of geographical separation and feel that no ecological conclusions should be drawn until more data ; are available from Alaska. Small samples of Marsh Hawk eggshells showed changes in most- areas where samples were available, and Ospreys from eastern U.S. (decline summarized by Peterson, 1969) showed the same phenomenon (Table ). ` Small samples of Florida Osprey from 19U9 and i960 sboweci;.gather substantial eggshell changes., as well. This is in contrast, perhaps, from a population viewpoint, to the Osprey situation in; Florida Bay, -` Floridaj where the Osprey was reported as stationary in 1966-69 ' f ! | (Henry & Ogden, 1970). .! Falcons.--The eggshell changes we observed for Gyrfalcons were .f considered statistically significant but were below 10$. We do not ; know that 10$ is a critical level,and it was arbitrarily chosen by us i here for discussion purposes. Our Prairie Falcon data exhibited f HARTOLDMON0029748 10? variable eggshell changes, many of which seem to be of a critical magnitude (Table 6). Changes occurred in southern California where population declines are known (Glading in Hickey, 1969), although we could not statistically demonstrate that these changes were real (Table 2). Changes were evident in New Mexico Prairie Falcons (Table 6), adjacent to Utah where that species has also declined due to various reasons, including some disturbances by falconers (white, 1969). Peregrines showed nearly universal declines in eggshell weight and thickness. The exceptions were one early clutch (1948) from Baja and 1947-53 specimens from British Columbia (Table 6). One normal clutch was also observed from southern California as late as 1952 (Figure 1). The two eggshells taken in 1966 in British Columbia were not tested, but all measurements were below the pre-194? lower 9$% confidence limits. The statistical significance of the eggshell change in Peregrines.(Table 2) leaves little doubt as to its recent occurrence. American Sparrow Hawks showed small eggshell changes in our data, the exceptions being two clutches from 1952-63 in eastern 6). This species possibly shows some `^recovery11 when very recent specimens are compared to those from the late 1940s and early 1950s (Table 6, data from California and British Columbia)*- , i,m#., Areas of low productivity for Prairie Falcons were found in central Canada by Fyfe et al. (1969) to harbor falcons with egg residues of DDE higher than in falcon eggs from areas of more satisfactory productivity. Also associated with the high DDE residues and low productivity were the thinnest eggshells in 1966-68 and a 3k% reduction in occupied territories over the ten years prior to their study. The overall mean level of . eggshell thinning reported by Fyfe et al. (1969) was ~ll. The various local breeding populations they studied tended to have similiar, local 8 HARTOLDMON0029749 108 DDE levels but dissirailiar interregional levels, resulting in a generally consistent'intraregional pattern of contamination but an interregionally patchy situation. Enderson & Berger (1970) have also established a relationship between chlorinated hydrocarbons, thin eggshells and lowered hatching success in Prairie Falcons from Colorado. The Peregrine story has already been summarized (Hickey, 1969), and the general syndrome is similiar and better documented with respect to population decreases. Arctic Peregrines now show a progressive eggshell thinning (Table 6) along with high levels of chlorinated hydrocarbons in northern Alaska (Cade et al., 1968), northwest Canada (Enderson & Berger, 1968) and in eastern Arctic Canada (Berger et al., 19?0). Apparently, the Peregrine situation is now characterized by eggshell thinning on a continental scale, : Whooping Crane, Herring Gulls, Great Horned Owls and Crows.-- Whooping Crane eggshells did not show significant changes in 1967-69 samples from hatched eggs (Anderson & Kreitzer, 1970j Table.7). Herring Gulls on lake Michigan experienced local,heavy eggshell breakage and reproductive failures in 1964 (Keith, 1966), and subsequent data from that area substantiated that 1964 was a period of extremely high levels of DDE and a period of maximum eggshell thinning, approaching -31$ in broken eggs (Anderson et al., 1970bj Herring Gull data.summarized in . Table 7). Changes were not detected until the early 1950s. Herring M Gulls from other areas exhibited lesser magnitudes of eggshell change (Table 7) and lower residues (Hickey & Anderson, 1968). Two species believed to be currently stationary, the Great Horned Owl and Common Crow, showed generally small changes or no changes at all (Table 7). One local change, in Great Horned Owls from Florida 9 HARTOLDMON0029750 10? as early as the late 19lt0s and eariy 1950s, exceeded 10$, however. This change is indeed worthy of follow-up research. Some General Characteristics of Recent Eggshell Changes The limited post-19l*7 data in Tables 3 to 7 seem to provide enough material for several generalizations regarding the eggshell-thinning syndrome. First, no increases in eggshell thickness or thickness index that we regarded as significant occurred in any of the eggshells we examined except for Common Crows. We are tempted to cite the work of Jefferies (1969) concerning another passerine, Lonchura striata, where eggshells became thicker-shelled (but smaller) upon DDT treatment. There were no volume differences observed between these two sets of crow data and their pre-19if7 measurements, however, and the occurrence 1 of slightly thin eggshells in some specimens of Common Crow (Table 7) suggests to us a sampling error instead. The changes we encountered seemed to follow four general categories: i! (l) Mups and downs'1 within a given geographical area, implying the potential for "recovery11-, sueh as that observed with California Sparrow Hawks (Table 6)5 such occurrences might be similiar to that reported by Lockie et al. (1969) for Scottish Golden Eagles, (2) steady declines of eggshell weights and thicknesses such as those observed with Ontario cormorants (Table 3), (0) early eggshell changes such as those observed with Peregrines in eastern U.S. (Table 6)3 these were followed by complete extirpation (Berger et al., 3.969) so that recent data are not available, and (1|) only very recent eggshell changes such as those observed with White n Pelicans (Table 3)* The eggshell thinning syndrome seems to be a general phenomenon among many of the species we studied; however, temporal variation ("spottiness11) in eggshell thinning is evident after 191:7. Some species 1.0 m HARTOLDMON0029751 r 110 did not experience decreases until the 1960sj others began as early as 19U? This seems to be an important characteristic of the phenomenon. In species where eggshell changes have occurred, not all changes correspond to the general environmental introduction of DDT in the mid~19h0s, but on the other hand, there is no evidence that DDT was a global pollutant as early as then. In all the species we examined, general eggshell changes were not. observed until the post-19h0 decades. Eggshell thinning on a widespread basis is a recent phenomenon. Another seemingly important generalization from the data in Tables 3 to 7 might be the apparent.geographical "spottiness" observed. Certain areas and populations show eggshell changes, others do not. Where data were available, a general but loose relationship seems to exist between population status and degree of eggshell change in many cases.- Such data are tempting indeed,, but we must stress that these associations are purely circumstantial in most cases. Changes in eggshells at least , seem generally and partially related to changes in population status-- not that all population changes are associated with thin eggshells--but when excessive eggshell changes are observed, say above 15$ to 20$ for a period of years, the population in question seems to be generally in 1 trouble. In species with significant population reserves of nonbreeding subadults, a sustained reproductive failure will not affect the known nesting population for several years. This occurred in the Peregrine ; Falcon' in the late 19h0s and may now be taking place in the California Brown Pelicans. '- Ratcliffe (1969) has argued that increased adult mortality--as well \ as reproductive failure--played an important role in the post~199j? . I population crash of the Peregrine Falcon in Britain. The excellent census ' . - . ' n - . I ih I* ii HARTOLDMON0029752 Ill data on this species provided by Rice (1969) and Herbert & Herbert (1969) for eastern Pennsylvania, New Jersey and southern New York clearly show a population decline far in excess of what one would 'estimate to be the normal adult mortality rate. The population decline of Ospreys in Connecticut (Ames & Mersereau, 1961+j Peterson, 1969) . displays the same phenomenon. In other parts of North America, however, known-rates of Osprey population decline agree rather well with cal culated decreases in net productivity (Henny & Ogden, 1970). It seems highly likely that North American birds are experiencing a spectrum of population-effects due to varying degrees of exposure to chemical pollutants. Adult mortality will always be difficult to trace on a population basis. Its occurrence in individual Herring Gulls has been traced to DDT-type compounds by Hickey et al. (1966) and to dieldrin in Bald Eagles by Reichel et al. (1969)* , . Identification of Pollutants , Up to the present time, the effect of mercury on North innerin an bird populations remains unknown, and the occurrence: of. this pollutant -in the North American environment can only be noted here'as generating - a tremendous amount of new and potentially interesting, research. , The discovery of high levels of PCBs in the natural environment (Jensen, 1966j Widmark, 196?j and others) and their frequent association with DDE in. the -lipid reservoir of the avian body (Reynolds, 1970) has opened up the possibility that these chemical pollutants may in some way contribute to the eggshell thinning now being found on two continents. Estimated residues of PCBs were found by Anderson et al. (1969) to be inversely correlated with eggshell thickness in Double-crested Cormorants but not in White Pelicans5 whereas, DDE levels were correlated with shell thinning in both species. 'The PCB levels in the pelicans were, ;. . 12 HARTOLDMON0029753 112 however, relatively low. The field evaluation of PGB effects certainly invites a broader survey. In our own studies of eggshell changes in Herring Gulls, we have concluded that more variation.in eggshell thickness could be accounted for by DDE than apparent PCB, but that both seem potentially important at this time (Anderspn et al,,, 1970b). The list of factors producing eggshell changes in poultry is a long one (Romanoff & Romanoff, 19h9'} Sturkie, 1965j Simkiss, 196?). The eggshell changes reported in wild birds by Ratcliffe (1967, 1970) for Britain and by various North American authors have been followed by controlled experiments testing the new hypothesis that these remarkable physiological changes can be produced by various chlorinated hydrocarbons. Porter & Wiemeyer (1969) demonstrated eggshell thinning in American Sparrow Hawks and the associated symptoms of egg disappear ance, egg'breakage and lowered productivity. These birds were fed a of dieldrin and p,p'-DDT. Heath et al. (1969) demonstrated similiar phenomena with Mallards (Anas platyrhynchos) fed very low levels of p,p'-DDE. Such eggshell thinning was verified by Risebrough et al. (1970) with mallards fed similiar levels of p,p*-DDE. Peakall (1970) induced thin shells in Ringdoves (Streptopelia risoria) injected with p,p'-DDT and p,p'-DDE. Bitman et al. (1969) and Bitman (1970) demon strated that eggshell thinning was possible in Coturnix coturnix fed high levels of p,p'-DDT and o,p'~DDT. Lehner & Egbert (1969) reported .minor eggshell changes in Mallards fed dieldrinj whereas, Dahlgren & Linder (1970) could not produce the same effects in Ring-necked Pheasants (Phaslanus colchicus) fed dieldrin. 'Nor could Peakall (1970)' produce eggshell thinning with dieldrin injections in Ringdoves. 13 HARTOLDMON0029754 9JW- 113 The recent shell changes found in wild birds in North America have been associated in 11 cases with DDE or'DDT-family residues in the eggs. These residues have ranged from to 2525 ppm on a lipid basis (Table 8); residues of 7 and 30 ppm apparently produced no detectable eggshell changes in two species. Some interspecific differences in response to these environmental contaminants appear to be suggested by the data now available (Table 8). Pesticides and other compounds in nature are mixtures, each component of a given residue spectrum may have unique physiological effects, and combinations of pollutants may have additional effects not manifested by each singly. The observation of gross physiological end-results in the field are probably in reality a combination of causes and each ecological situation may be unique. SUMMARY . . Analysis of eggshell thickness-index changes on a decade basis, carried out with 2,088 eggs representing that many individual females of 11 species and li*. geographical areas, disclosed an apparent decrease in Golden Eagles in the 1890s in western North America; this represented 1 significant decade-change (a decrease) out of 91 decades compared. Comparisons in the other 13 analyses disclosed no significant changes. When this comparison was enlarged to include a total of 2,801; eggs through 1969, significant decade differences were apparent in 12 of the li; cases, 9 of these at the 0.0QJ? level of probability. In this arbit rarily selected group of species, we conclude that eggshell changes were rare before 1939 and common sometime thereafter. Expansion of this sample to 20,651; eggshells taken prior to 19W> and to 3*00[; taken since then reveals that 9 out of 25 species have . . sustained shell-thickness and 'shell-weight decreases of 20 or more . '' 31; HARTOLDMON0029755 nil percent, at least for brief periods: the Peregrine B'alcon in at least three regions, the Marsh Hawk and Brown Pelican in two regions and the Prairie Falcon, Cooper's Hawk, Double-crested Cormorant, Black-crowned Night Heron, Bald Eagle and Osprey in at least one region each. In eight of these, regional declines are known; and in some cases, these declines continue. . ... These eggshell changes appear to .be absent in Whooping Cranes, Broad-winged Hawks and Rough-legged Hawks. They have reached 15-19$ in. Ontario Common Loons, lij~l6$ in some - White Pelicans, 1-9% in some Great Blue Herons, 8-12$ in California Goshawks, 9-13$ in some Sharpshinned Hawks and 10$ in Great Lakes Herring Gulls. Changes generally under 10$ (with some local exceptions) were observed in Red-tailed Hawks, Red-shouldered Hawks, Golden Eagles, Gyrfalcons, American Sparrow Hawks, Great Horned Chris and Common Grows. The species we reviewed are not regarded as a representative cross-section of North American birdlife, and the population significances of the data are restricted by small samples and time spans, often representing only a few years C . in a given region. The shell-change data seem to characterize regional differences in chemical fallout, contamination that varies with diet and phylo genetic differences in sensitivity to pollutants. DDE is the pollutant most often associated with these physiological changes and population decreases. The importance of additional chlorinated hydrocarbons, including PCBs, as well as mercury remains to be worked out. The threat to North American species is geographically widespread and not limited to the site of pesticide-application. It probably involves a small . 15 115 fraction of the continent's species and often only some geographic fraction of a species population; but it seems to be mounting^ and its occurrence in the tropical parts of North and South America remains to be worked ou. REFERENCES AMERICAN ORNITHOLOGISTS' UNION (1957) Check-list of North American Birds, Fifth Ed. Amer. Ornithol. Union. Baltimore. AMERICAN ORNITHOLOGISTS' UNION (1968) Report of committee on conser vation, 1968. Auk 85, 669-677* . AMERICAN ORNITHOLOGISTS' UNION. (1969) Report of committee on conser vation. Auk 86, 738-714*. AMES P. L. & MERSEREAU G. S. (1961*) Some factors in the decline of the Osprey in Connecticut. Auk 81, 173-185. ANDERSON D. W. & HAMERSTROM F, (1967) The recent status of .Wisconsin cormorants. Passenger Pigeon 29, 3-15* ANDERSON D. W., HXCKEI J. J., RISEBROIBH R. W., HUGHES D...L. & ' CHRISTENSEN R. E. (1969) Significance of chlorinated hydrocarbon residues to breeding pelicans and cormorants. Canadian' Field- Nat'. 83, 91-112. . ' ANDERSON D. W. & HICKEY J. J. (1970) Oological data on egg and breeding characteristics of Brown Pelicans. Wilson Bull. 82, llj-28. ANDERSON D. W. & KREITZER J. F. (1970) Thickness' of 1967-69 Whooping . Crane eggshells compared to that of pre-1910 specimens. Submitted to Auk. . ANDERSON D. W., LUMSDEN H. G. & HICKEY J. J..(1970a) Geographical variation in the eggshells of Common Loons. Accepted by Canadian ' Field-Nat. ' 16 HARTOLDMON0029757  116 ANDERSON D. W., HICKEY J. J. & KEITH J. A. (1970b) Chlorinated hydrocarbons and eggshell variation in Herring Gulls. MS in preparation. BERGER D. D., SINDELAR C. R, JR. & GAMBLE K. E. (1969) The status of breeding Peregrines.in the Eastern United States. In Hickey (1969)* pp. 165-173. BERGER D.' D., ANDERSON D. W. & KEITH J. A. (1970) Shell thinning in eggs of Ungava Peregrines, Accepted by Canadian Field-Nat. BITMAN J. (1970) Hormonal and enzymatic activity of DDT. Agr. Sci. Rev, it, 6-12. BITMAN J., CECIL H. C., HARRIS S. J. & FRIES G. F. (196?) DDT induces a decrease in eggshell calcium. Nature 22k, kk~k&* BLUS L. J. (1970) Measurements of Brown Pelican eggshells from Florida and South Carolina. Accepted by BioSclence. CADE T. J., WHITE C-. M. & HAUGH J. R. (1968) Peregrines and-/pesticides in Alaska. Condor 70, 170-178.- DAHIBREN R. B. & LINDER R. L. (1970) Eggshell thickness in pheasants given dieldrin, J. Wildl. Mgmt. 3k, 226-228. ENDERSON J. Hr & BERGER D. D. (1968) Chlorinated hydrocarbon residues in Peregrines and their prey species from northern Canada. Condor 70, 11*9-153. ENDERSON J. H. & BERGER D, D. (1970) Pesticides: Eggshell thinning and lowered production of young in Prairie Falcons. BloScience 20, 355-356. FIFE R. W., CAMPBELL J., HAISON B. & HOBSON K. (1969) Regional population declines and organochlorine insecticides in Canadian HART OLDMON0029758 117 HEATH R. G., SPANN J. W. & KREITZER J,. F. (l969) Marked DDE impairment of Mallard reproduction in controlled studies. Nature 22k, I(.7-I|8. HENNY G. J. & OGDEN J. C. (1970) Estimated status o.f Osprey populations in the United States. J. Wild!. Mgmt. 3k, 2llt-217. HERBERT R. A. & HERBERT K. G. S. (.1969) The extirpation of the Hudson River Peregrine Falcon population. In Hickey (1969), pp. 133~15U HICKEY J. J, (Ed.) (1969) Peregrine Falcon Populations; Their Biology an6 Decline. Madison, Wisconsin. HICKEY J. J., KEITH J. A. & COON F. B. (1966) An exploration of pesticides in a Lake Michigan ecosystem, J. Appi. col. 3(Suppl.), litl-lh'iu HICKBI J. J. & ANDERSON D. W.'(1968) Chlorinated hydrocarbons and eggshell changes in raptorial and fish-eating birds. Science 162, 271-273. JEP'FERIES D. J. (1969) Induction of apparent hyperthyroidism in birds fed DDT. Natore 222, 578-579. JEHL J. R. JR. (1969) The Brown Pelican, a vanishing American. Environ. Southwest I4.I8, it. JENSEN S. (1966) Report of a new chemical hazard. New Scientist 32, 612. KEITH J. A. (1966) Reproduction in a population of Herring Gulls (Larus argentatus) contaminated by DDT, J, AppI. Ecol. 3(Suppl.), 57-70. KEITH J. 0., WOODS L. A. JR. & HUNT E. G. (1970) Reproductive failure in Brown Pelicans on the Pacific Coast. Trans. N. Amer. Wild!. & Hsi* Conf. 35, in press. LEHNER P. N. & EGBERT A, (1969) Dieldrin and eggshell thickness in ducks. Nature 22lt, 1218-1219. . HART OLDMON0029759 118 LEWIS H. F. (1929) The natural history of the Double-crested Cormorant (Phal.acrocorax auritus auritus (Lesson))# Eu-Mi-Lou Books, Ottawa. LIES M. G. & BEHLE W. H. (1966) Status of the White Pelican in the United States and Canada through 1961*. Condor 68, 279-292. LOCKIE J. D., RATCLIFFE D. A. & BALHARRY R, (1969) Breeding success and organo-chlorine residues in Golden Eagles in West Scotland. J, APP-T- Scol. 6, 381-389. '. . MARTIN A. C., ZIM H. S. & NELSON A. L. (1931) American Wildlife and Plants * New York, PALMER R. S. (Ed.) (1962) Handbook of North American Birds, vol. I. New Haven, Connecticut. PEAKALL, D. B. (1970) p,p'-DDT: Effect on calcium metabolism and concentration of estradiol in the blood. Science 168, 392-591*. PETERSON R. T, (1969) Population trends of Ospreys in the northeastern United States. In Hickey (1969), pp. 333-337 PORTER R. D. & WIEMEIER S. N. (1969) Dieldrin and DDTs Effects on Sparrow Hawk eggshells and reproduction. Science 1,63, 199-200. RATCLIFFE D. A. (1967) Decrease in eggshell weight in ,certain birds of prey. Nature 2l3, 208-210. RATCLIFFE D, A. (1969) Population trends of the .Peregrine Falcon in Great Britain. In Hickey (1969), pp. 239-273* RATCLIFFE D. A. (1970) Changes attributable to pesticides in egg breakage '' ' frequency and eggshell thickness in some British birds. J. Appl. Ecol. 7, 67-113* REICHEL W. L., LAMONT T. G., CROliARTIE E. & LOCKE L. N. (1969) Residues in two Bald Eagles suspected of pesticide poisoning. Bull. Environ. 1 : j ! | it I HARTOLDMON0029760 119 REYNOLDS L. M. (1970) Pesticide residue analysis in the presence of polyehlorobiphenyls (PGB's). Residue Rev. 3h, in press. RICE J. N. (l969) The decline of the Peregrine population in Pennsyl vania. In Hickey (1969), pp. 195-163. . RISEBROUGH R. W., DAVIS J. D. & ANDERSON D. W. (1970) Effects of various chlorinated hydrocarbons. In The Biological Impact of Pesticides in the Environment. Ed. J. R. Gillett. Oregon State Univ., Gorvalis. - ROMANOFF A. L. & ROMANOFF A. J. (191|9) The Avian Egg. New York. SBIDENSTICKER J, G. IV & REYNOLDS H. V. Ill (1970) MS in preparation. SXMKISS K. (1967) Calcium in Reproductive Physiology. New York. SPOFFORD W. R. (1969) Hawk Mountain counts as population indices In northeastern America. In Hickey (1969), pp. 323-331. STEEL R. G. D. & TOHRIE J. H. (i960) Principles and Procedures of Statistics; With Special Reference to the Biological Sciences. New York. STORER T. I. (1930) A crititique of oological data. Auk it7, 329-33)4. . STURKIE P. D. (1969) Avian Physiology. Ithaca, New York. ' VERMEER K. & REYNOLDS L. M. (1970) Organochlorine residues in aquatic birds in the Canadian Prairie Provinces. Canadian. Field-Nat. 8I4., in press. WHITE C. M. (1968) Diagnosis and relationships of the North American tundra-inhabiting Peregrine Falcons. Auk 82, 179-191. WHITE C. M. (1969) Population trends of Utah raptors. In Hickey (I969)3 PP. 399-363. WIDMARK G. (1967) Possible interference by chlorinated biphenyls. J. Ass. Off. Anal. Chem. $0, IO69. HARTOLDMON0029761 120 Figure 1. Raw eggshell-weight data plotted against time' for California Peregrines. These data suggest an eggshell change in 19k7> 5.0- oo s' <c cc CD z 3.0- 99 9 9 99 9 99 9 99 9 99 99 99 9 99 99 9 9 A9 9 999 9 9 999 9 99 9 9 9 99 9 99 9 9999 9 9 99 99 999 9 9 9 999 9 999 9 9 9 99 999 99 9 9 9 9 99 9 99 99 99999 '9 9 99 $6* 0#69 99 99999 999 99 9 . 9 99 i-- , CD ' IxJ 1.0 40 YEAR 999 9 9 9 9 99 99 99 99 99 99 9 HARTOLDMON0029762 .121 Figure 2. Eggshell weight versus time in Florida Bald Eagles, again suggesting a general change beginning in 19lt7 The'196? data (A. Sprunt IV, personal communication) suggest some "recovery11 or represent a series of females less affected than previously. HARTOLDMON0029763 122 Table 1. Relationship between Ratcliffe's (196?) thickness index and measured eggshell thickness in nine species. Species Regression Correlation coefficient!/ equation Std. err. F~ (!=<.+ bX) of slope value?/ Pelecanus erythrorhynchos 0.842 Phalacrocorax auritus 0.850 Nycticorax nycticorax Aquila chrysaetos 0.855 0.878 Pandion haliaetus 0.918 Faieo mexicanus ' 0.675 Falco spar.verius 0.866 tarns argentatus (Midwest) 0.881 Larus argentatus (Eastern) 0.901* Corvus brachyrhynchos 0.711 0.277 + U.I48IX 0.290 0.261* + 4-342X 0.272 -0.025 + 5.025X 0.308 0.860 +' 3.782X 0.208 -0.018 + 5.096X 0.223 1.51*7 + 1.003X 0.111 0.001* + 5.0351 0.291* 0.052 + U470X 0,21*3 -0.1*31* + 5.85ix 0.280 0.21*8 + 3.1*781 0.31*7 238.7 255.5 266.7 330.4 521.7 81.8 29l*.l 339.1 1*37.3 100,5 1/ These represent 100, randomly selected eggshells from each species* HARTOLDMONOQ29764 T a b le 2 . D ecade .d iffe re n c e s in s h e ll th ic k n e s s in d e x f o r T a b le 2 c o n tin u e d 123 to OOI j S ^5<<< dw $ *>!< * $ $T>!< w o mC2N O (C!O) E-c Cr-Ni fl) a, w H (f!l) afl) P o CD s CO a> $ CO O <3 9 brlD CO o u o CO o (1) 6o-1 OOOn CN r~i XA H CN* CV o co CO CO CO ot r*4 CO CN Os c- H CA D~ -3 ON .* CO (0 fl d m & d$ co * d (04 fl tn CO -d co CSI On CM CO Os XA * Os Cvi * o H H CSi H H co Nx| l 5O3 bbOo jo<0 pa i=> CO *d o(!) I P) CO Os H1 co CO OO OJ r~t Hs^ Vf--l/ SO oo SO co 3 Oo iH H Sp* C"~ r- o H 0-- CO On -fl ' CM co <A oo H *** CO 3 o rS kO--ss C- so O g CN CN On -2) 7? CN CM On CO CM CN e' 0 8* er) w> S o Oo 0) CaO. fl a co o *g So3 A 8 a &cLl! ' fl 1 to o 5** $ 3 -p CO to I o -p o o 0 1I CO 8 8 & fl _(0 I a 8o HPM fl A fl a oo 0 o o a> a -p f! & tzo; <D g $ -p H nmi -P (A O, Is ctf fl fl O' < S p to fl H 1 0) a> * t) 'g d CO o d 0) 52! CO H to d fl dO > H r~! 4 fPli o fl -P CO -P O co g W .5 fl o fflj I HARTOLDMONOQ29765 HARTOLDMON0029766 2 / The t o t a l number o f v a ria te s a v a ila b le th ro u g h 1939 ( in p a re n th e se s, th ro u g h rOd 0 1 *H <FNh T2<wi 51 XOOA ri o W s V P p-<f B %' +> a! 0) *P +> oo 4> rH "d a> >3 H r, t*0 CO ho a) * XA SO- ?o5 O o oH o 0) <D d 3 H H O VV fM fM s> H fil W b ni $ 1oa xi <D o /*s H 54 i1d3 w R Os sO ho a> Os H H CO <A| I | HARTOLDMONOQ29767 TABLE 3 . 'E g g s h e ll d a ta fro m N o rth A m erican museums and p r iv a te egg c o lle c tio n s , show ing 268(102) 15.70+0.21 0.676+0.010 3.26+0.03 126 <wD E o *rt fCet5 O CO 0 0 M 0 0 $ rd as MM J>s rH P 0 <d a w oe 0 U CM 0 M cd 4J <d *d M as 0 *H *-v* U 0 h| o< CO CO 0 0 a *rl o 0 MM 4J 0 *rM 0) d oo o d *r! O CO . * 0> CO to 0 0B l 0 Xo o <44 0 *H d ctf d 0o a) 0 d rH <u 0 to PM t LA A! CO X O CO 0 rl 0 *0 d 0 f-i to CO rH 0 rH 0 U <y .0 to a H 0 W OLAn X H +1! as d rH d rH <u <13 *CCO 4J H 0 to O. I tn *CtMO o CU0M 03 0) X) O .0 <33 > (X rH to O *o > d0 CO M s 0 wO 0 vi *H to U0 0 P4 H P< n0 CO pm 00 Os oO * ?l ?l &>? LA CO r-i H SO 1 * to CM to <T rH rH o o. * ?l Mf O. + VO CM r~4 to in SO LA # oo 1 00 Mt* o. +1 Mf H SO rH r~l LA ?l to r-4 CO r-{ Cs r-4 i rH o i--I O. +i H o o rH CO CO o mT rH O oSO ?l o, -H cs LA o rH vD SO oO CO CM ?! CS CM * iH CO LA 1' o <r LA rH /'""N vvoOy C~Mv CO OSOs CCMO /C--M <Vsr^ /\CvsM/ oA- CO #AM j-* <up1u< srSOOHoIHOss mT aM1) <CLro-OA|sr * /Nt CM* v vO U0) tP<13 O0' y4) *v oo H 'ri *ri i-< u t=J *3 M.Q M *H >0 40J 0 O i0J oa t> O << O 0 .o 4oJ *ri C sJ3 0M 40J M<u . M 0 rSH> .* ,m < - 00 t*oHH I P0M 0 0 0U pm .CH mj* VD to o SI o . uo M Si & CO 0 wd C0 H 3 u 3 0 0a U 4 P0M P40000S.3J wFH i I ji f .1 HARTOLDMON0029768 127 1 fwt & o vM 0 C mH u ml to * Oi 0 p0 H * w r4 - * o rH rH o rtOo M W <u c V /n v-/ tn H ON 4-1 0w00) rH H0} to 4-J *C "fHt Vto> 5 ft rH rx tfgot str<NM|y CO CM r-M ?l p CO CM CM oCM gM CO +1 ON H NO O mm- o, ON + MJ* * CM* rH in VCM' in CM 12% NS 14% NS NO -3* O rH o rH o, O, CO ?l S'? + CM m1 S3 ON H CO i--i ON 1 4 , ' CO CO CM tn . CM 4 ?l sD to CM- 1 CM ON H O NO c o, o, i4n* Hh | CO* CO *3* NO N0m oo .tn rH o ?i . r- CM NO o S^ 010 to <o o ?i &-S CO ON rH1 m o NO CO ON o, o P4'-1 s+f*1 St** p* NO sr pH rH 1l CmO o CO O. SH rH. 4-1 ON NO rH rH c4o-1 I CD NO rH I si* CO rH rH tn o4 CCON o, CC4OO* i smr CO * CO O rH O. ` CM * CO yoM oa pp--*. o oNCMO4 o, C+O VCOO o o, vop4O- o to S3 . PsUV4ONr1.i O oCO CO 4 o, Vm+O 1 CO 4* pON*4 ' tSo3 VO NO H rH pCOv o +1 NO4 tn H CmM sr CCOO o rH /P~-N VsSpJ*f* /-S rH VrH CO rH /-*N CM SONH . sONr Ntn/ tn . /NPNvy p* C\O> ON SP"f \UfPt* SOITN p rOHn CO vO ON CO VO1 SsOr rOHN r-. . <fr ON 0) Sf ON n rH n CM OinN H <U > rH o> p M 0 ft O X3 ft to o ><3 P3 S3 4J 03 0) r0 4J u o u H 4-J O ft FH td *rH 0 U o UH *H rH ft a o S3 ft 0 f-i o iw 'H H 0 a - o S3 . ft ' 0 rH O a. w r< 4J *rl V< fot ,-1 n a) 5 < 4 o s S3 g (U Vi *0 oO ri 4J Vi g 4H w . . 00 M CO 0 tO o w 0 <3 2d *i it TABLE 3 C o n tin u e d HARTOLDMON0029769 128 CM ON O o o, ,VoC4MO-*. o. CC+MM 1 CM in rrIi Mf- o io M **c0d imn o0 o 11 rCHM rH S'O. 0+0 H ON CO 1 17% o o. ci4nr>-*1 CM rOH O, CCH4MM"| tHoS * C*HM 1O* so o. cd f imn* do o ?! tVCMO voO o, vcs. o o vO ON (V 1 o n* rH o CrHM o. O, . rh+-ni CO4M-* CM CM irOH CrO-Oi ON m O41, CM oo MCM*f \oa I1 o, 4-1 m COON o H CM /*N rH o w V/ CO ON o CM /~s CO <* o V in VO rH rH CO CM N*y. m CO cn in . 1 o 01 n ON fX rH r- rH n 01 ON u rH tx r^. cm 1 vD 01 ON rH P< to CO fi OJ 0 sd *r*i o rH o W cd sd cd *-d o *ri V4 U o- O o rH rH cx W 0 O sd H rf .0 Uo p Ho 4M e t-ow *H rH rH rH cd cd w to o U tS x <u td. K 01 o o o H H *4 00 HARTOLDMON0029770 w Q), u 8* H JP H o H rH o 0 rG CO tOns +1 0CcdO) H0) wk CO Q) 0 -0 0 H w CO Q> tt o. g O *H >4CCJO feO CmN I r^. CS &S vO CS <U r<HX mMH -0 C*<uowHy 129 COO m o ! i. &r>S O Cs I CS CSI o. & nRsJ CS o < a CO <r o rH o. l. o, <4*r I so rHI sf Os CS O <fr in co cs rH <r "t *O<sr <u Os upu rH CM CO oo ?i 0?0i OS I ?l 0 rH 1 in OrOH, BS + vOO vDI o sf rOH o, .4*1 Os r-v I Cs CO nc>s CcOs i' SS a, + BOS IOs CO rH vo uo I in sr /--\ o mH CO rH w o rH r^- rH CO **>. vO nS-' ih co r-. cs m <y os H&< rH Sr^O. stnoI Os rH r o *H *H <U a *H 0*H 4OH ' *H rH 0 o * CO 40J 'S *H. n U P 'oH 0H 404 td wrt cd *4or4J 0 $4 k 0 ' <g * rM CO W0 . 5 < ( |. f ft | TABLE 3 C ontinued So HARTOLDMONOQ29771 -21% 130 to 0 tc d d. q3 a *wr/t hJ # u J1i o H h IT> ON -H leans rH rH 2 CO S h e ll |A to A to O "d d0 M fw/N to <U d U *hH (im il) o o fl S--S r~> tA 1 O CM d 44 d * HO o0 1 l 4J rH JtGo ^*0rHJ. '--w/ O. or-s1 i 7% 14% NS Is- o S'S o, O A1 rH 1 Oh rH o ot cO 4-1 CM co vD * rH d 44 d no o0 * 1 1 \0 rH ?mcoi l sf O H O o. Oh OCO o CoO ?l ^2 O tn CM o* < -28% A- O o4 ) cn Oh *3 CM CM CO rH O & C/5 M<rf o *C<Or o. 4 CM CM0 in hoO ?l r***. CM to to <r rH S' rH CM 1 rH `rH ' to rH O O 41. Mf CO o B-S rH CM O CM O O A1 CO o CO o CM* S3** S' CO CM to* ot > O 1 to GJ ***^ rH & 0 N *H Sow^ Oh O o CM 00 sC.Mx' to KO \0 f'-v A* cC'"nO`s tn cm to to Ofl co *r4 M<U Oh rH tino Oh rH 0M0D tr1> to Oh rH Oh rH rtOH 1 Oh Oh . vO. m Oh Oh rH; rH 19% to H *d 0 -fl U H S to I *x)d O M to *H fuHrt* *rl CO o c3 O d H rH rH M d *H m do o 05 a fl ' Ti 0! o fl ri OM t) m 4dJ ? od o *ri 4Mtt4f oH o4d04 *dOH rH 4M c3| p4 HARTOLDMON0029772 131 1 co ,,<; <tut, u *H s 43 .e H U r--1 fH 0> a 43 COA 2^ in on +1 co d dd 4o2 CO tn{ to A to <y 0 *d ad HM CO to a) dg OE HH vCaO CO <u r0-l 2< C9dO <u N *H CO O *r4 M 0 P4 CM rH O H o. U4Df o, C4A1 CO S3 on o rH CM to in o o* ?l CO on C4M| O' I CO* CO* rH . rH oO 4d4 d mm <dp ?l ndd cCoh o 1i *d0 O ii -da I 1 CO d d d r- On o CO o. t4o* I |l *ST 00 -tf VD CM * o( CO1 SW3 m -d* CCOO * +o,| S-4 rH <H m+ <r P'S Is oo * c> * <tD o CO CM <3* CM OO OCM 0N I C\ Isco co * ao vo o CM i CO O. <4N ON m mn i rI no CrHM o voy wo co i> ON rH o CO V"\ VrwHV v<3-^ c- CHM CM r- *<r "l r- kt CO n o n> ` 0 ON ON ON U rH rH rH P4 f'- CM NO 1 0 t co (U n M M ON tt) P- rH Sd 44 d Q> O no O IOn n* 0) a> CM * rl U o *d <>u vD a tu o to rH O d *W co CO< rH O S H Od P<tDoi td d n3 *a H M OO rH rH fH fH d 'O M O rH P~t d H U o rH Pm 3 0d0 d ^o4 <u *od 4u03 d a u % aM <du X3 ud coo d o WM <5, <\ $ |: i) i; ina TABLE 3 C o n tin u e d HARTOLDMONOQ29773 132' 1 0) H! o er, rl S' J3H <5 in to V) o o *d 0a M (0 w <0 4 rH H d o rCJ o to H ' in H on r\ g Os +1 w r-4 44 % r4W<-3yi '>43t0H35:o to rdH & two ^1<Nu to is in O 0 0, O . r+H +1 CM rH O 4 CM CM CO O O* <d 44 o, <d + d Mf 00 <r 0 vOO in rH $1 ?l BM <Ir VO vCCOMM*/ y*~S SO-' CCcMOo tn co r^. rH vO| 0) co On OU r-i COM O0n ?l COM c?oooi OtN cs rH i00n 00m C?OCOOni gl ON CO O0 CO O o wj. rH O, mhi r*H VOI VO m ro OO oo o. OONN A1 CO On CO o O* ?! 4fidddJ CCOO o0 * o ' , II o rH Hco * , ro-* o, rc+*o1 tPo3 * m in v<VOr/ Mv<*jy--s r0CHO ON tr> eMa1 . OInN vmO . ON rH . CO /-N vn o v/ v/ ro cm m cm rH ' *r<*-0u<1x c<umQo13o\ rH "A fMooo CO *0H to O 0) O CJOX .n C3O r>H O P M- *H N c0d tc > rHed 4M4 * . n 0) u aCD ml g *gri *OH to ' O 44 * u I *d 0 tao ' 4 4c3d 44 > 4o4> I 44 H >-i id rl SQ5 ssid IM 4 iwHJ 43| id P4 <<l rOOH 44 4W3 uodo <u, '*ctrqdo 4-H4 a ocd * * ej 4i . d iwo raHtd C0O 4i i i. 1 TABLE 3 C ontinued h w a rd ! HARTOLDMON0029774 2.62+0.10 0.285+0.020 1.39+0.04 25(4) 133 i CO <y o u 0 <d a J M0 o C/3 in On 41 xn 0 rH 0H <u 0) m| M CO <D <4 t3 00 M CO <0 <u 0 O `ri to to D<sU tH0) A< N *CH/3 XI o H <U A< ml nO O o VPm/ 1/5 <D *H 'O O 0- <U > 0. Cu rH .2 w O 0 43- > h 0 6 44 CO M 0 o0 o 0) O <y *H o*\ ca. "k *H U <D to o p3 *H Cu 9 to Pm Co\| o rH O. o, mo4-1 41 00 0>* 00 525 rH ioon 0 44 ?! frf XI ao o <r 0 o O ON rH O, o. + ! 4* j to rH CO S3 CO rH \D nD rH 00 O s-/ 'w' CO t'** rH rH r- 0 <r NO 0 o1 m u ON p* a o u 0 JXJ 44 41 60 SH3 *0 <U I u V Jt Toi~di ea rOri mo O. c4m-1 o, 41 rH* 6r0Hi0^ rH rH ro oo * o. id 4 ON 00 CM COM* sOO ?! Mf rH Q HrM4Hf m oo 404 ?i trf *d ncCoM- o 0 CO js;. ro o o rH o, 41 ?! CM CM -H CM CO C^l 1 CM CO O rH ?! C, 41 CO CO rH !25 CO r- CM CM ~N On On rH O 'w' m CO 0*1 ro r-s Mf <o rH O 'w' CO CM mH rH ~s CM m o ON u rH 0* r-x SJ" ... "l . ON m 0 ON U rH m 0 rH 0 rH PH rH Crf U m td o o CM U rH O 0 rH o PH CM nO i X erf U CO o o ` *H 05 44 O rH *H erf d rH M >N 44 44 o X O 0 to utrf cd <y o 0u o ol o 44 *H 44 41 0CO ` 0 O r^ m J25 & l erf <U H <4 m cu 4^ 0o ri 4.1 M 0 J o 4J M O oC . 1960-61 O ntario HARTOLDMONOQ29775 13H I CO a! w M 03 u to <u ot o nd a ,G G ' cd H M o *> to to CO rH o> rH 03 G a* *--S g a *g u -fa oo *H t LTi H o> ft to dH cd rH 0) 2 *d Kf) P /-~N 'Eb toO 0) 0) 0rH <f\ ft* cd *rH CO CO ny o H P 03 ,*** rot A o o wPm CO a) *H tf U 0) nd 0 > CJ Cc H d to O d> tf d M o *a o <0 o a *H on w *4 (N I rN H GH CoaHpO>* ` #oC5O <H3 IcHd fov * ?! 6^ 00 CN * i CA * H fo--i o. tN + SJ- rH <r 1 cn * o o CN * ?l CO CO rH to 1 CN O' h V~/* CN H <f sO 1 CO VO 0> rH o *rl u d P od * 03 w d O <0 rH o li Vt in 1 P pO <u o > P M~i x) H & 0) (!) to 6 P u *H` cd P 0) 44 pO O cd ft px 4fc <D to d H p: cd o to #- Sh U it .cl P CN p & 0) d PH Uo 0> to P to 0 4-i U G ^3 O p O 44 to co P O c o 44 (0 cd 0) V d 03 G CH 03 W 03 0) tn (C a o on p > CO o H 03 *p e P p c -d o P cd 6O cd Pto On H 03 vy o op 0> p p* cd o 44 & ,o 1! 03 0) rH d <1) H P 0) P P P p >N 03 G A! U p 0) cd 0) rC > fp * > P o Ja to od 44 a CN *w 0 VO 03 o G d *H CTv X3 P rH O o to is P< W o >N nj - *H HH p to P P d no d a 03 o P 0) to to g rtf to P -P 0 rH H w o0j a) 2 x td O <y no S fH to 03 H H to <U a to d toO *tf too Pd d o ni d ft td up d H d cd G r4 P cd u * nCtOf Hr-t P p03 O rtf H 03 P d lA II *H P rH <u to d ON NT o s0 f> <0 to VH-y* " i `44 to? CO d 03 to d 03- rH <u a a> *H to O o3 *rl 03 HI .P N *U 03 CO *H rH G P. in nd cd 03 P H cd 44 -H 'to 03 U G 0) rrrMH-< K) G cd G O to p 03 *0 G < /N G O *H P td o rH G G d 03 cdi p cd H in a> G) 33 p a to pj 0 > 03 to !i 0 .p 03/ *rl r- 03 PO 03 o 44 03 H <u c p p cd p d cd o H CM rl d toO *ri ' CO to 0) too d cd G o tn rH d O r~l 44 4-t *H nd P G cd u H tw rl d toO `ri CO 0 P P O G* 0) P tPu cd * d P 0 d cd o *H P' tw m H CT\ a po rH N-X *H to P (0 P -H 2H rH 1 if r*4 U ra03. a p d> 0 q 0 <4 P pj p 0 P t<M3 CN d H P P cd >4 8 o u 44 0) Nrt rH cd p 0 G 03 too >\ HPoCO CO ds ' > o to ii 3 <n O P 0 CD > 0) *ri da to M /3 *aH* IU li 4-1 vO d *4 >' C5 H to rH cd 60 BE Cd H 0 4J <u taoo H P rfti cd 03 H * S cd to 03 P cd to 03 to 0. P o r^. On rH v r~^ vP \ H v-^ G 03 03 44 p 44 to *r4 G- rH G* O P -H G <00) Pd. P Hd 03 P >cd n P 00' G O rH rH o 44 0) P 8 o p 44 0) N H P cd g v0uGo cd G O to p 03 wd. a 03 *0. CO (3 G *4 * P rM tf 03 P O .d HARTOLDMON0029776 13$ in! 0 W k t>0 0), V I CD Hi TO T) 0P H P .0 a O C W to <0 H 0 rH TO 0 /~N w TO 0 g P o a0 U CO *H 4HJ 0 in H O (D iH--f +1 O 0 60 w 0 .T0O trMH 0) , .p 000 /~N 00 TOto X* *ri a) CO <u M A Vi p. rHTO TJ P & CTOO CO to ttoo 1 -d o <u p< r~- *4 oo i. o. Cc4nN- O CO I CM CM in CM CM oO S' i\ iS--M{ I r~i o <h O co t-H o, & 5-2 CM <Ts CM m tn r~i 0\ O O o, A1 I' 'e* CO oa H <7* I O coa oo Sisa. 1 CO VO CM C\J CM CO oO 1' O -&S, XAI O|n OA H /N co o> s*-* co CM co r*H a\ rH <t v-> CVOO in VO 1 1 0 0 *4* o\ 04 t~i r*4' * i to ; - n . P. 00 m 1 r^* Q\ rM 0 gj <-s . Hi 0 O o m 5*|I <D *H no /~S u to uTO to > Pi r~i mo 0 CO os > i3n w0 l 0' *P TO w o p *--to *H *H <N I *H 60 {*> O 0> H <0 Cd o *rt CM TO S CO s<-!7 9 Cm , w <1 ci M ' cc*dH H-i Cm H U a < is 0 to o o /^S * vmy to tt H rH 0 r-i H H P H 0 P0) 0 0 5o| U H H 0 M00 tu. 4J o W 0 4-4 *rp<X H rH *H TO u CJ O <3 <4i TO *H 0 U a MH *ri H TO O TO .*0}- 0 *H 0 W I u J3 CO in w to .0 A-l TO U -M to X o i--1 to > * w TO TO '* 0 TO O' 0 TO U v ' ' TO o CO < & u <u p. o o o A. A. <! HARTOLDMON0029777 136 m| .M m k O V) <u 0i-l cl ti ja 52 HH 10 CO * CO rH o rH S3 /*N 6o <u 43 trt fisd V *H @ w sm rd ionv -4 -H (00 H Q> H OJ 43 in U rC "s, es 60 *H Jg CM o oo mro+^.! <?Nl i$n5 vO o o 4J o. rd 54-) 03 co CO o0 o 1 i CO CO o rH o. O, 4* CO v+o1 co 25 rH rH CO CO co Os oo , n- vo rH rH <# 4J 4-J 1 l 11 oo 52 5J CO o ot +1 w <r 23 rH CO m rH O, 4* 1 rH o rH CO 1 C\) o S' VO vO / 4-1 * *XJ i i o fS rH o. S'? 4d VD 1 CO * CM in H o co ?i in rH rH Os o o o, + in m 23 CO co o d 4J CO o a CO rH vO O. o + CO 4r-s. {23 CM o * O CO CO CO <y rH ^r| pu d g cd H to CO *0 o rH M <1) PH CO /*\ CO o CM CO o CM CO co <r KT ~\ 1 a> Mf u Os D< rH o v/ C\ rH O UO Cv <t OS rH Os vo^ O \-r^ V/ fO H . <* CM CM CM CM Os m uo 1* rH VD mm cr .. ON rH' rH V0 m CM m Os rH O CO lO m cr> rH oo pH X> 4J 'w U <u W a ci id 0> 01 *H *H *H t3 cx H CO O> Wo 51 M O CJ ko 54 O s o 36 0 MH 44 44 44 u co . . Cl M oc iH O *H CO rH d *U *H rH d u *rH rH cd O *H H HH cd oj o ^2 rt , aoN O ofcO <u gO <* o o o *3} *rl i-< H COl* P4 .< CO CO CO CO ^3 aw :i I, i I I; ! i HARTOLDMON0029778 TABLE U C ontinued HARTOLDMON0029779 1947-50 15(5) 2.97+0.10 0.324+0.011 1.61+0.07 to Q>, 43 o to *4 a in' cv H to 0 flj 0) o CcO rH H to ,0 to m w% to 0 <y *o c Ma w to o rM0 a 4J 43 oc H VC-O/ 0 pi to rH f* <d to to xs o *rl M <U PH -A T3 O O VPm to 0 H 0 0 O. to 41 P Vi "A. rH 1 P* to 0 ft u 0CM Vi tJ 0 A o 6 M 0 O rl feO Co4 ON O CO rH vO I * H -w 0<3d r i rH H ?| VO <T rH CO I CM rH Mf o O o. ot A1 CO m l. CM rH CM CM Mf* iH O On JlO o, rQH c4o- in l CM o sf Mf Oo in oH * ?i cCnM ?! CCOO in l V0 m CO rH 0 O. 6^ A1 1 CM cd td n3 1 1 o 0 Cl r- o. A1 ON M3 1 in CO f- o o o. o4 A1 vtO r-l , O CM CM 6M ? CO o o 4CJd o, cd 4- x> -so o 02 o i i O CM Al o. NO 1 CO in * mm l CM O a. c4n CM CM <fr o o gCMl < o CO o o 4 CM CM <> /N VO r~\ O 'CO /N VrH-' *CwM* Vo wCoO <*-N vo-/ . CO vrH/ ON CM r^. vO VO rH ON ON rH rH CM CM co OtnN rqj <D r-1 0) rx OinOronHtNn . VVOroHOD1N VO VO t i rH 0n H On 0< H *-MT 1 <3? *U nIo <S? ptf * r"N <Tt * m a 0 <u H cd O too tA *H v> 0 0o) G3 d *rl M 0 rHd <d T") tod tHd 0 O 4H o 0 A rH 0 4J o MH rH td a o to UH *H H td u o to 4J (00) o d too 0 rt K 0 % Mto 0 0 a voi to 0H M ' M Pm . o a d pp PQ| i f. ij fcif TABLE.U C o n tin u e d HARTOLDMON0029780 139 t0fco)O i ort PrdSt ,HC b <mw0 inX0) Ma * w * .j u an* m CM +1 wa to ja) is? HH0) CttoyO ,a u w *H .G H rH U H i*G ai xmt CO *H Q r\ V./ to \rH0 I' rCHN HO co 0ro0 O |N in o, + vO CN <o r<Hy cl. t<6o0 tM oH POHJ VCO' v> cmIo CM NS NS -9% 14% NS 14% vOO o rH O. 4> CM* C$O5 Jl o &C-O1S N . rM +cJd Vfd *tao o 1 <0 o 11 00 in vo S' rH , VO H ro +i VO vO * in to rH o. CC4N\ H VD rOH rc?n tn i CO CO *' o rH rH S' CO i m CON ton I1o4 C4v <T tJ2oJ CN CN CN ` CN CooO o, 4 rCoO CO rOH CN CN O* O CO VoO Ov CN o. 4' tn CM CN * vO vO oW o tn CO cionIn. rH m tn CM cHm H vf CM V - CO m V-/ Nj* M*. CO ` rH CN m co <r vO * 1 vO I CT\ ON 0> <srH U CN P< rH table h Continued O *0 Uo roH <* r^J to A d0 ej o 0 M 0 C5 ksO~Ut to -Q) M to *H rH IaQi Ca>j H r4 O & 4toJ a) nl IH 0 H H cd M 0 O n| JmI HARTOLDMON0029781 4.16+0.11 no data 1.85+0.04 61(0) lUo w<D, & 4Oc3d i jtHeoH o 4tdo3 mOn ft <tao>y rH rH 0 CO 00CC3l ma*tl)) ww 0a a 43 H p /-N to rl 3 0 rH' MfJ s< d 0 id H to CO *0 o H J-t G) cn TfOonS 00U<pH3u). t03na33 nC3l 3 o C1 HH 0COH3l w0Cu ot6ril0. 3prf Vo| O o, CO+Mn 1 rrH-*i SmMI CM CM CM rOH <d S' *Xid5 <COr o0 CM CM rH CM * i m i O int m CM NO NO <o$ oU> I1 CM tH- CM CM COI CHOM O. 4ccJdd .c+o *C4M* O in m H CM i Si 3 o CM \o vO HO HON f*4- I ?00l co Is*. * rH rH* W}25 moo 4ccJdj sO' CO ao o -o4 C*<Mr Jl tH i lOA Cis/3 CM <n VOh1 m On rH CO rH r- n> NO | CO 0 4- Hft ON rH W 0 X 0 H ,13 4H ri o w s X 0 H 43 4J 0 0 to c** rH /"N CO o v/ NO O CO f"- 1<r NO * l ON <u *4" u ON Cu rH 03 3 eJ H M rO id f3 TJ 3 *iH H M i--*il rH0 O . rH PM in (4| rH /"s vv3 >O--' <f CO (0X0) r-i r** On pci o 4 w I <r^Ir 0U0* *4 on rH HroH3 ,3 -0aI3 er>f /<"fNr r-* in 0H cn 8 St HaG)i 0 &04WG(0J> UU<o43 *H ra*<HdI coo cal 1950-55 So. C a lifo rn ia TABLE U C ontinued HARTOLDMONOQ29782 p ra - ' 47 300(38) 4.224-0.05 0.353+0.004 1 .8 4 + 0 .0 2 Ikl T h ic k ness In d e x d a ta -1 1 % ) <tut ato 43 a H H <D u& W in c* +1 to 0 (O a) 4a3 00 nl CO to 0) d /-H l 44 00 to *H f2 m o w I' a vO rH m o o 4- 8to*5 <3- I to H P3 44 td t1i 1 T3 1 o0 d o> in rH H co o, S'? + CO cn H ?l cs CO I \0 I o to <? 0 rH <f| & 04 9N *H V) 00 *\3 O *H V4 04 P-t /*N ""N o vy vOy to i-j to o to Vvd vo 1| vO rH to VO <Js o% rH rH rH s0 Oo <p p4 T 1 H <3* CD- cn 1 ON 00 * rH rH <N H O to oO # o. o. -- [ 4-1 <r to to < cn to o .0 to 1 cm o- H o, O. + to m to to ^r i O rH O + SO>n to 1 to *1 rH 44 ttf O d to to o, + to iCh1O5 to *4* /-N o to to to vy vy . O to rH * O rH rH to to -3* to - 1 * 1 o\ 0 <r u o> *x rH vOy rH i--1 CO vO 1 o vO CTi rH toA (0 *0 o Pom a) to >H 44 to to y O a) o a H M 44 +i d d 03 0) <u sHuoH 4vo-*i d *X> n 44 <0 d o* ' a- *H *H <0 rH #* rHto rtHO 5 'DO o d H 0 1-4 *rf rH oo W Vj G! ttf 4- 44 It too too Ha o d o H M snl G u lf C oast, Piedm ont ' HARTOLDMONOQ29783 T A B L E 1|. C o n t in u e d %i 11*2 1/3 0 tt d 0 u * m m* >4 O m On +1 05 d d Cj 1 *m^Tli 0) X o to <u *H Ci *0 fi d 0 HM * H H G) & to 03 03 <U d ** o *H H /*s s E S--' H 4J H ,G /*N <11 y/Z 00 v0->0 to a) ON wC Xf\ o o O * o * o ?l ?i s-s vO o. 4* 1 w ?i 6-5 -cf o( CO 4-1 rn1 cm CO iz f^. i On CO r** * H rl H H 0 00 0 4-1 4-1 4J 4J M 0 id d (0 cd +a *0 no 1 a I -0 i ! i o oooo d d d a .d in rH O, +| m a o CO CM o, &M H O o' H CO * CO CM CM O. -f | c/d on S3 H Mf o H Q 6-5 X I CO o1 H * <? i*-* rH o, 2 4* 1 525 rH rH . CO O * i wO CO H in o ?! o r-* rH `O CM O <?i 4* CM nco o 0 4J 0 *0 o d CM CM rH HI d i O CN CO NO * <r CO <u -v^ H fX <y gN 0 H to CO /-*N O CTv CN *0 o CO U 0 ON CH H /***% o o, H /^s o rH o *w* <3* m ON ON <3* m CM 1 -. f m Ph o ON <3* m rH ON ON H rH /^v Voi-^ ON . <r ' co \D I."O WO On H in W0 V-/ V CO in r-* rH LO * 1 l On 0 U ON P H CO d H 0 rH nHdo M o H * nHM0o O H PH ' 0} dd 0 wX0) rH w cd <XU H r4dH ^o to |tHX0d <-Ms o to HARTOLDMON0029784 TABLE h C o n tin u e d 6% Ht3 W <ffPcl0)t, I ao ,,C-< o 0) u rP C/5 t/0 ON +1 to ttPoo rtHo tttooo 003) c Ma tCtooO oH H -M /-s to *fwcO/ <y CO o ioH * MCrl\ I of 4 CO <?i r-* O, 4I Kn rHI 4tcnoJd oa CO o &$ o, uo rH v4O- ` I 1O Mf CO mo o ?l CO uo co CMl CM o in CM ot 41 CO rH uo 1 (O 04 CO oo o, +1 or** ?l <3 to 4* o o 4to3 ?l o R) *0 CO CO S o vr> MD o rH ?i O, 41 cn GO CM Mf CO ` CO 00 53 CHO . ?l W in' k rs O S' CO MCOf o. O CO o <? S' ? O CM M3 <fr UO m C5O3 to Ss% 1 r~i <rl p0 gN to *H V3 CO /"*> o CO 0\ 0*Or3l PMt-o4 MUf10r . CCUHMO\ /\ CM V <r . r* MOUUOrH1OO>x o ov0f*<Oof3<yb *-ttottapoarooot 003 r>~i Q MP> *rM I *0tut(tr3oooXl OP cttxooj * HttKtooo rACJ ' 0oP3 ttoo H0 4J tPOo <s> o. > uo CM v^' o \o o --! r*H -tC0HoO CO /-"S CM f" #*-N rM r-* s-x V*' in r^. r--1 CM ? to tOo > r^. a!n1h)- CivUrMH^D1O0.t r^ * t1o tvvM*OD1H3 up. rH uoI. g Mf VCM/ 0tt3oo t1o3 rtHtoo 1 A<yJ t3o p. M>J0 rRPHS to PU A0J 0* 4^-> r*at-of o u0 oH HU o <6 ^3 o S3 MOH <wU . *atro( * & * 4 0 M0 J3 Q05 a. 4N 'uOo t1OPo3 *taHo Mto <i - 5o25 tc H * Oto u *OUrU4 U < * H i3: s; <. >tttooo P p ppp F0Q HARTOLDMON0029785 lUU HARTOLDMON0029786 22(3) 12.00+0.49 0.510+0.124 2.76+0.09 145 CO 0) rH fcb 0 O) 'w' c rl <Si J* 0 O -d .TCO *H +-> * *H p, 0 *H oO H O M S O 0 rH rH to O U H JO to 0 to X 01 0 0O 0O 0 4-4 > H 0 0 iH JO 0 *0 H 0H 00 > 10 0 0 X 4-1 0 *H CO 3 ** CO p 0O 0P U P< 0 UM 0 & 44 <! O *0 Cfi 4J 0 0 *rf aO; 0 0 P< $o 05 Ss, u 4-4 d Hi 00 0P 0 P M xi 0 0 H *0 O 0 o H 0 H H *H X} 00 Awd to (D PM to W) to d ft W0 rG u 1A rH cd w. PJ M X rH 0 CO CQ 0 ^4 <0 & H to TO G) tc ct m js o ftTO* t-4 a m on +i CO 0 0 0 ft 1 U H ,0 H rH rH 0 X to rH rH 01 .0 CO 1A| CO k CO 0 0 *0 d0 H CO (0 0 ag o H S-/ rG 4 4-1 X /-s to to *H 0 V~-v . rH <rf Cu 0 6 Cd d *H to CO nj 0 H M Gi PM ^1 00 X) 0 0 'Nw' ct. *-4 0 0 U-i w 0 H U 0 P< CO /s C3O ft CO 0 *H U 0> P< CO Td 0 r~>i O h-i 0 O *H to a) c4 <N CTN 0 rH o. O' 4*} 4-1 CO CO i-A & 0O ro CO CO CM P4* i1 40 C\ CM 0 0 0 ttJ a, 4J 0 + X) rO OO O lA 0 O rH 40 O ! ?! I CO co 40 0 O CM r-C rH O, rH. 4*1 *f | CM vf CM CM * rO CO H rH to 'SCt rH rH *--1, 4-1 rH CN rH /*s 0 0 CM O V' <r CM CM rH CO 0 rH to CO r^. <U Sf *. r^. 00 1 0 ON *0 u rH rH P. O O * y*N r^. ft in w u> W 0 0 X> 0 0 0 0 0 V H 0 0m a) 1 4J 0) 0 W 0 0 S5 H P0 0d tJ P MO ai 44 0 w H rH TO rH *r4 0 0 0 cs a V` . V<f'/ CM rH CO <fr ON rH 0 *H 0 U O 44 rt rH 0 a CM CO rH rH CO CM CO ro 0 0, 4 to uo 5? 40 40 O 40 VO * O. + 0 CO 0 CM H. to K ON Kj 0 44 0 *0 90 CM vO O. + ON ON CM rH /**S CO O w V-' CM 0 rH rH CM 40 <ONr f O ON 40 rH ON rH Gj cj iH *rf 4*H4 4*H-4 rH rH 1953-55 C a lifo rn ia . HARTOLDMONOQ29787 li*6 0 0, 21 4Ut3d J: 4HoH3 CcJ in<*|u TMC3 <0 0 rH Q> 4 rH C o *e 0 Ato u H to O' 41 -s Ej E sy a c<Dj r-4 r~i CAD< 4J .C fcO *A to w to 0 *> vj- Oo> co HH <K' OI ' ?! lA W5a O' o. 5^ O-H' >43 Ir-i O, to rH 25 ?l nO 2CO5 o. i1 w S3 o |l s-c 1AI oo CM CM . CO CO CO CM <r sO rd 4J cd GJ 4J Cti o RJ * 4.J o. Rj d 4J CO 'd 1 i 4 V5 *d i H Io I o to o 55 o i 0 Id d SO d d o R3 4J cd 8 d SO sO vO r^. H OS CO so * 6-S ON O. f 1 CO CM* ?| ' Ae A. ?l CO I i | CO CM i ro * ?! i to i CM CM CM <3- -cf rM H H H uo r- ?i m CO 25 Os CM r-t CO sO o, rH CO A? AI CM rH TABLE 5 C ontinued HARTOLDMON0029788 1U7 V) 0) a 6C H Pi rfl rt H m. w w <u fl> U c M hj a rc0Co <0iaf0)i o in on f I to Scd r0H5 4J *0to. to CO n0H*J **3* rH o, i1 rH m f^v m o o. *f 1 CO CM vO o 00 * o. -H CO CM m CO rH NS NS St rH o, S00t1 C1O CM ft* P cd 1 O vO o, a* r+V o H r~l 1 CM rH CO rH o'. NtO 1 i On CM H oO CM CM rH o. o. + m CO A1 +1 CO sr i H S3 CO ON CO CO CM* o r> ON CO in CO o o o Cd i o. & 1 9Hi co o. 41 1 O CO CM CO a lO NO m * o Oo co o. +1 \D* N! vO1 CM H o t*** St CO 0o0 rH, 4* I to rH , ~H o. + co CM <t 53 co CO CM CO rH 1 CO - <t rH rH rH- rH rs v-v rs N! si* v/ o w O V-- t*-* w r-* V-'" Ns--t<* r-^ St rH St CO CO rH rH rH ON Mo0 NO | CinO 1 CM VD 1 St St O * 1 inrl ON CO rH nD I CM U sf <U ON ON NO ON On rH - a) NO V4 On rH rH rH P, rH leucocephalus -(6 ,7 ,5 ) Bald, eagle TABLE f? C ontinued > g cd cd ft* i4 4J CO W 0 cd cd O rH rH rH <5 ' <OOdu rH HARTOLDMON0029789 lua w I u cf! cr*d *XHri ww<r<u * *d MP u w t-3 u 6^ Aa) v> tn OV m<u M sHVHi +1 tn <<PUd H r-i ,WHGi M rP vf0l)i0 to co o o rH t o. O o4 4 CO rH C\ CO cm o o* 4-> i <d 'O CinO 0 P o m rH vC * O4 , o, 4 CO CO CM o\ VO 4 H rH rH rH da' 15% 17% 17% 19% VO rH O. S4- 1 VO c CM , in H 1 cd 1J 1 1 o P 00 CO o. Cr1\ O1 o r-4 CO rH 4 o, fc'S 4l 0\ rH m( CM .f^ m o o( 4* 1 CO <r o r> rH 1 vD \D O. 4 vO CO l OV n* o o, S'? 4 cm CO CM sr 1 CM 03 U 000 1 1 O 0 CO m o, 4 o\ CO 1 OV vo O 4 o'( ^2 4 VO rH in > f CM U cd 1 1 O 0 cv CO S' vO 1 Ov O rH o 4 in rH 4 CO H rH o, 41 CO o CO CM O td 9 i <d ro co o -0 vO , 0 o MT to vO i'i C?Ol mO CM CM rH r~{ 1 ,9 9 + 0 .1 0 no d a ta 1 .2 0 + 0 .0 8 1 2 (0 ) 11*9 co rdI <tut o 0 rdd <M0 <di> * n<3U a H u <0 r4 O rH <rDj 0(00)) 5 o i5nM in *H ^3 H on +.1 rQHJ rPtCo CO 00 *H <D H<dd< *<fflJ1 0B30 oNHo X) Q *H M0) {P CO H o. s-? ^1 o Os rH m1 CN P d *d 11 o d ON CO oo ?I 5M CM 525 in rH o! rH rH vo_y nO cm m o o. o * o, 6^ o NO N+O CO m H1 . rH rH OO O o * 4tdJ o. Ctf H rd' On rH o CO d o 11 CO On OO o. O. -H + ON n CO rH NO co * 1 CM CM <r o * o. C-MH vfr<-ST CM CM * H CO p to X1 o a o o. 5-2 fl in CO CM o\ CM rH NO O O O. O, + -H On CM m in rH rH CM O O CO P a. cd + xf ON rH CO o0 o o o CM * o, +1 |i SM O rH rH NO CO * CM CM /"-v CD rH r\ ' ON O v /-n O v/ sh vmw* o V--1 CO O r^. r-- On rH CM m CM f-H CM O CM in l <r m NO St* m j* * I I * 1 o flj l o CO ! fv m W a> m m ON p ON ON CD P CN H > &< rH rH &4 H co <n( ch 'O o o vpHy CO (D *H O0 a* CO f0fl CO k CO o *ri o *rM o Cu TO to fX) "d <u> ."i o > c M a o H to a PS CO ts 8 H rC3 P 'd o to v * siny CO 0 *0H O CO X2? d co 0d) d d o CO d o P * *H a CO d> 0 P O P CO (D 3 id H a U o IW H n) o r d o 50 Q) P. o <5 * p0 H U Q) P0 M "* d p a) p CO >d d p p <D rQ rH 1 9 5 6 -5 7 , A lb e rta HARTOLDMON0029791 lESI'i, 150 M w a), 0 # *rH & 0H JG U i4 o r0C w m +1 03 0 rH 0 rH 0 0) +*~> fin tni W to s0) a0 H 0031 0 o 4J .c to. /"V v--to' 00 sf m o H o tsCM 1 O, 5M 4- CO rH H CO 1 O. 4! <A co o. H in I rH 0 o S' vff l-i MI co m id o 40J 0J -d ! I ?i 10 C/D K Id 03 otf CrHO o 0 0 U 0 *0 CM CM CO rH CO rH CM CM t?-s 5^ Mf CM O , CTx C*M* rH O4, , | 00 52$ rH O, +O co 9H-l CO rH r*v i I rH in VO CM CM CM CM CM 'Kf o ?l CinO CM rH o, +i CrHO CM o> o o 03 4idJ CT\ O o CO rH 0\ <r I1 o, 1 CO H CO \D <y *H & cn w /~s o O v-' vo/ . o tA in m 00 V0 /s *<3* O CO CO 00 *0 0 rH f-* Cv *H vD Mf vO m. U Ch CT\ CT\ 0 rH rH rH rH (U Mfr cr* tn u (U 0 - Q\ (4 rH O, fx CO o \X> 05 *H m 0 0 0 H H O H 0 u w a 0 0 Jn 0 a uo 0 *rl *-t> W 0 !S w flJ So 0) 0 o 0 05 *rl ES H rH 0 xt 0 id 0d Q 0 h *H Ho S3 *r 0 0 O O 0 H rH Pn PM 5 PM r :1 TABLE 5 C o n tin u e d HARTOLDMONOQ29792 to o tc d ^3 O to * * o fr-s m Cn + to 0 s !5d t o H Is H t--t i-t <U fl Vi r-i H a) **-* to mf to to <D o 0d M * to co <u P g u j~ ri .X3 H IJ rth fcO H V/ <3J <y rH sr| p- d 0N d *rH to to *o0 ft u 0) Pw Q) HW u0o XA w *4 <E-* w X* O O wf*< W <u H *0 >U tu Ci Cl rH (0 O 4* > ad to W s to do <N | o *tHo UO rH <u *H 0t$ to Ox *H to rH o oi 4 . o . o. 05 ' 4 4vo CO + 1 rH H t~4 o* 04 i 04 1 in o 1 04 04 04 <r tn o co 0 oo 4J * * ot o. | "0 1 *H +1 H | 1 tn co 04 O oo 1 d m co # O' o rH x4* 04 ro o CO ocr-4 04 r*H 1 ?i m O (4 04 * Y o. ?! 8-5 + CO o* rH rH CO * 1 vO n. tn tn - /*% vO O V co v> vO CO O VO rH Cr o -<r O r* Ox v> OV <D mcr* rH U H Cl g d to oo * D *rt 0 d S3 *H <u U 44 O CO 4 Hd W VO 01 44 O 0 U 0 4o-i 6Q d 44 a, G) o (U 4h rH 3s fx 0) r4 n S~-' H VH O 0) wo 43 4H tos rH rH O 4-4 w (D 4J O d 4J o o MH rH Jr3H (0 d rH & 0) P3 t? d d M <y ^4u rt 4J -W d Oi d 01 to .0 O u ct 151 ' . mm, \ i | i <{ (if 1 ( j HARTOLDMON0029793 HARTOLDMON0029794 West. U. S. (West, g ra ssla n d ) p re -'4 ? 243(38) 4.15+0.04 0.374+0.008 1.94+0.02 153 , w <u bt 0 j X! CJ * to t*. 1-1 o 5-5 in as +1 V) B k<--yi i 0 H rG H H i-l (tf .d 00 rH r--1 O to cn to A w a) o no 00 M CO CO o 0 /> JlA* u H J& H 4J .tco 'wto' <u rH <r\ C00U <u N H to CO o o -f 1 vD CO CO rH 4tOJ <d no o 0 on o o. 4i vO CO CO -7% 11% NS -3% 18% CM O o rH o. o. s4 4| o CM rH On rH rH . <o O O ttJ 4-J o, td A' *d V0 o CO 0 o tn CO o CM . o, i+n A1 rH r^-* 4* CO i CM CM Q. 4 1 to ON 53, ON rH VO CM O * to vO S25 VO CO o vO CO ?! co ON* CO ro * 94*.| so O1 CO * rH VO rH O ?! <3to CO a <r rH 6- S' 1 00* co rH O, s-$ 4 lO CM r-i vO 1 * rH CM CO O o, A1 CM COa o` n r-i 1 <* <* o, 4 CM Mf e CO CM CO O rH A9. a, 4 ON 1 CO m rH rH Mf ON o CM oO * o, A1 r-* VO o <o* CO oo CO CO o CM o, o, A1 Af On rH CO CO o. m <oo s-y min rH f-s. vO rH v-~V o S/ v-/ vO 00 CM <"% ' CM rH N-> h* rH -`n r-. w r~* vO /'-X <? V/ o CO 00 CM t*. r-- vO <1- | m 1 m *1 rtrH\ vO V0 ) ON On ON . M H Hi rH . B. OO ton ' *ooH DC* to ' JHU 1 0>) 55 ! !' : ji 'ii. ] ^ TABLE 6 C ontinued HARTOLDMONOQ29795 154 0 <y o fcCi H a S r~4 a ini wX w 03 o *d dd M a S'S m ON 41 0 a d o rH rH 0? rd CO w 0 03 d u H X5 IH rH U rH yC 03 60 *rH CO 03 rs j 'w&' to ICO CHO *0 O ,*rl U 0> P* y y > rH a > d M d o H to o C4 vO O o. --j eg v4O1 CO CO l CO 1 rH Gj d 03 rH |o d 1 1 m rH o, 5M rH CM c4o1 co CH 1 CO 1 VO-> u0 <3- to Oo ?i ?l H cr\ \0D0 o m H I rH co OH oo ?l o, 41 ON voO co ON rH 1 CO CM o ON m o rH a. 4 ON o |i eg S"S rH eg 1 co o O. 4 eg mt i--i CM 1 co o c, o, 4 > 3 CO CO vO I rH <3H O o. 4 CO r^ CM r^. 1 CM o VO o o o 4i CO vO co o m CM o, SM 41 sr tH CM rH 1 CO On o o 4 CO ON CO CO co o o, &n$ 4| n- CO I CO Mf rH & r^ i vO co /-s CinO /n CO y^s CO MOir co N*/ rH r^ orH co ON r>. m vVO\ ON H o u *rl X b0--34 rt4 S ss rv <r t 0) u p< d o a rH Pd4 S UbO a) a) p*< /> r-> w 0 n00 *r4 d H -rH ec w y 'd 0S at. d 0 M 0 y fN rH 0 P<* Nl d 03 d *j < o *H U S 03 y H 4-3 a nv m l CM m ON rH a 03 d vO ON H d 03 d rH 1^. 1 03 U Ch A. m ! *<r ON rH 03 O V *H UH O H <3 .ts * 53 U o H d 0) % U to dd H to TABLE 6 C o n tin u e d HARTOLDMONOQ29796 IS# I w. a a) u K tH a ,c C43J H rH rH 0) o as 41 V) <13 rH rH 2 CO m WX w a) 0 1a3) f3 H w w <u 4a4 u *ri b-* 4J "So *H 'wto/ 4J "s. rH CX 0 N H CO CO 13 O H UtU ' CM rH rH O, 8^ + tA Cs H in* 1 rH Os rH O O, &$ crHo O o *sf o. 4" 1 O uo rH m1 * CO /> SO CM CO CN os so I ssoo cs rH CN vO O o o, +1 9. 4- CO rH o & OS SO 1 Cs OS rH , H rH in rH o CO oo O, + CO o, 4"1 so C5O5 O S^S cn SO v> <N CO CO CO Oo o CM o rH ?! ?i CO CO CM CM & n* CN CM SO * <* CO rH Cs O O o. o. 1 4* | CO os* rH OS * m 1 rH *H CO O o O, * vO CO CN *<r o A CO CO 6^ CO I in 'O *' o, +1 CO* H 4<J3 Ctf *0 O CN o CN ot +1 i cs LO CN CO rH * O. i1 rH * CO CO Vin/ f". rH Os m "s VCN CN Os CN CN w so CN CO rH m ***-% O rH rH <fr CtoO V I r- 0 'CT U OS P, rH SO S0 Crs r*-. CO Nf* *. J 1 no <r u Os o< rH O in 1 Crs <} 0s rH HARTOLDMON0029797 1$6 I Wa Gu t1 rGH ti H js o o ttGCo 41 w G R! w rH rH CD Ww tn| 0X <u *G dd H mw <u d jEvH{ V-' H U fcD H G /*N to v. *4 G N rt CO o H U0 PH CO CO co CM o o co vo o H CM OO * 1 tlS' 25-5 O, M? O. o: MS o. o, M2 v> 0 CO CO $' A* ? str {25CM r-{ m* fv toI r- 1 co r- CO < CM ON rH rH rH rH rH * rH rH sf to On vO o O O oO 2?! M o. ?!$ ON H 1 to rH On rH i ON to CO CM ro CO o Oo co ON N to o rH * 2 2M ?! M o. ?! 4vO CM A CO | rH M NO 1 i-- H CO i *4 * co co %6 8 % VO O O ?] co <r co o 00 o e o. 4-1 vO ON CO 4G4 5G T o d Si* co O, +1 CM M- CO i i 5w3 to -o O o, + to p* CO o to CM O 2 H CO O <r o ?I ts* * -4 o CM o + to CO * CO 5-2 rH I M$ lCO\1 I 00/-N /*N rH /*\ CM M* rH w w 00 On CO f"" rH 4CM * to <r to 1| **1H CM 0to to ON M ON rH P* rH /N -4VO /-N o w to 4 O rH t~-> sir 0) N p. C4O ON rH 4 ON V-/ CM ON CO /^N co VO C*- -4 1<k> . o to 0 r4- u ON p* rH &x-l tl GP* co rG wl < * CD > rH O *H ud * >a o 5 4-j G M *H CO , CO o * & d o *vH to CD C4 rH G a * o CO <d A M S-i G 44 0) &' rH r~i <3 < GG *o. G GPO fO . d * G 44 G GG GG W TABLE 6 C ontinued HARTOLDMONOQ29798 TABLE 6 C o n tin u e d NS <(wo0tujc ,A ,*u0H H s rH o rXCH0O mox 4* I <0nU3 ri--<HUI rC CO in| oo0 0K) -0 wp <m(Ay $i i.a X4CJO /6s0 H <D 0 r--j 0< <y 0 CO 00 'toJ O <u PH *o CO oO o, o. ot A1 g o O rH H tH O O id u o. to A1 T3 HO CM p o 11 rOH vo o ?l ?l co in in oo rH MO O o, VtO o rH , Ox O o o. +1 CO C$O2 o CM o in o o, <4t J to 15 o r-1 C-- o ll to 1 CN O 0 4H *00 00 r^o 0 o + VO Cv # o CO o o. 1 CO #33 o H CM rH O. -f I cn ox o 5^ CM rH 1 CO vO o ' rH o O ?f . O* 1 O, -H <T rH rH CM CO 1 CM rH 0 oO O S' CP5/i o> o o, S'? -f n rH H Cx 1 o CO o o, H in CSV o o rH 1 0 4-1 0 TO o 0 CO o ?l rH CO rH Ox 1 * o CO CO mo w rv n o VD 'w' O w rH V/ n CO n O v-^ rH r-> rH MT CO CO CO in tH rH H rH CM t--{ CO m cn rH CO 00 n VO VC . n 1 r-. ) j CO <3* rH CO 0 <3- cx m m CM n CM n u crs rH Os cn Cx OX tx rH rH rH rH rH *0 O O Pi V-/ 05 <U H O O 04 C5 H0O CO k 0 o H & *H d O t<0oy4 Pm . "d t>u H O > d H od 0to c Ma> G X* u o o 1 J-4 O H H CU U0 M 40J O K 0 s-5 4 0 U rH <3 4M H rH0 o o ** X0 ' E03C 4H *rl rH ,0 o o $3 . 4-* rl rH0 o o 000 *H "03 M . r. s \ rHo r0H o0 oo >* ** u M X .0 ^3 0 0 to to 000a H HART OLDMONOQ29799 oTO I TfcOC.j H xTdsO H o t-4 H <3? u ,0 to &s? m 0N -H w 0<0u rH rH 03 rO CO *n| wX w 0) o *0 00 M CO (0 o a o H .0 H 4-1 ,0 60 60 -H G) 3* <t\to rH Pm d 0 0 *H to CO O VHl P0H3 <r o oO i<o, t+o oo w 5 <O? o. Co+M w sor * o, 41 mt O o ?l mo t5o3 m o o |i 400J *o *H CM 0o ;i1l 4c0Ja -d 0o i i 400J X) 0o ,11 4TTJOO -d^ft d 11 O rH ON m CO OO o o o * O. o. + CO + 1 CO 3 Ot + 1 CO ON* 9Hr.I \0 r- f o + CO to oo ON ON o rH rH o o rH r-N ON <r So-'' >O--' Vow/ so-y CM rH rH <r rH CM rH CM ON CO tn *1 1 in 0 in M ON P* rH * rNO 1 m NO CN tH ON ON in 1I N CO in ON ON rH rH o ?i o, A1 to CM O o to o o ?l CO 'o CM 4004 d o0 rH VO Oo |l o. + 5CO3 ON On Oo *~N CO 4--S \000m voy o NO CM r*- ON m * i Xo om u ON sx rH lCO 'O /^ o Pom V< o u <T3d3O TO u R o u VO w 4 to 0 *ri OCl (X 0 0 to Ni rH H 0 O *rf U03 to T-j T>O rTO O > d H d o TO . TtOo u00) *H x o Q> d 55 * * 00 O u *H <e1 <M4 O o *H X(11 JSJ sts !3J td 4UO0H-t rf r0H O * ' 0o3 0 *0H Vi 4OH *H r0H u o to TO 3 rTO d 0 0 TO- *d dj H 0rH 0. O TO H H Pm pH tXi .- kmtA HART OLDMON0029800 . 15? HARTOLDMON0029801 TABLE 6 C o n tin u e d . TABLE 7 . E g g s h e ll d a ta from N o rth Am erican museums and p r iv a te egg c o lle c tio n s , showing 1 in ] 44 w K O to 0) *H 0} "O Si Ci HM w w 160 CO O ' O. &1 CM cn cd 4J <d -U O r-* rH in O0 1 1 CSO3 Hm 00 , 0. 0, + CO &1 hs VO * rM tH o 11 Ch in CM CM O rH OO roo**. O, I*mf's CO co CO 525 co VOO m CM CO to o.. S3 CO CO O Cn r*^ VD in m o\ VO <r CO H h V/ vy 8 vO m *4* to 0H MU<U f<''-r 4* iH 00 1 CM u<$u <T OHh CM CM , VU* CO CM Cm1O rm~S . tOHh CO CO V M 44 <d M P Id t$` 0 G> UM OO 00 0) M . td h? U td CD M O HARTOLDMON0029802 161 I to * u to <0 *H 0 0 cd IH H tD 4 rH O rC0CO LA on CO <0 0 M v *H tO d rH U r<HD ,6C2 ^60 rr: H ^ to <i> 10% 1% CM o ?l m in i rH rH <r Oo O. o, &> v0 * rH . rH S'? 1 CM O o, i1 CO rH V0 O 0 + 01 VO * rH CM CO o o oO oO o oo l`l s-s o ?l ?i O. O , SM + + 1 00 i H CM CO . i + CcOo 1 00* V0 CO CO CO 00 ** o oo oo o rH* m rH o CM VP+0i| m 0*0* i , OC+MN ?l oo ON* in | m VO m rH rH VO CM o. C+M m ?l CM rH VO vO on 0 rH vO o o VO 00 o to o *4* *4 0 rH P0 gN d H CO CO o OH P 0 P-t "A tt~ PooM CO *3 a) *H u 0 *0>0 d *4HJ P. (0 H o rA d 5 CO d O o 0) od 0H Pit*- 1 *r4 0 60 0 3 rH 0 *H P P9 CO <d h Pm rH CO in rH rH 4 /~s rH in srts/ vy ON CO v/ O 0 VH ON * On MJ" rH in CO CM o rH rH ON ON CM VO | <r vO * 1 4 vO 1 O i CO 1 -sf vO 0 VO 0 m ON P ON u ON H PX rH p rH 44 <U at to to i dd <3 < to o o 0 oa 4 oG 0 oo 23 5 H H H 4J M Oy P dd -H ri d d d. 4J 4J a). rH rH UO 44. 44 MH O< <3 <d -4 *0i P P* <0 0 uo J 4J d 0) P o to to to d a H I <u d u H CO o S <u MM H . *H rH rH dd *ri pj o ffl0 0 a01 > 'D * :> o to l Hco <\ / HARTOLDMON0029803 162 1 CO <D o or. G td Ch u * >0) fH >4 o rH (D 4to5 lA On 41 to G 1--l td --i 0) 4j CO *nl to <0 <XD o GG H to 0TO a 4O<J *H w1 & 4J G 00 JV a? rPHu <0\ d <d to *H CA in o ?i CO 51 ON rH <0 4J XtdJ o G rH ICO 1 VD <T co CO o O o, o4 r4- 4 o CO 5 On 4 ON H r-l ton o Rj W o. td O4N|* X3 V<Oo4 O 0 o 1 1 t^ O o 4 CN o. o, it1 <4r CO f4 On* CO <T Nf --> /r*N> /% Vo-/ wVO so~> rrHo CN vO ON rH rH ON tn1 hN CN vt m . 1 Io -d* Q> m ON U ON rH P< rH NS NS CO m o 4 o o. o, <4r C4O w $3 ON ON rH H tn o o td U o, td 0401 vD XJ o co a o I1 o CN H rH 4 S' ?! m V^* CO /--S vKo_frx /--'N o CO ON vNOT rs VO <* r vD ! (iU r-* nr u. ON Pl, r*H CO CN o rH o, O, H4 v4dO' ON CN rH 00 CO o rH o O ?l CV h- O, m4 VO CO# CO oo o CN rH CO* S' o4 * CN ON m <1* /<Nr ---s CO tH wrH 'rwH' m f-> ON rH rH r** t"vOI U iOnN rH /-N col A Xoo? spM/ TO U *rl O O Of w rQ G C/3 * ToO rl O <y CO X# 0) > roH > B H G O H 60 O td G O iw rl rH d o o CO o a -H 3X 4fc- to 0) cd X <y 8K -f H 0 G rH *rl cd CO d cd P0 (U XJ u *r4 cd O <u OM OO ,* 1 CO u Gu <y 0 o CO CO . 0 <D Q rH rH d cd a* 0 u * to > * 43 td 4-> a O o *rl X CD ^4 0) ^25 TO 3 G td *aH *H fcC }4 -H > J TO rt> . rHv +TOi ,, ' : 0 X) H O 0 0 > . u 5) i < * $23 . 4J m S? s T3 VJH 0 *tH <D 4J r-0*i fp ' td 4J rH < to u p #, <d frdj M do ^4 HART QLDMON0029804 163 i m o) W H a .d r* H CJ to rH rH rH o BM tCHoO m on *H tn cad I Jd3> CO m CO A CO <u i) nd cd M <0 <U (3 U *rCHS H P C> to H P C\ o 1 n00* rH to rH o VO CO o f-* CM O + o CM in <U rH & cd co CO rv./ r~ Td O *H P O fH CO VO | <r in ON rH -o o o f* CO <y tH U 0> CX; C0 o & 2 CO H to CJ O <y *H *tHo UQ (cUx #5 0 CO pH O u td p d 0 <0 00 1 1 0b vO 00 vO H , rH bm CM rH ( On O O p ?1 cd *a in co o O VO rH rH I1O , S* v+n1 r-l H 0 in 1 * m <r vO o &M r v0 H in l rH td P cd *d rH CM a, H ON rH rH rH 1 ON O O rH <r On 0 CM 0O 0, P, C4M 4n- OH CM CM V O0 CM O rH a. O. + n- H| <0 rH co rH O CM 0 'w* V/ nvO CM * <3" ** 1 0) P cx CM m l ON -3' ON rH \Om*S CO CO VO 1 f*^ m On rH m td H d *H bC P cd > H U O rH Cn d H . P O rH fH d d rl U O rH P4 H 'cwo' cm in CO rH Is-t--. "l f" <r <U o> V( rH P Vt U C Og . to g0 O rd uO > d /N ,X3 CO P *n xs mO 0 cd d P &JOr* jj 1 "0 e cd r-t Ai Vi cd n. n 0 a. 3 VI to *ri <u P H O <D P te P- rH nw dd x> (D P P P-t P S3 m J03) EPi jo! -O 3 H 0 u U| o o. TABLE 7 C o n tin u e d HARTOLDMON0029805 TABLE 7 C o n tin u e d -2% NS l6h i wM <u (J fcf. art H u .w *H rH 0 o 4CO2 iNi iOnn ml w M <u O 0 "Md2 03 CO <U 0 u 'w' 42 E~* co o Hr-H o, ^8 in Sl 4- ON o pd cd Cxt--* 'd o d l l ' rH CN O O ?t sr O' O, r4HON w !5 .O o ON o ' o ?! r> ' ' rH ch* o cd P d o d 1 I CO ?! ?!o O CO m O' CO 03 & oo m o o d u CL o+ o CN4 d 'd oa o m. i o o, tC4HY CO ' S3 - O CM O o, .0O40N o VO ?!d u oo d X5 1 1 UD Oa o 1 ?! S'I I 41 CD d rH 1 CO 4J 4030 *H Vtwo/ co o o, co 41 tH 4- rH* rH CO CO o* o O' m o o vD O ?! vO o, C4O 1 CO rH* rH rH rH CO rH o, C4N 1 in 2 ON o o, o rH o Hm rH 1 d N' m O v r-H vO o 'w* v> is>. in r-. -s o v.-* w' *<r On CO rH CM /***> ON o < Ny- W CO ` uo CO CN| ~a o H VO o1 VO o ON pH rH S- 1 ton <y ON u rH p< ON ' *cr *- t 1 CO o <T . u . ON (X rH vO LO CO uo ON rH r> <? m 1. <y u d-s dd d H H doto a 4o-f 4dPoH H *r-{ wH 4dOM4 *H 0) ou taHCO rUdH ' rUdH M CD 1 C4O2OJ HARTOLDMON0029806 165 400CGb3O3)e i uH h o * CO * o SN Hi4C--<O2U! tons 4-1 0a3 H s r4C<HO2u 0t<aoD3 m*MQA00) 00<d3u5 *4HOH2 6 4*&40tH24o> 'wto' o Co+N* rO-( + rH *40004 Oa 11 o v?CN>! <oN! rH + rH <U -sr 1 fX d H 00 CO o Os *p*<UOH04y vmo Os OH OCN oo*i O' CC+Os 1 6CM14 rH o cn oo* o. <+t rC--Ni *40d4 o0 O 11 oCN OO*r Cc+sV1 ?n.l rH0 6U"I4> H r-i aCN CO9tOsi CO o *t(oMo0 oa <ot0 CCNNi Sto3 rH CON 0# 64 + Os t O 4fi4S *0 o0 11 oo o. O+S " <EN1T> rH rH ' Ho o, - CO+Os 1 o 0o0 o oof rH mo o ooC9N. n400o4 Od o O<N CoO O, SCtNfi iC9+nN0 H rH in o /*-s /*S rH o OO v> S-X vO o O St cn cn St VP CN --N H rs CN O w Os Os rH n- o Os O' sr in vn VD 1f1 r-> rH O 03 <r in VO U as OS Os cx rH rH rH 1^* St 0- CO | o Os u rH (X *% -d *4gH4 g o wj 'dOO K *0<O0fHUx5) *<0u > 4cC0n*O o w r*HH *CU00cH3J)x g to fj o 60 e60 &OH <u m0 4o0>2s 4140 a) rHtf 4M2 44dOd02 43 4h00W4^) 44 OCud) a* o4oHRHJS 6oMrM$l a o 0 MtrSM-t O0 r0Hd3 2 l <0 2 o0W03J. "i0dH 43*.| rUoHa O) 0 r*MOHH Pw N PM HARTOLDMON0029807 166 <or , f-1 CO tA PS cs oCM * o, ON o. r4tC0-oO* 4tcoJd do OoN mo # ?'l CO o. *h1 CM 5c2n5 CO CM rH Oo Vw' CO n CM ro om1 NIO ON tn ON ON rH n<MOoH0 rH t0 rHO Pm c6rt0 4> iHU1 to t0o H0 CU XI eto <2Ou ' <0 vO <> S t<o11 4od-J twoo f- ON r-1 Cd| 5o4 uMoutt) 00' <ul *4P0H-J>-. Ftumoot 4> ns Ft <S a60 (X I CO 0 rH nS Ft <u *d d . x) *H3 /--N 0 r- rH 4M O ON rH rO H (0 ttooy M 0! N to O P U to 4J *ri a) HO <U U i4 u d0 u rMoHM ns Ft d tu 4-1 4-1 u<od0Uu) 4QO-1 Ft o m Ft at ns Ft < *H *0 4J d d 0 *H 4M 0d HrH O W to *H & CO NO C- A3LE 7 C o n tin u e d . HARTOLDMON0029808 TABLE 8* In te r s p e c ific d iffe re n c e s in th ic k n e s s as i t r e la te s to egg le v e ls o f DDE o r B B T -fa m ily re s id u e s ^ ' found in f ie ld specimens. 167 <ri <5 *=4 0O C"-~ t- Os rH On rH *. (0 to 'd CO d O H NO H C-- | ON Os , O On H MD & H 0 Os . H Cl) n on vO OHN P0 H drtPi O ) to ttJ s oa Vi On 0) to vr -d Vi Vi Vi a> P0) X0 ri t) t! p 0 0) X) M 0 ri s rt~tfi p0 P cJJ pP a CQ 0 to 0 P P N w d to 0 00 Xl 44 ri 0 Vi 0 03 ri Vi O <ah> 0 10 H W 4 $ CJ w IP 8 co CD O t-~ CO O p0 On sO r- H Os Os p H rH 0 O A to (0r- ** O Vi 0ON -d d t>- HH O 0 co 1 d d 0 Ri a> H > 'd On H *Hctf -p * <D rj C! at! to 04 d <2j NO ja NO P b0 On 0 U0) V? 0 CQ Ua) 0 Vi P> <4* P ,w0oh), $M <D 30) w do O(h 0 to ri P4 H ,3 PO--i fca44> On rri 1 1 On 1A VA NO CM CHM CO-- C-- no CO On m Pi 0<0n rri $ tOR VI H CM, W. W. Pf H H CO . i . r~ti rrii l VS. rH CM1 VH*. VOI HI ' W, w Pmtt 'CLIAA m OHtn 0 MP 0 d Vi ri Pi to Pi a> <P <4 iH a> > <D rH * ^ ^ CM C-- 6 1 CO rH CSJ 1A A 0 XA C~- E-- CM QA CM csi Csi On CM CM U^ H CM CM I r- NO N-yO Wto o On H OiHn PcPTPOtOS S -OPH d r<HPsj Hp00uu0ri R) 8 6 CO -5 dcUOQS OHO V 4R35 p *H4 VPS0PVOriiii 0 jobdJ3 CrppRp0rii ctooo to S a o 3 H fi o a Wri aoo to ri 3 CO CO H 'A rH Ctf 0 P p- dp to w Q> 0 p P to *p rH ' *drl ri cd Vi ri 4j O O O O txO r) d O O 0 0 0) .. P u b<3 CO (0 00 Pj 43 d Oj 2 SCO' i O cd CO 0 0 ' 0 p a> 0 r) l-J rH rri Vi d <D 0 W4 Pi HARTOLDMON0029809 d P(1) 0 H d HqqO0 $ ?>> 1 1T5O O bdfl S' H csccdoo r a) H * d Icd P s to d to txO bfl 0 d a) o q 40 co bbffli 0) Oo <D & p0 dp 0 o p Xf a cd 4> 0 1 HQ> M o 1 OHH PO <R 0) u is to OCO qH dCdHO 3 is dm 4K0 X0) t Ro d q& to 1 to o 4P3 Pd0to) Sd & ss Jj4d Jjoo oo &M qo A dH O H PO a) c *H 0 &O cuod J- o *u3 3 T-cd0Pd*) fO-i S3 o 10 1 5 H. a p to 0 5= M qCD i>> *Sgr| <P Pd P qo Wpp Atoi qH p0to *c3d p*4r5l IS dtt) x q X S3 m .15 H rR qo to oto a 51 a H. to 0) A- u '43 a>do) o iqs 43 4qr3l l P O Pto Q G 0tmHo PI drl &H H GP P to d to W f a> to w erf apH 44*Or33i *O pcao> t0Hdo) q aH p r-- PO0to) po SR3 vO do "oH to >qCOPD bh<1>Dfl O dH ^ ip a) >a) H 0H dH SH3< U O d o H P0 H d H Xoq ddHd p &0; Hqd qp q to ffl H0, > 4PO3 O' -d OHn <qOn TJ 1PwO0)< 0CO IS H <d g H d i 51 4p3 to d q 403 Si rr\ p fH H| p43 1J5e CMi sS33 \t^l -dl 168 HARTOLDMON0029810 169 appendix 1 a i HARTOLDMON0029811 170 GEOGRAPHICAL VARIATION IN THE EGGSHELLS OP COMMON LOONS DANIEL W. ANDERSON1, HARRY G. LUMSDEN2 AND JOSEPH J. HICKEY1 Upiv. of Wisconsin, Dept, of Wildlife Ecology, Madison, Wisconsin 53706 2' Ontario Dept, of Lands and Forests, Research Branch, Maple, Ontario. ' INTRODUCTION Our major objective here is to report geographical variations in the eggshells of Common Loons (Oavia immer). The data, besides being of interest from an academic viewpoint, have a potential practical application. Numerous authors have recently reported eggshell changes, namely in shell thickness and weight, in certain bird,species. These findings dictated, but not necessarily so, that our cutoff date in analysis of "normal" geographic variation be no later than 19^6 {see Rateliffe 1967, and Hickey and Anderson 1968). INDICES TO GEOGRAPHICAL VARIATION Rand (19^7) has used two criteria to report geographical variation in the Common Loon, bill length and wing length, the latter being more, but not entirely satisfactory. The wing-chord measurements of Rand (19^7), although somewhat variable, were adequate to. suggest geographic variation in size on an intraspecific basis in' North American Common Loons. Despite potential weakness in the use of wing chord as an index to body size (see Welty 196L:b57~58 and Scholander 1955) it seems generally satisfactory (Mayr 1956 and Araadon 19^3a), especially on an . intraspecific basis. Since many readily taken eggshell measurements (such as. volume, thickness, and weight) often seem to relate to body HARTOLDMONOQ29812 171 size on an intra- and even sometimes on an interspecific basis (Romanoff 1 and Romanoff 19^9:150, Amadon 19^3b, Lack 1968:235, Anderson and Hickey 1970), most of our measurements here probably reflect body size. The general relationship between wing chord and body size reported by Rand (19^7)s then, provided us with a basic hypothesis to test statistically, applying our eggshell data to an already hypothesized trend. MATERIALS AND METHODS Eggshells were measured from various museums throughout North America as already described by Anderson and Hickey (1970). The basic variables we examined were as follows: empty-eggshell weight including membranes, egg length (L), egg breadth (B) at its widest, portion, L/B ratio, estimated outside volume, and eggshell thickness including membranes and cuticle. Tyler (1969) has described in detail the shell structure of Gaviiformes and it is obvious that our measurements of HARTOLDMON0029813 r 172 were as follows: subelliptical = 0.64, oval = 0.62. We multiplied these by the volume of a cylinder of equal L and B for each egg to obtain our estimate of volume. This procedure is essentially similiar to that described by Stonehouse (1963). The hypothesis which we tested is restated as follows: (l) geographic variation'in eggshells occurs in North American Common Loons, and (2) it occurs according to the trends (wing chord as an index to body size) described by Rand (1947). Statistical tests were (l) analysis of variance, and (2) Duncan's new multiple-range test (Steel and Torrie 1960:112-15, 107 09). The hypothesized trends are summarized in Pig. 1 as we vision them from Rand's descriptions. RESULTS AND DISCUSSION Eggshell Measurements All six variables tested showed highly significant differences(P<0.01) between our various subjective groupings used to test for clinal variation (Table l). The F-test generally confirms Rand's (1947) description of variation in Common Loons. The Duncan's-test, although not entirely satisfactory because of its one-dimensional character, yielded additional information regarding a description of the variation (Table 2). Volume comparisons (Table 2), for example, gave 12 significant differences in 21 tests, whereas only 1-2 would be expected at the 95$ level due to sampling error. Eggshell weight, volume, and thickness generally followed similiar trends which were much alike and in accordance with previously described trends. Since eggshell weight, volume and thickness are generally interrelated parameters on an intraspecific basis (Asmundson and Baker 1940, and '3. HARTOLDMON0029814 FIGURE 1. Map shoving trends in the wing-chord measurements of Common Loons as described by Rand (19^7)--dotted line. The breeding range of Gavia jmmer is adapted from Palmer (1962:27) and Godfrey (1966:10) with modification from the AOU Check-list (1957) concerning California. Circles indicate, subjectively, relative differences in wing chord, and arrows point to increases in the trend. Solid lines enclosing geographical regions represent subjective groupings as we would see them for statistical testing and as best fitting our sample distributions. Eggshells from area 5 represented inland nests, and Ontario eggs from the Great Lakes were included in area 4. Areas of especially noticeable overlap in measurements, according to Rand (1947), are between 5 and 6 and between 1 and 3. HARTOLDMON0029815 17k TABLE 1,---Analyses of variance, comparing six.eggshell variables in Common Loons over seven geographical areas . Variable Name Eggshell wt. (g) Length (crn) Breadth (cm) Length/Breadth Volume (cm ) ' Thickness (mm) Degrees of Freedom: Among Within 6 295 6 295 6 295 6 295 6 ' 295 6 239 F-value'*' 20.44 7.77 22.86 5.33 23.31 7.15 . F~values at various significance levels are as follows (Steel and Torrie 1960:440)! P.05 2.10, P.01 2.80. HARTOLDMON0029816 TABLE 2 .-- G eographic v a r ia t io n i n fo u r e g g s h e ll v a ria b le s o f Common Loons from seven g e o g ra p h ic a l areas Means n o t s ig n if ic a n t ly d if f e r e n t from one another share a -common l e t t e r in th e " t e s t '* column ro * (oa0> mu<<0uu 44 a xj 00 10 44 p aoa0)t Uoo(HS3 tn ON +| aei nH2) fW-4 > 44 to 0 H to to 0 0 V H H H4(M-U1 a) I rOH > 44 CO 0 S-4 ' PQ P 44 CO <d H 44 'PbO 0*) w Op p pq <3 u PQ &04 fM w o m < <J --1 r--1 CM r-4 r--{ CM O o o O O * o O o. +1 <3* o, 4* CO o, + o o, +1 rH o, + CO +1 m vO vD vO NO VO VO VO sO Oo oOo o o w U P3 tu fn w po opp M 4 m si- CO o Os ' CO, O+N 1 CM. P CO. t n co. A1 + (O fsni om CO CM o <r CM VO MJ* r-4 r~i rH r~{ rH o * 4co o n rH w M P U P <3 Q O PQ <3 Op w p Osf- cm oo I1 o, 4- m CmO vO OCO CM co cm o oo S'I' m vO o. o, +1 4-1 vinO H a Pn PQ M W8 Q (X| P P3 m * o, c+o n+ VO rH nP cn in O. O. o, 4- co A1 -hi sf m vPO --{ o, P ys. CO CO CO v-^ CO s* CO CO 17$ to , il CM * <* O c<d4 < O H rH O 44 CO to P0O 40a4 o p 0B CO &4-1 PfcO fM P4-4 P 0MUcOUc0ad3dd) a P X) M0 <U s 01 td w rt a) n -4 53 to ta *aH 525 P po <d a S3 II V0 d o P <y a 4U<-1d P dM 44 P4*4 P0M CO *2n od to MO *M CM d o CM II VO C^v ?--{ in $04? B O0 tc--0uI o 0a) to PC2O 1 Pu P P <0504u o g u0 0 P 3 4004) 0 na3 .0d cd P U0 H 0 4J P X) r4*-4i 0 o0 P * td .w M 52i- fn r-l 0O CO too q Hg MO 44 0NW to 0 r*4 CO tHo W0) x4C0JOl) a U0 cd 04 /s 0 0 rl CO 0) 45 O 44 R CM HARTOLDMON0029817 ' . 176 Asmundson et al. 1943), similiar trends are to be expected. The general uniqueness of middle groups (geographical areas 2, 3, and 4, Fig. l) vith significant differences moving out in both directions from those areas, suggests a linear, but branching variation either converging or diverging from that area, possibly related to latitude and Bergmann's Rule, Even though, in all cases, the values' on each end of the spectrum were not significant statistically from one another on a ranked basis, we believe them to be biologically significant due to their geographical separation. Perhaps two statistical tests would have illustrated the data better, using these interior areas as a "starting point" for each. The patterns observed with weight, volume, and thickness seemed nearly opposite to those observed with L/B ratios, but by no means were the differences large or of sufficient magnitude to yield noticeable differences except from the interior areas (5 and 6) out toward each extreme. The large F_-value for breadth as compared to length (Table l) suggests that B is the most variable. Somewhat different L/B ratios, in any case, suggested possible differences in the shape distributions from area to .area. We tested this hypothesis with 2 Chi and found that significant differences occurred in a p' attern ' ' generally similiar to that observed for differences in the L/B ratios. Areas 5 and 6 (Fig, l) tended to have significantly different - *' distributions of the two shape categories' (area 5 was significantly * Ir different, P<0.05, from areas 2, 3, 4, and 7; area 6 was significantly different from 3)- L/B ratios (Table 2) from area 5 were not significant only from adjacent geographical areas, 3 and 6. When areas 5 and 6 . were combined and tested against the remainder in a 2 x 2~table, it was ' found that their tendency toward a larger proportion of subelliptical ' ' 4. f l 'i . > t:' f-j '|i--; | j Vj \.j. . mm HARTOLDMONOQ29818 177 eggs was significantly different (P<0.01, Chi2 = 21.3). The mean ratio of subelliptical/oval in areas 5 and 6 was O.56 as compared to 1.7** in the remainder of the areas. Since areas 5 and 6 tended to have the higher L/B (Table 2; admittedly, there is a great deal of overlap, and differences are only slight), our tentative conclusion is that different tendencies in L/B ratios must at least in part be related to different shape tendencies. The loon eggs with the greatest L/B ratios > were the eggs which tended more toward subelliptical and those with the smallest L/B ratios tended more toward oval. Clutch-size Much discussion has been presented regarding geographical variation in clutch-size (see Lack 195**, 1968; Cody 1966; and others). There seemed to be little difference between the mean clutch-size between most areas, except in our samples from area 5, where differences seemed apparent (P<0.05 differences from all but 1 and 2). There appears to be a rough trend for clutch-size to increase out from this area (Table 3). When areas 1 and 2, and 3 and ** were analyzed singly, no differences or even trends could be found, thus their combinations in Table 3. Unfortunately, we had no data from Greenland, Baffin Island, or Northwest Territories to test for iUrther changes in the trend. Mean egg volume showed no relationship to clutch-size due to the variability in clutch-size. We do not know, however, if the clutch-size data were biased by failure to collect one-egg clutches in some areas. Such clutches are reportedly fairly common for <3. immer (see Olson and Marshall 1952, and Palmer 1962:31). We doubt that our Iceland sample, for example, truly represented the tendency for a lack of one-egg 5. HARTOLDMON0029819 178. TABLE 3.--Clutch-sizes of Common Loons from various geographical areas Geographical Area"* Sample Size Mean Clutch-size 95% Confidence Limits 1+2 3+ 4 5 6 7 all 23 65 58 20 17 183 1.87 1.92 1.64 1.95 2.002 1.84 0.15 0.07 0.14 0.10 0.00 0.06 ^See Table 2, footnote 1, 2 All clutches we found In museums were composed of two eggs, therefore the zero values for 95% Confidence Limits. Host of these eggs were collected by one collector, and might be biased away from single eggs. HARTOLDMON0029820 179 clutch from Iceland. We believe that additional data to increase sample-sizes are necessary before conclusions can be made relating clutch-size to some geographical gradient or to egg volume in loons. There is good reason to believe that these relationships exist (see Lack 1968:233-34 and Mayr 1965:325). ` GENERAL DISCUSSION The data presented in this paper confirm geographic variation in various eggshell measurements of the Common Loon. 'The eggshell measurements which most likely relate to body size showed the same trends as described by Rand (194?) for other measurements relating to body size. Snyder (1957:26) recognizes a "limited" correlation between size and latitude in Common Loons, but stresses that considerable _ overlap occurs in intermediate areas. Todd (3.963:75) discounted any evidence of a cline in this species on the basis of bill and wing ' measurements, but suggested trends, nonetheless. He did riot believe r that smaller forms to the south should be recognized taxonomicn.ilv due HARTOLDMON0029821 | I ( 180 eggshells from this area). The isophenes (see Mayr 1965:362) of, say, egg volume might be visioned as shown in Fig. 2. We regard the clinal variation described here for the Common Loon to be "ecotypicl! as described by Mayr (1956, 1965:^15) and therefore probably of little taxonomic relevence (Mayr 1965:363). It is interesting that in the area of proposed isolines outside the breeding range of Common Loons (dotted lines in northwestern and north-central North America, Fig. 2), another species, G. adamsii, breeds (Band. 19^8). Palmer (1962:35) reports that G_. adamsii is the largest of our loons, but eggshell measurements are nearly identical, comparing the two species (Palmer 1962:31, ^0). Godfrey (1966:11) does not believe there is sufficient' justification to consider these two allopatric species as "mere races of the same species", as proposed by some authors, and it appears that eggshell measurements might help justify this conclusion in demonstrating different'body-size/egg-size ratios. Studies concerning the eggshells of G. adamsii are needed. ACKNOWLEDGEMENTS . This study was supported by the Bureau of Sport Fisheries and Wildlife, Patuxent Wildlife Research Center, Laurel, Maryland. Lucille F. Stickel and Eugene H. Dustman advised us throughout the study. We wish to acknowledge Austin L. Rand and W. Earl Godfrey for critical advice. We are grateful to the many private egg-collectors and museum curators who aided us in obtaining our raw data. Keypunching and computer facilities were provided by the College of Agricultural and Life Sciences, University of Wisconsin. 7* HARTOLDMON0029822 181 FIGURE 2. Hypothesized isophenes for eggshell volume in Common Loons, assuming that the dines as we interpret them generally cross the isophenes at right angles (Mayr 1965:362). This figure is only roughly sketched and on the same scale as Fig. 1. The values given are mean volumes from each area. The value for Baffin Island is strictly hypothetical hut assumes that this deme, having the largest individuals (Rand 194t) will have the largest eggs. It HARTOLDMON0029823 REFERENCES CITED 182 AMERICAN ORNITHOLOGISTS' UNION. 1957. Check-list of North American birds, fifth Edition, Baltimore, American Ornithologists' Union, AMADQN, D, 19&3a. Bird weights as an aid in taxonomy. Wilson Bulletin 55s164-177. * ------------ . 1943b, Bird weights and egg weights. Auk 60:221-234, ANDERSON, D. W., and J. J, HICKEY. 1970. Oological data on egg and breeding characteristics of Brown pelicans. Wilson Bulletin 82:14-28. ASMUNDSON, V. S., and G. A. BAKER. 1940, Percentage shell as a function of shell thickness, egg volume and egg weight. Poultry Science 19: 227-232. --------, G. A. BAKER, and J. T. EMLEN. 1943. Certain relations between the parts of birds' eggs. Auk 60:34-44. CODY, M. L, 1966. A general theory of clutch size. Evolution 20: 174-184. . GODFREY, W, E. 1966. The birds of Canada. National Museum of Canada Bulletin 203, Biological Series 73. 428 pp. HIGKEY, J. J,, and D, W. ANDERSON, 1968, Chlorinated hydrocarbons and eggshell changes in raptorial and fish-eating birds. Science 162: 271-273. LACK, D, 1954, The natural regulation of animal numbers. Oxford University Press, London. 343 pp. .` --, ' 1968, Ecological adaptations for breeding in birds, Methuen, London. 409 pp. Mtf.jacga-v-. HBRK.' .. 1 HART OLDMON0029824 B 183 , OLSON, S. T,, and W. H. MARSHALL. 1952. The Common Loon in Minnesota. University of Minnesota Museum of Natural History. Occasional Papers 5. 77 pp. PALMER, R, S. (Editor). 1962. Handbook of North American birds. Volume 1/loons through flamingos. Yale University Press, New Haven. 56? pp. PRESTON, P. W. 1953. The shapes of birds' eggs. Auk T0:l6O-l82. RAND, A. L. 19^7 Geographical variation in the loon, Gavia-immer (Brunnich). Canadian Field-Naturalist 61:193-195. ------------------- . 1948, Glaciation, an isolating factor in speciation. Evolution 2:314-321. RATCLIFFE, D. A. 1967. Decrease in eggshell weight in certain birds of prey. Nature (London) 215:208-210. ROMANOFF, A. L., and A. J. ROMANOFF. 19^9. The avian egg. Wiley, New York. 918 pp. SCHOLANDER, P. F. 1955. Evolution of climatic adaptation in homeotherms. Evolution 9:15-26. SNYDER, L. L. 1957- Arctic birds of Canada. University of Toronto Press, Toronto. 310 pp. . STEEL, R. G. D., and J. H. TORRIE. i960. Principles and procedures of statistics: with special reference to the biological sciences. McGraw-Hill, New York. 48l pp. f STONEBOUSE, B. 1963. Egg dimensions of some Ascension Island sea-birds. Ibis 103b:47^-479. ' TODD, W. E. C. 1963. Birds of the Labrador peninsula and adjacent areas. University of Toronto Press, Toronto. 819 pp. ' TYLER, C. 1969. A study of the egg shells of the Gaviiformes, 9. ! HARTOLDMON0029825  l&k Procellariiformes, Podicipitiformes and Pelecaniformes. Journal of Zoology (London) 150:395-412. WEL1T, J, C. 1964. The life of birds, W. B. Saunders, Philadelphia. ' 546 pp. HART OLDMON0029826 OOLOGICAL DATA ON EGG AND BREEDING CHARACTERISTICS OF BROWN PELICANS Daniel W. Anderson and Joseph J. Hickey jfces in Made in United States of America Reprinted from The Wilson Bulletin Vol, 82, No. 1, March 1970 pp. 14-28 Copyright1968 by the American Association for the Advancement of Science HART OLDMONOQ29827 OOLOGICAL DATA ON EGG AND BREEDING CHARACTERISTICS OF BROWN PELICANS Daniel W. Anderson and Joseph J, Hickey he Gulf Coast population of the Brown Pelican (Pckcanus occidentalis) T is now considered to be endangered by the AOU Committee on Conser vation (1968). The circumstances surrounding its decline are not clear. Murphy (1936:102, 808-822) suggested that some breeding populations of Brown Pelicans "normally" fluctuate in response to fluctuating food supplies in relation to such factors as Humboldt Current changes, as well as other factors. Conney (1967), Kupfer (1967), Peakall (1967), Risebrough et al. (1968), and Wurster (1969) have explained some potential physiological effects of chlorinated-hydrocarbon and related environmental pollutants on mammalian and avian reproduction, which might apply, as well. This paper presents information, obtained from major oological collections in North America, regarding some egg and reproductive parameters of the Brown Pelican. Ratcliffe (1967) and Hickey and Anderson (1968) have utilized oological sources to document changes in shell thickness and shell weight among seven species of birds. These changes were related to (1) the widespread introduction of persisting chlorinated hydrocarbons into the environment and (2) reproductive failures associated with shell-breakage and loss. The lack of field data regarding certain breeding and egg characteristics from prior to and possibly during the decline of the Brown Pelican necessi tated our attempt to glean whatever information possible from museum and private-egg collections. An understanding of the present situation, in addition, requires an evaluation of the geographical and temporal variations in the characters of interest. METHODS Measurements.--Eggs were weighed to the nearest 0,01 gram (g) on a torsion balance. Improper cleaning undoubtedly influences shell-weight and possibly also shell-thickness measurements. We used four criteria to determine if eggs had been properly blown: (1) a tendency to settle to one side when rolled on a smooth surface, (2) loose contents, (3) roughness on the interior of the shell, and (4) visual examination. In the course of measuring over 34,000 eggs of 25 species, we found about 200-300 broken or cracked eggs and a larger number with large holes. These lent themselves to close examination, and all proved to be satisfactorily cleaned. Eggs with holes larger than 7 mm were either not measured, or their weights were corrected to those with a 3-mm hole. This was ac complished by taking a small piece of shell, weighing it, and visually "filling" the hole. Egg lengths and breadths were measured to the nearest 0.01 centimeter with a standard, precision vernier caliper. Egg shapes were determined by comparison with the shapes 14 HARTOLDMONOQ29828 m- Anderson and Hickey BROWN PELICAN EGG AND BREEDING DATA 15 Table 1 Clutch Sizes and Incubation Stages of Eggs Taken by Oologists Prior to 1943 Est. Stage of Incubation Incubation Rating Sample Size First egg-3 days1 4-12 days 13-21 days 22-30 days All combined i 72 2 137 3 27 40 -- 236 1 This stage represents a period of approximately 9 days. Mean Clutch Size 2.94 2.93 3.07 -r 2.95 n 0.32 0.31 0.50 --' 0.27 described by Palmer (1962:13) and Preston (1968). Shell thickness was measured to the nearest 0.01 mm with a specially adapted micrometer, the procedure being described by Hickey and Anderson (1968). Tliickness included shell and associated membranes at the girth of each egg. Injormation from Data Slips,--Data slips, giving species, date of collection, stage of incubation, location, collector, and other pertinent information accosnpanied each set of eggs we measured. Due to the inadequacy of incubation terminology and the inability to identify incubation stage accurately (Storer, 1930), mean dates of set-collection (corrected on the basis of reported incubation to give date of clutch completion) can only provide an estimate of breeding phenology. The dates together for an area really only represent a mean over the years, but do suggest general trends and provide an -index to length and variability of breeding season from region to region. We felt that oologists' .estimates of incubation could, at best, only be categorized to the nearest one- fourth of the period from first egg to the end of incubation. The incubation period of the Brown Pelican is not precisely known (Palmer, 1962:277). We have used Mason's (1945) estimate of about 30 days for our calculations here and have estimated the mean number of days that our samples were incubated on the basis of our four incubation categories (Table 1, col. 1). In our series of samples, mean stage of incubation in days subtracted from mean date of set-collection provided an estimate of date of clutch com pletion. Unincubated ("fresh") eggs were included in the analysis of clutch size, after testing to determine if fresh sets might be biased by the collection of incomplete clutches. When sets of fresh eggs were separately compared with those of later incubation (t-test), no significant differences in clutch size were found (P > 0.05, Table 1). There remains the possibility that some egg collectors sought larger clutches. . Calculations and Indices.--All data were analyzed with an IBM 1620 computer. Sta tistical analyses followed Steel and Torrie (1960). A size index for eggs was calculated by multiplying length by breadth and was used as a crude index to volume. In a study of White Pelicans (P. erythrorhynchos) (D. W. Anderson and J. J. Hickey, unpublished), we have found displaced volume to be correlated with this size index (P < 0.001). , Geographical variations in egg size, shell thickness, shell weight, clutch size, and egg dates were determined in a stepwise manner as follows: (1) current subspecific range boundaries were determined from the AOU Check-list (1957) and Palmer (1962:275), and the range was then subdivided into small geographic units such as a single state; (2) the eggshell data for these were then tested for significant differences and regrouped until a region was obtained containing a maximum number of subunits that were not significantly different from each other; (3) groupings never included more than one ^ol * a >es in Vi Copyright! 968 by the American Association for the Advancement of Science HARTOLDMONI 16 THE WILSON BULLETIN March 1970 Vol. 82, No. 1 Table 2 Geographical Variation in Eccsheias of North American Brown Pelicans, 1879 to 19431 Subspecies Area occ West Indies car car Fla,, S.C. Ga. car car car La. Panama Texas cal Baja Calif. cal So. Calif. Number Wt. (g) 95% C.L. 6 8.05 0.90 43 9.46 0.35 208 9.78 0.12 42 9.87 0.32 7 9.94 0.49 115 10.00 0.26 174 10.99 0.18 85 10.59 0.24 Size Index (cm2) 33.2 95% C.L. 0.6 37.6 0.9 37.6 0.3 38.2 0.7 37.4 1.0 38.5 0.6 40.0 0.4 39.0 0.7 No. Subelliptical No. Oval Thickness Index2 95% C.L. Number Thickness (mm) 95% C.L. 3 21 109 20 2 3 22 99 22 5 2.42 2.52 2.60 2.58 2.66 0.24 0.06 0.02 0.06 0.10 6 0.510 23 0.557 172 0.557 24 0.554 ---- 0.031 0.021 :0,004 0.014 52 63 2.59 0.04 94 80 2.74 0.02 44 41 2.71 0.04 43 83 28 0.557 0.569 0.579 0.012 0.008 0.014 i The pre-1943 means that were not significantly different at the 95% level in Duncan's New Multiple Range Test (Steel and Torrie, 1960:107-109, 114) are underscored. 2From Ratciiffe (1967); Thickness index = 10 X wt, in gj (length X breadth in cm). described subspecies; and (4) plienological subdivisions were kept at the smaller units without regrouping. BESULTS ANB DISCUSSION Geographical Variation in Egg Parameters.--Egg-size index, shell weight, and shell thickness (Table 2) tended to vary with the size of the bird as discussed by Romanoff and Romanoff (1949:150). Our index to body size was obtained by using two common standard measurements that tend to measure skeletal size (tarsus and culmen) (Fig. 1). These skeletal measure ments were taken from Wetmore (1945) and represent those of female birds, Wetmore (1945) ranked the size of the three North American subspecies, from largest to smallest as follows: P. o. californicus, P. o. carolinensis, and P. o. occidentalis. The general shape categories (Table 2) were, nonetheless, not significantly different (P > 0.05, Chi-square test) from area to area or' between sub species. Ordinary shape changes in the eggs of domestic poultry have already been shown to have little effect on the shell present as a percentage of total egg weight (Asmundson and Baker, 1940). Of the subspecies carolinensis, birds from Texas tended to have the largest HARTOLDMON0029830 11<\ * 1 *?&,, BROWN PELICAN EGG AND BREEDING DATA 17 I ^Ol v> cn c<ac CO Fig. 1. Relationship between two egg measurements and index to body size in three subspecies of Brown Pelicans. The index to body size was calculated in mm2 units and is shown on the abscissa. Eggshell size was taken as the product of length and breadth in cm,2. eggs. Louisiana eggs tended to be intermediate between those from Texas and those from areas to the east (Table 2). South Carolina birds tended to have smaller and lighter-shelled eggs than birds from farther south in Florida and Georgia, although not significantly so (Table 2). The Baja California eggs (P. o. californicus) were represented mostly by specimens from Los Coronados Island but suggested a similar gradient, with egg size decreasing from southern to northern colonies. Lack (1968:279) mentioned this trend among certain congenerics in certain tropical Procellariiformes. A con tinuum in egg size and shell weight between different populations from different areas was suggested in our specimens, especially in carolinensis, although shell thickness in the various subspecies seemed relatively stable. Whether or not the intrasubspecific tendencies are genetic is unknown. They are likely genetic, but standard measurements from museum skins are needed for further comparisons. The intersubspecific variations in egg size are most likely representative of body size (Fig. 1). If one assumes that egg size provides an index to body size, the large Texas birds may represent an intermediate between californicus and carolinensis. Brown Pelicans along the Pacific Coast (californicus) have the larger and thicker-shelled eggs (Table 2). Asmundson et al. (1943) showed that larger eggs in several species tended to have the thicker shells, but the essentially ies in ' i Copyright1968 by the American Association for the Advancement of Science HARTOLDMON0029831 18 THE WILSON BULLETIN March J970 Vo!. 82, No. 1 equal thicknesses from all our Gulf and Atlantic Coast eggs suggested that this relationship was not present on an intrasubspecific basis. The small sample of eggs from Panama suggested that these eggs were most similar to the subspecies carolinensis, as Wetmore (1945) has shown with museum skins. Unfortunately, we were unable to obtain egg measurements from Ecuadorian or Peruvian Brown Pelicans. Murphy (1936:820) reported that the Peruvian pelicans are very large and we suspect that their eggs would also be larger and thicker-shelled. The ecological significance of egg-size difference within a species is largely a matter of speculation. Lack (1966:7) suggests that egg-size differences between different species (and larger groups) are mainly a matter of heredity. The differences we observed on an intersubspecific basis in Brown Pelicans at least implied that these eggs are represented by relatively distinct gene-pools. Perhaps such gene-pools are even distinct on an intrasubspecific basis. Mason (1945) showed that Florida Brown Pelican movements, at least, are some what restricted under normal circumstances, suggesting potential isolation between breeding groups. Welty (1962:408 quoting Murphy, 1936) also suggests that this species is potentially sensitive to isolating barriers. . Possible Factors for Bias.--It is not our primary objective here to specu late on taxonomic relationships on the basis of eggs; nonetheless, the varia tions in eggs are expected to relate in some ways to taxonomic characters (Tyler, 1964, 1965). Our interest is mainly to examine natural variation in order to better understand if unnatural change has occurred. Egg size and shell thickness and composition are known to vary with heredity, age, adult physiological condition, diet, and chemical influence (Romanoff and Romanoff, 1949:152-157, 359; Preston, 1958; Sturkie, 1965: 464, 487-4-88; Simkiss, 1967:157-197). Shell thickness also varies in dif ferent areas of the egg of a given species, the most notable examples probably being the rock-nesting murres (JJria sp.) and other seabirds, where thickness tends to increase at the most vulnerable parts (Tuck, 1960:25). Some inter specific differences in thickness have been shown to be related to the hazards associated with placement on different nesting substrates (Belopol'skii, 1957: 133-134). Fortunately, egg collectors drilled their specimens at the girths, the most uniform area for most species (Romanoff and Romanoff, 1949: 157-158). Shell calcium (about 5 per cent) is utilized, as well, by developing embryos (Simkiss, 1967:198-213); hence, shell weight and also possibly thickness may be biased low if eggs of late-stage incubation are used in the shell-thick ness or weight comparisons. Data combined into carolinensis and call- forrdcus categories indicated this trend (Table 3), although not significant statistically (t-test, P > 0.05) and only amounting to a small percentage HARTOLDMON0029832 Anderson and Hickey BROWN PELICAN ECG AND BREEDING DATA 19 Table 3 Shell Weights of Pre-1943 Eccs of Two Subspecies of Brown Pelican at Different Incubation Stages Subspecies Incubation Stage carolinensis First egg-3 days 4-12 days 13-21 days No. 98 230 53 Mean Wt <g) 9.75 9.88 9.76 0.21 0.14 0.28 calijornicus First egg-3 days 4-12 days 13-21 days . 92 10.75 121 10.97 30 10.59 0.20 0.25 0.40 Both First egg-3 days 4-12 days 13-21 days 190 10.23 351 10.26 83 10.06 0.16 0.14 0.24 in our sample (1-3 per cent). Therefore, we do not believe this bias to be important in the oological data examined here. Furthermore, the data sug gested that most egg collectors tended to collect eggs that were about onethird or less incubated (Table 1), thus eggs in late-stage incubation repre sented a small percentage of our sample. Although effects on the egg stemming from the age and physiology of the laying female would remain undetectable in oological samples, they would not be expected to affect an overall random, or essentially random, sample (see Asmundson et ah, 1943). Eggshell Changes and Pesticide Residues.--The small samples of post-1949 specimens suggested thinning in all eggshells measured (Table 4). Florida specimens showed a -17 per cent change in shell weight, Texas specimens a -20 per cent change, California specimens (Anacapa Is.) a -26 per cent change, and one set of eggs from Panama a -15 per cent change. All were significant (P < 0.05) changes. We could detect no change in shape in these post-1949 eggs (P > 0.05, Chi-square test). The incubation stages were essentially the same for both pre-1943 and post-1949 eggs (6 2 days vs. 9 5 days, 95 per cent C.L.). Size indices were not significantly different (P > 0.05), although the post-1949 eggs from Texas and Florida were slightly smaller in mean than those of pre-1943. Whether or not these changes in weight and thickness were associated with either recent declines of the Brown Pelican or environmental pollution, or both, remains to be determined. Stickel (1968) has stated that in Gulf Coast Brown Pelicans, pesticide residues were of approximately the same general magnitude as those of herons i Copyright1968 by the American Association for the Advancement of Science 20 THE WILSON BULLETIN March 1970 Vol. 82, No, 1 Table 4 Post-1949 Eggshell Measukements of Buown Pelicans1 Subspecies Area carolinensts Florida carolinemis Texas carolinensis Panama californicus California Number Wt. (g) 95% CL. Size Index (cm2) 95% C.L, No. Subclliptical No. Oval Thickness Index2 95% C.L. 9 8.10 0.14 36.5 0.9 1 8 2.22 0.09 6 7.96 0.60 87.6 2.4 2 4 2.12 0.10 3 8.45 0.99 37.6 2.0 2 1 2.25 0.22 9 7.89 0.66 39.0 1.4 7 2 2.02 0.12 Number 3___ -- 9 Thickness (mm) -- -- 0.457 0.424 95% C.L. -- -- 0.012 0.018 1 Post-MO eggs were collected as follows; Florida--1950, 1953; Texas--1951; Panama--1952; California--1962. 2 From Ratcliffe (1967):: Thickness index: ~ 10 X wt in g/( length X breadth in cm). (Ardea cinerea) from Great Britain and Bald Eagles (Haliaeetus leucocephalus) in the United States (see Stickel et al.. 1966; and Moore and Walker, 1964). Risebrough et al. (1967) analyzing two Brown Pelican eggs from the Gulf of California found them to be generally "low" in pesticide content (0.7 ppm [wet-weight basis] DDT and metabolites and about one-fifth as much polychlorinated biphenyls [PCB's], an industrial pollutant; endrin and dieldrin, were also identified). They found an average of 0.8 ppm DDTfamity residues (61 per cent DDE) and about two-thirds as much PCB in six Brown Pelican eggs taken in Panama. We converted the above residues to a ppm wet-weight basis by assuming 7 per cent fat in the eggs. We measured two of the eggshells from Risebrough's study (Baja California specimens) and found one suggestive of a "normal" egg (11.7 g, 0.59 mm in thickness) and the other suggestive of thinning (9.3 g, 0.50 mm). Another study (Anderson et al., 1969) showed that egg residues as low as 1 ppm of DDE, and possibly less, could be associated (P < 0.05) with detectable shell changes in White Pelicans, although egg residues may not always necessarily reflect residues in adults that could influence egg-shell deposition. Risebrough et al. (1967) reported 84.4 ppm of DDT-type residues, 91 per cent of which was p,p'-DDE (77 ppm) in the breast muscle of a Brown Pelican collected in California. These levels are only slightly lower than those reported from Lake Michigan Herring Gulls (Larus argentdlus), which averaged 80 ppm DDE in the breast of adult birds (Hickey et al., 1966). Reproduction in the HARTOLDMONOQ29834 Anderson and Hickey BROWN PELICAN EGG AND BREEDING DATA 21 Table 5 Mean Dates op Clutch Completion in Brown Pelicans Fiiok Various Geographical Areas Area No. Clutches So, California No. Baja California Texas Louisiana Florida South Carolina 29 61 36 14 75 14 1 Numerically coded with Table 1, cols. 1-2. Mean Date ~ s.n. 8 April 16 days 10 April 54 days 9 May 18 days 27 April 31 days 29 May 125 days 5 June 17 days Mean Stage Incubation3 1.4 1.8 1.5 1.6 1.9 1.7 Wisconsin Herring Gull population in Green Bay (characterized by eggbreakage) is known to be severely affected by DDE and other residues (Keith, 1966; and Hickey and Anderson, 1968). Egg residues from the same population averaged 183 ppm DDE in 1963 and 1964 (Keith, 1966). Breeding Characteristics.--Pacific Coast data suggested that between north ern Baja California and California, the breeding dates were somewhat closely related (Table 5). Gulf and Atlantic Coast birds, on the other hand, showed much variation, especially in Florida (Appendix 1) as discussed by Bent (1922:295) and Palmer (1962:277). Palmer's (1962:275) distribution map suggests that on the Pacific Coast, the major breeding populations of californicus are concentrated into a smaller area than those from Gulf and Atlantic Coast sites (carolinensis). Bent (1922:296), Howell (1932:85-87), and Lowery (1960:113-114) noted that Brown Pelicans of the subspecies carolinensis tended to utilize trees as well as coastal beaches and islands as nesting substrates. Murphy (1936:810-814) mentioned diverse breeding sites for South American pelicans as well. The Brown Pelicans of northern Baja California and California seem more generally restricted to ground nesting on islands (Bent 1922:301; Williams, 1927). Bond (1942) reported tree-nesting for the California Brown Pelican as very unusual. In Florida, where the Brown Pelican still persists (Williams and Martin, 1969), a long breeding season and diversity of nesting substrate seem to char acterize breeding. They nest year-round in Peru, although considerable shift ing of sites occurs (Murphy, 1936:821-822). The Gulf of California Brown Pelicans still persist as breeders, although there is no evidence of a longer breeding season than in colonies farther north (R. W. Risebrough, pers. comm.). Clutch sizes showed no significant variation (P > 0.05) between any of the geographical areas listed in Table 2. The means, and our best estimates for clutch-size in the Brown Pelican, are given in Table 1. Bent (1922:297) Copyright1968 by the American Association for the Advancement ofScien im. 22 THE WILSON BULLETIN March 1970 Vol. 82, No. I and Palmer (1962:277) stated that three eggs, and less often two, is the normal clutch size; nests with four and five eggs have been found. Breeding Records.--The population estimates by egg collectors cited in Appendix 1 must be viewed cautiously. These estimates were subject to observer error; however, they can provide an approximation of changes that might have occurred. Data-slip information, although most likely sketchy, can also provide documentation of past breeding locations. The records we found in egg collections did not provide a complete picture of breeding lo calities but suggested possible fluctuations in numbers over the years (Appen dix 1). On the other hand, none of the major colonies seem to have been completely without birds since at least the late 1800!s. Numbers probably increased on Anaeapa Island, California, during the late 1920's. Williams (1927) reported a colony as far north as Point Lobos, California, during this time. The late 1920's may represent a period of population increase. Bond (1942) reported the estimated numbers on Anaeapa Island from 1898 to 1941 to be highly fluctuating (estimates ran from about 200 to at least 2000 pairs). Banks (1966) reported eggs and young on Anaeapa and essentially "normal" numbers of breeding birds, at least in 1963 and 1964, two years after the thin-shelled eggs reported here. The Los Coronados birds seem historically more stable (Appendix 1). It is certain that both Anaeapa and Los Coronados breeders were historically present in large numbers (Banks, 1966). Eisebrough (1968) and Schreiber and DeLong (1969) suggested that the Brown Pelican has decreased considerably in recent years off Cali fornia, including no known breeders on Los Coronados in 1968. Perhaps the -20 to -26 per cent figure in shell change represents or approaches the lower limit to which eggs may survive to be collected by egg-collectors. Certainly, some production occurred in the California colony with these shellchanges, although present numbers suggest a declining population. Lowery (1960:113-114) mentions large colonies in Louisiana; yet Winckler (1968), in a popular article, summarized their nearly virtual disappearance from the Gulf Coast by 1968. In the light of the better-known demise of Gulf Coast Brown Pelicans, we believe the status of California Brown Pelicans and populations farther to the south needs immediate study. SUMMARY Mean clutch size in 236 sets of North American Brown Pelican eggs was 2.95 and did not vary geographically between North American populations. Shell weight varied from 8.05 g to 10.99 g along a geographic continuum. Shell thickness averaged 0.510 mm for Pelecanus occidentalis occidentdis, 0.554 0.557 mm for P. o. carolinensis, and 0.569 0.579 mm for P. o. ccdijomicus. The ranges of breeding dates for the more southern populations were wider than those of northern ones. Small numbers of eggs taken in Texas and Florida after 1949 were 20 per cent below normal weight; 1962 eggs from California were 26 per cent below normal; and three HARTOLDMONOQ29836 $:tmZy BROWN PELICAN EGG AND BREEDING DATA 23 taken in Panama, 15 per cent below normal. Shell thickness had likewise decreased 15-27 per cent. ACKNOWLEDGMENTS This study was carried out hs part of a contract with the Bureau, of Sport Fisheries and Wildlife, Fish and Wildlife Service, U.S. Department of tire Interior, Patuxent Wildlife Research Center, Laurel, Maryland. We are grateful to the many curators of museum collections listed in Appendix 1. The private collectors cited in Appendix 1 were extremely cooperative. We are especially grateful to Ed N. Harrison and Wilson C. Hanna for their personal assistance and extremely helpful suggestions. Mr, Harrison, in addition, located our 1962 samples of California eggs. Lucille F. Stickel and Eugene H. Dustman provided critical advice, and Ralph W. Schreiber suggested the immediate consolidation of our Brown Pelican data. Mrs. Pearl Davis punched our data cards, and the College of Agricultural and Life Sciences, University of Wisconsin, provided computer facilities at no cost. We are grateful to J, 0. Keith and R. W. Risebrough for critical advice on the manuscript. APPENDIX I Brown Pelican Breeding Records taken prom North American Oological Records and Collections. Date Location Southern California 27 May 1893 5 June 1910 7 Mar. 1916 2 Mar. 1917 15 May 1919 7 Mar. 1920 8 Mar. 1922 28 Mar. 1927 24 Feb. 1929 1 Mar. 1936 1 Mar. 1936 12 Mar. 1939 19 May 1919 25 May 1927 Anacapa Is, Anacapa Is. Anacapa Is. Anacapa Is. Anacapa Is. Anacapa Is. Anacapa Is. Anacapa Is. Anacapa Is. Anacapa Is. Anacapa Is, Anacapa Is, San Miguel Is. Point Lobos Baja California, Mexico 18 Apr. 1894 19 Apr. 1894 4 Apr. 1895 19 Apr. 1898 27 Apr. 1898 6 May 1904 6 Apr. 1908 Los Coronados Los Coronados Los Coronados Los Coronados Los Coronados Los Coronados Los Coronados Estimated Numbers; Remarks Observer Museum* of record 500+ pairs -- -- -- 5,000+ pairs -- -- -- -- 2,000+ pairs "large colony" -- 8-10 nests A. H. Miller G. Willett M. C, Badger M. C. Badger -- S, B. Peyton S. B. Peyton -- C, W. Ashworth E. Harrison L. T. Stevens L. T. Stevens -- L. Williams (1927) 2 3,5 2 3 1 30 .5 3 2 3 14 4,7 1. 2 -- -- -- -- -- 500 nests E. Parker -- A. Hewitt A. J. Kellog -- O. C. Polling P. I. Osborne 27 1 2,22 24 3 2 1,4 ^ol. ges in i: 1i, I Copyright1968 by the American Association, for the Advancement of Science !S HARTOLDMON0029837 24 THE WILSON BULLETIN March 1970 Vol. 82, No. 1 Date 6 Apr. 1908 1 July 1908 4 Apr. 1910 2 Apr. 1912 1 Apr. 1913 29 Mar. 1914 31 May 1915 26 Mar. 191? 4 May 1917 11 Apr. 1919 12 May 1921 30 Mar. 1922 15 Apr. 1881 26 Mar. 1917 6 Apr. 1920 6 Apr. 1920 17 Apr. 1921 2 May 1921 7 Apr. 1932 10 Apr. 1932 2 June 1932 Panama 15 Feb. 1942 15 Mar. 1952 Texas 10 May 1886 20 May 1888 10 Apr. 1889 16 Apr. 1889 14 June 1894 14 May 1896 28 May 1910 30 May 1910 3 May 1912 18 May 1913 27 May 1915 19 May 1917 (appendix 1 continued) Location Estimated Numbers; Remarks Los Coronados Los Coronados Los Coronados Los Coronados Los Coronados Los Coronados Los Coronados Los Coronados Los Coronados Los Coronados Los Coronados Los Coronados Mexican coast: So. Coronados, SE slope Todos Santos Is. Todos Santos Is. San Pedro Nolasco Is. Granite Is, San Benito Is. San Martin Is. Asuncion Is. ___ -- -- -- 500 nests -- -- 500 pairs -- -- -- -- -- -- -- -- -- -- -- -- -- Observer Museum* of record A. Van Rossem P. I. Osborne P. I. Osborne C. S. Thompson L. M. Huey W. C. Bradbury I. D. Nokes N. K. Carpenter D. S. DeGroot N. K. Carpenter W. C. Hanna -- G. Bancroft J. Burnham -- E. Harrison E. Harrison E. Harrison 25 9 29 23 3 2,9 5 6 2 23 4 1 1 1 4 26 1 1 3 3 3 Chaina Is., Panama Bay, Panama Taboga Is., Panama Near Corpus Christi Neuces Co. So. Bird Is., Laguna Madre . So. Bird Is., Laguna Madre 25 mi. from Corpus Christi So. Bird Is., Laguna Madre Near Corpus Christi Near Corpus Christi Laguna Madre Neuees Co. Padre Is. Big Bird Is., Laguna Madre A. Wetmore 13 -- A. Wetmore 13 ' F. B. Armstrong 12 -- T. S. Gillin .4 J. A. Singiey 2 ., J. A. Singiey 4,25 . F. B. Armstrong 1 D. B. Burrows 2 __ C. E. Farley 30 -- J. M. Carroll 4 -- J. M. Priour 4 -- F. B. Armstrong 9 -- F. B. Armstrong 2 _ R.W. Quillan 19 HARTOLDMONOQ29838 amfHickey BROWN PELICAN EGG AND BREEDING DATA 25 Date May 1919 15 May 1919 5 May 1922 . 24 May 1925 1951 Louisiana 29 Mar. 1893 28 Mar. 1894 29 Mar. 1894 3 June 1919 5 June 1919 26 May 1938 13 Apr. 1940 (appendix 1 CONTINUED) Location Is. off so. coast Laguna Madre Neuces Co. Pelican Is., Aransas Bay Refugio Co. Estimated Numbers; Remarks _ -- -- -- -- Observer _ H. Brandt G, Stewart R. D. Camp T. C. Meitzen Museum* of record 3 10 11 31 18 Lost Is. Marsh Is. Shell Keys Pass a l'Outre Errol Is. North Is. La Fourche Par., Timbalier -- F. A. Mcllhenny 2 -- F. A. Mcllhenny 2 -- F. A. Mcllhenny 2,23 -- E. R. Kalmbach 13 -- J. D. Figgins 9 -- F. Tobin 10 G. H, Lowery -- (1960) 17 Florida 15 Mar. 1879 1 Apr. 1880 15 Apr. 1880 29 Apr. 1880 12 Apr. 1890 12 Apr. 1890 3 May 1890 26 Apr. 1891 12 Apr, 1892 10 May 1893 5 June 1893 30 June 1894 21 Jan. 1896 3 Apr. 1896 30 Apr. 1896 18 May 1896 15 May 1899 19 Apr. 1908 20 Apr. 1908 3 May 1911 19 May 1911 27 Apr. 1913 27 Apr. 1913 Near Marco Indian R. Indian R. Old Tampa Bay Lee Co. Charlotte Harbor W. of Pine Is., Lee Co Pelican Is. Tampa Bay Pelican Is. Mullett Key Tampa Bay Pelican Is. Pelican Is. Seminole Is. Rookery Is., off Diston City Brevard Co. Boca Grande, Charlotte Keys Charlotte Harbor, Devilfish Key Pelican Is. Hillsborough Co. Lee Co. Roco Bay, Pinellas Co. -- -- .-- -- -- -- 225 nests -- -- ---- ---500 pairs -- -- _ C. L. Gass -- -- H. R. Jamison S. Reiff N. K. Jamison M. Gibbs (1894) D. P. Ingraham J. M. Southwick B. T. Smith .--B. W. Evermann H. E. Pendry H. E. Pendry 1 26 1 1 4 21 4 9 27 4 26 1 23 3 5 W. Meyor 8 -- F. S. Webster 10 200 birds P. B. Phillipp 12 ---- -- large colony in trees P. B. Phillipp P. B. Phillipp O. E. Baynard O. E. Baynard O. E. Baynard 12 12 24 3,9 8 '3-oL m NCE Copyright1968 by the American Association for the Advancement of Science HARTOLDMON0029839 26 THE WrLSON BULLETIN March 1970 Vol. 82, No. 1 (appendix 1 continued) Date 15 May 1918 20 Apr. 1920 17 May 1921 17 May 1921 27 May 1921 28 Dec. 1921 20 Apr. 1926 1 June 1926 10 June 1929 28 Mar. 1930 25 Apr. 1930 10 Apr. 1931 6 June 1931 7 June 1931 3 May 1932 22 Apr. 1934 9 Mar. 1950 10 Mar. 1953 Location Tampa Bay Tampa Bay Tampa Bay Tampa Bay Tampa Bay Pelican Is. Pinellas Co. Merritt Is, Merritt Is. Lee Co. Near Bokelia? Mosquito Lagoon, Brevard Co. Pine Is. Res., Bird Key Matlacha Pass Res., 6-mi. Is. Bird Key, Hillsborough Co. Rattlesnake Key, Levy Co. Is., n. side of Cocoa-- Cocoa Beach Merritt Is. Estimated Numbers; Remarks -- -- -- -- -- -- -- -- 2,500 pairs -- -- 2,000 nests -- -- -- 375 nests -- Observer Museum* of record J. L. Vaughn -- J; L. Vaughn W. F. Lewis J, L. Vaughn T: D. Burleigh C. E. Doe K. Squires J, C. Howell, Jr. C. E. Doe G E. Doe 4 3 2,20 8 23 10 16 2 12 16 16 W. H. Nicholson R. W. Williams 23 13 R. W. Williams R. E. Gammell 13 7 C. E. Doe 16 C. E. Carter H. Brandt 15 10 Georgia 16 June 1898 Chatham Co. on beach T. D. Perry 1,16 South Carolina 10 May 1901 20 June 1901 23 June 1901 23 May 1915 18 June 1915 7 July 1916 3 June 1925 14 June 1934 20 June 1942 10 June 1943 10 July 1943 Bird Bank, Bull's Bay Near Charleston Bay Point, near Beaufort Bird Bank, Bull's Bay Bird Bank, Bull's Bay Bull's Bay Bull's Bay Georgetown Co. Bull's Bay St, Helens Sound, Beaufort (Bird Bank) 18 mi. e. Beaufort -- on beach "large colony" -- _ -- -- ----- -- -- -- -- M. T. Cleckley -- M. T. Cleckley A. C. Bent A. Sprunt, Jr. M. T. Cleckley W. B. Savary H. L. Harllee E. J. DeCamps E. J. DeCamps E. J. DeCamps 9 3 3 13 30 28 5 14 14 4 14 Cuba 8 Sep. 1930 Cacachita Bay -- P. Bartsch 13 * Museums and collections are numbered as follows: 1. Calif. Acad. Sci., San Francisco; 2. Mus. Vert. Zoo!., Univ. Calif., Berkeley; 3. Western Found. Vert. Zook, Los Angeles, Calif,; 4. San Bernardino Co. Mus., San Bernardino, Calif,; 5. S. B. Peyton, private collection, Fillmore, Calif.; 6. Oakland Publ. Mus,, Oakland, Calif.; 7. Santa Barbara Mus. Nat. Hist, Santa Barbara, ^Ol Anderson and Hickey BROWN PELICAN EGG AND BREEDING DATA 27 Calif.; 8. San Diego Mus. Nat. Hist., San Diego, Calif.; 9. Denver Mus. Nat. Hist., Denver, Colo.; 10. Carnegie Mus., Pittsburgh, Pa.; 11, Philadelphia Acad. Sei., Philadelphia, Pa.; 12. Amer, Mus. Nat. Hist, New York, N.Y.; 13. U.S. Natl. Mus., Wash., D.C.; 14. Zoological Mus., Clemson Univ., Clemson, S.C.; 15. C, E. Carter, private collection, Orlando, Fla.; 18. Fla. State Mus,, Gainesvilie; 17. L.S.XJ. Mus. Nat. Sci., Baton Rouge, La.; 18. T. C. Meitzen, private collection, Refugio, Tex.; 19, R, W, Quillan, private collection, San Antonio, Tex.; 20. Univ, Kans. Mus. Nat. Hist, Lawrence; 21. Univ, Nebr. Zool. Dept Mus., Lincoln; 22, Cleveland Nat. Sci. Mus., Cleveland, Ohio; 23. Royal Ont Mus., Toronto; 24. Joseph Moore Mus., Earlham ColL, Richmond, Ind,; 25. Ohio State Mus,, Ohio State Univ., Columbus; 28. Univ. Mich, Mus. Zool., Ann Arbor; 27. James Ford Bell Mus. Nat. Hist,, Univ. Minn., Mpls.; 28. M. Pollock, private collection, Edmonton, Alta,; 29. Burke Memorial Mus., Univ. Wash., Seattle; 30. Puget Sound Mus. Nat. Hist., Univ. Puget Sound, Tacoma; 31. Zoology Mus,, Ore. State Univ., Corvallis. LITERATURE CITED American Ornithologists' Union. 1957. Check-list of North American birds. Fifth ed. American Ornithologists' Union. 1968. Report of committee on conservation, 1968. Auk, 85:669-677. Anderson, D. W., J. J. Hickey, R. W. Risebrough, D. L. Hughes, and R. E. Christensen. 1969. Significance of chlorinated hydrocarbon residues to breeding pelicans and cormorants. Canadian Field-Naturalist, 83:91-112. Asmundson, V. S., and G. A. Baker. 1940. Percentage shell as a function of shell thick ness, egg volume and egg weight. Poultry Sci., 19:227-232. Asmundson, V. S., G. A. Baker, and J. T. Emlen. 1943. Certain relations between the parts of birds' eggs. Auk, 60:34f-44. Banks, R. C. 1966. Terrestrial vertebrates of Anacapa Island, California. Trans. San Diego Soc. Nat. Hist., 14.; 173-188. Belopol'skh, L. O. 1957. Ecology of sea colony birds of the Barents Sea. Israel Prog, for Scientific Translations, Jerusalem, 1961. Trans, by R. Ettinger and C. Salzmann. Bent, A. C. 1922. Life histories of North American petrels and pelicans and their allies. U.S. Natl. Mus. Bull., 121. Bond, R. M. 1942. Banding records of California Brown Pelicans. Condor, 44:116-121. Conney, A. H, 1967. Pharmacological implications of microsomal enzyme jnductipn. Pharmacol. Rev., 19:317-366. Gibbs, M. 1894.. Nesting habits of the Brown Pelican in Florida. Oologist, 11:81-84. Hickey, J. J., J. A. Keith, and F. B. Coon. 1966. An exploration of pesticides in a Lake Michigan ecosystem. J. Appl. Ecol., 3(Suppl.) :141-154. Hickey, J. J,, and D. W. Anderson. 1968. Chlorinated hydrocarbons and eggshell changes in raptorial and fish-eating birds. Science, 162:271-273. Howell, A. H, 1932. Florida bird life. Coward-MeCann, New York, Keith, J. A. 1966. Reproduction in a population of Herring Gulls (Larus argentatus) contaminated by DDT. J. Appl. Ecol., 3(Suppl.) :57-70. Kupfeb, D. 1967. Effects of some pesticides and related compounds on steroid function and metabolism. Residue Reviews, 19:11-30, Lack, D. 1966. Population studies of birds. Clarendon Press, Oxford. Lack, D. 1968. Ecological adaptations for breeding in birds. Methuen, London. Lowery, G. H., Jr. 1960. Louisiana birds. Louisiana State Univ. Press, Baton Rouge. Mason, C. R. 1945. Pelican travels. Bird-Banding, 16:134-143. Moore, N. W., and G H. Walker. 1964. Organic chlorine insecticide residues in wild birds. Nature, 201 .-1072-1073. ' , Murphy, R. C. 1936. Oceanic birds of South America: Vol. I--II. Amer, Mus. Nat Hist., New York. ! 1 . : i 1 ' ges in HARTOLDMON0029841 aoo 28 THE WILSON BULLETIN vet"7? Palmer, R. S. [Ed.] 1962, Handbook o{ North American birds: Vol. I. Yale Univ. Press, New Haven. Pea kali,, D. B. 1967. Pesticide-induced enzyme breakdown of steroids in birds. Nature, 216:505-506. Preston, F. W. 1958. Variation of egg size with age of parent. Auk, 75:476-477. Preston, F. W. 1968. The shapes of birds' eggs: mathematical aspects. Auk, 85: 454-463. Ratcuffe, D, A. 1967. Decrease in eggshell weight in certain birds of prey. Nature, 215:208-210. RlSEBROtJCH, R. W. 1968. Pollution, wildlife and science. Canadian Field-Naturalist, 82:241-243. Risebrouch, R. W., D. B. Menzel, D. J. Martin, Jr., and II. S. Olcott. 1967. DDT residues in Pacific sea birds: a persistent insecticide in marine food chains. Nature, 216:589-591. ' Risebrouch, R. W., P. Riecre [= Reiche], D. B. Peakall, S. G. Herman, and M. N. Kiryen. 1968. Polychlorinated biphenyls in the global ecosystem. Nature, 220: 1098-1102. Romanoff, A. L., and A. J. Romanoff. 1949. The avian egg. Wiley, New York. Schreiber, R. W., and R. L. Deionc. 1969. Brown Pelican status in California. Audu bon Field Notes, 23:57-59. Simkiss, K. 1967. Calcium in reproductive physiology. Reirihoid, New York. Steel, R. G. D., and J. H. Torrie. 1960. Principles and procedures of statistics. McGraw-Hill, New York. Stickel, L. F. 1968. Organochlorine pesticides in the environment. 2d Meeting on the Unintended Occurrence of Pesticides in the Environment, 19 Sept. 1967, Tay- mouth Castle, Scotland. Mimeo. Stickel, L. F., N. J. Chura, P. A. Stewart, C. M. Menzie, R. M. Prouty, and W. L. Reichel. 1966. Bald Eagle pesticide relations. Trans. N. Amer. Wildl. and Nat. Resources Conf., 31:190-200. Stoker, T. I. 1930. A critique of oological data. Auk, 47:329-334. Sturkie, P. D. 1965. Avian physiology. Comstock, Ithaca, New York. Tuck, L. M. 1960. The murres. Canadian Wildl. Ser. 1, Canadian Wildl. Serv., Ottawa. Tyler, C. 1964. Wilhelm von Nathusius, 1821-1899, on avian eggsheEs. Berkshire Printing Co., Reading. Tyler, C. 1965. Egg-shell characteristics as a guide to taxonomy. Ibis, 107:131. Wei.ty, J. C. 1962. The life of birds. W. B. Saunders, Philadelphia. Wetmoue, A. 1945. A review of the forms of the Brown Pelican. Auk, 62:577-586. Williams, L. 1927. California Brown Pelicans nesting at Point Lobos, Monterey County, California. Condor, 29:246-249. Williams, L. E., Jr., and L. Martin. 1969. Nesting status of the Brown Pelican in Florida in 1968. Quart. J. Florida Acad. Sci., 31:130-140. Winckxeb, S. 1968. Brown Pelican epitaph. Texas Parks and Wildlife, 26(121:24-28. Wurster, C. F., Jr. 1969. Chlorinated hydrocarbon insecticides and avian reproduc tion: how are they related? In Miller, M. W. and G. G. Berg (Eds.). Chemical fallout: Current research on persistent pesticides. C. C Thomas, Springfield, Illinois. DEPARTMENT OF WILDLIFE ECOLOGY, UNIVERSITY OF WISCONSIN, MADISON 53706. 12 MAY 1969. HARTOLDMONOQ29842 Reprinted from 11 October 1968, Volume 162, pp. 271-273 ^Ol Chlorinated Hydrocarbons and Eggshell Changes in Raptorial and Fish-Eating Birds Joseph J. Hickey and Daniel W. Anderson Copyright1968 by the American Association for the Advancement of Science HARTOLDMON0029843 CKP<* Chlorinated Hydrocarbons and Eggshell Changes in (9). Analyses were conducted on Raptorial and Fish-Eating Birds chromatograph (Barber Coleman, i GC 5000, and Jarrell-Ash, mode 700) with electron-capture dete Abstract. Catastrophic declines of three raptorial species in the United States The glass column (0.6 cm by 1.2 m have been accompanied by decreases in eggshell thickness that began in 1947, packed with 5 percent DC 200 <12 have amounted to 19 percent or more, and were identical to phenomena on Cromport XXX. The column reported in Britain. In 1967, shell thickness in herring gull eggs from five states perature was 210C, and the nit) decreased with increases in chlorinated hydrocarbon residues. flow rate was 75 cmVmin. Each tion of the ground and dried sai New perspectives on the role of chlo declining populations; golden eagles was extracted for 8 hours or more rinated hydrocarbon insecticides in our (Aquila chrysaeetos), red-tailed hawks Soxhlet apparatus with a mixtui environment have come into focus in (Buteo jamaicensis), and great horned ether and petroleum ether (70: recent years. Successive discoveries have owls (Bubo virginianus) were selected Portions of the extracts were fu demonstrated that these compounds are as representative of reasonably station purified by putting them throuj systematically concentrated in the upper ary populations that may be slowly de Florisil column. trophic layers of animal pyramids (/). clining as their habitats are gradually In California, where the pere Raptorial bird populations have simul destroyed by man, but for which wide falcon population is in "a serious taneously suffered severe population spread reproductive failures are cur dition" (10), a change of 18.8 pe crashes in the United States and Western rently unknown. In addition, 57 eggs of in shell weight occurred from \9* Europe (2, 3, 4). These involve repro the herring gull (Lams argentatus) 1952. Ratcliffe (6) found a cones; ductive failures which, at least in Brit were collected from five colonies in ing decrease of 18.9 percent in Br ain, are characterized by changes in 1967. The shells of these were dried at The change in California involved calcium metabolism and by a decrease room temperature for 4 months before crease in shell thickness and ha< in eggshell thickness resulting in the being measured, and residues of the en precedent in the previous 57-yea: parent birds' breaking and eating their tire egg contents were analyzed by the corded history of the peregrine in own eggs (4, 5, 6). Such a derangement Wisconsin Alumni Research Foundation state (Fig. 1). In the .eastern U of calcium metabolism or mobilization for chlorinated hydrocarbons but not States, where the nesting populatic perhaps could result from breakdown of for polychlorinated biphenyls. Analyti peregrines has now been wiped ou steroids by hepatic microsomal enzymes cal procedure followed that outlined by fragmentary data indicate that the: induced by exposure to low dietary, the U.S. Food and Drug Administration change took place (Table 1). Br levels of chlorinated hydrocarbons (7). We have examined the possibility that the eggshell changes reported in Britain (6) have also occurred in the United States and that the raptor population crashes in Europe and North America may have had a common physiological Table 1. Weights of raptor eggshells In museum and private collections. Citations (2 refer to the data for the population trend; S.E., standard error of the mean. Region Period No. Weight. (8) Mean S.E. Change <%> Popu tren( repr- tii mechanism. The population changes are without parallel in the recent history of bird populations (8). They include the pending extirpation of the peregrine falcon (Falco peregrinus) in northwest ern Europe, the complete extirpation of the nesting population of this species in the eastern half of the United States, and simultaneous declines among other bird- and fish-eating raptors on both sides of the Atlantic. We examined 1729 blown eggs in. 39 museum and private collections. Shells were weighed to the nearest hundredth of a gram. In 29 percent of these, we were able to insert a micrometer through the hole drilled by the collector at the girth of the shell and to take four measurements of thickness 7 mm from the edge of the blow hole; these were then averaged to the nearest 0.01 mm for each shell. Thickness in each case then represented the shell itself plus the dried egg membranes. Peregrine falcons, bald eagles (Haliaeetus leucocephalus), and ospreys (Pandion haliaetus) were se Calif. (23) Calif. (23) Brevard Co,, Ha, Osceola Co., Fla. Md.-Va. N.J. B.C. . Calif. (23) N.H. to N.J.* Vt, Mass. N.J. Calif. (23) 1885-1937 1943-44 1953-67 1889-1939 1940-46 1947-65 1886-1939 1947-62 1901-44 1959-62 1890-1938 1940-46 1955 1880-1938 1957 1915-37 1947-53 1895-1939 1940-46 1947-52 1888-1932 1946 1947 1950 1886-1936 1948-50 Red-tailed hawk 386 6.32 0.032 6 6.09 0.237 8 6.49 0.214 (Tolden eagle 278 13.03 0.083 28 12.70 0.161 33 13.41 0.232 Bald eagle (24a) 56 12.15 0,127 12 9.96 0.280 25 i2.32 0.240 8 9.88 0.140 Osprey (24b) 152 7.05 0.054 21 6.91 0.164 3 6.85 117 7.08 0.069 6 5.30 0.446 Peregrine (25) , 4.24 0.061 15 4.18 0,081 235 4.20 0.031 49 4.07 0.038 31 3.41 0.084 56 4.38 0.034 3 4.30 3 3.47 3 3.24 Great horned owl 154 . 4.50 0.033 12- 4.62 0.119 ~ 3.6 + 2.7 - 2.5 + 2.9 -18.0 - 19.8 - 2.0 - 2.8 -25.1 - 1.4 - 3.1 - 18.8 - 1.8 -20.8 -26.0 + 2.4 Statii Statii Statii Statii Dec! Decl Stati Stati' Decl Stati No Decl Stati Extii Extii Stati lected as having one or more regionally * Including Vermont and Massachusetts. ^03- eggshells in a North American peregrine eyrie were observed for the first time in 1947 by J. A. Hagar 60 miles (9.6 km) from the Massachusetts eyrie cited in this table (11). They were next inferred In Quebec in 1948 when egg-eating was observed at the same site in 1949 (12), and were observed in Pennsylvania in 1949 and 1950 (13). Chlorinated hydro carbon data for this now-extinct region al population are completely absent. For nine surviving adult peregrines in Cana da's Northwest Territories in 1966, the data are reported to have averaged 369 parts per million (ppm) (fresh weight) in fat (14). For four adults in another migratory population in northern Alas ka, values were even higher (IS). For the five other raptorial species we have studied, the data do not permit a precise delineation of the onset of the change in calcium metabolism or mobi lization, but the decrease (Table 1) in shell weight (and hence thickness) has involved only declining populations and not stationary ones. Change in shell thickness occurs in poultry as a result of dietary deficiencies and age (16, 17). This phenomenon would probably not occur simultaneously on two continents 1 year after the chlorinated hydro carbon insecticides came into general usage. Other chemicals affect shell thickness in poultry (17), but the find ing of high concentrations of chlori nated hydrocarbons in the eggs of wild populations of raptors and the time correlation of shell changes with the introduction of DDT fl,l,l-trichloro-2,2-bis (p-chlorophenyl) ethane] tend strongly to suggest that chlorinated hydrocarbons are the major contributing ^R.l. ME, '** 036) U tX-- g 034-) X r.-0.9863 0.32-1 t- ^* 0.3807-tXOGOSJ X , >vMINN. A MICH. MNS. 20 * 40 60 TOTAI. ppm RESIDUE 80 Fig. 2. Variation in shell thickness and DDE concentrations in the eggs of herring gulls in 1967. The eggs were taken off Block Island, R.I.; Green Island in Penob scot Bay, Me.; Rogers City, Mich., on Lake Huron; near Knife River, Minn., on Lake Superior; and the Sister Islands in Green Bay, Wis. Some polychlorinated biphenyls probably occurred in these eggs, but they have not yet been identified. Fig. 1. Measurements of 614 California peregrine eggshells collected since 1891. The dotted horizontal line is the midpoint between the 95 percent confidence limits for 1947-52 and the lowest of any pre ceding group. Solid horizontal bare are means; rectangles, 95 percent confidence limits; heavy vertical lines, standard devia tions; narrow vertical lines, range in sampie. The thickness index, calculated as ten times the weight divided by the product of the length and breadth (in mm) of each egg, was devised by Ratcliffe (6) for the study of museum eggs and appears here to be a meaningful statistic. The sizes of samples for measurement of weight and of thickness Index for the periods were 71,49, 36, 85, 155, 30, and 31, respectively; and the sizes of the samples for measurement of thickness were 24, 31, 29, 23, 37, 7, and 6, cause, although it is not unlikely that other chemicals could be contributory. In order to test the hypothesis that these recent changes of thickness in raptor eggshells were the result of dif ferences in exposure to chlorinated hydrocarbons we analyzed 10 to 14 eggs taken in 1967 from each of five colonies of the herring gull (Lams argentatus). Mean shell weight and thick ness in 55 eggs collected in the same five states prior to 1947 disclosed no geo graphic gradients or significant differ ences. The 1967 mean thicknesses for each colony were therefore compared to mean levels of residual DDE [1,1dichloro-2,2-bis (p-chlorophenyl) ethyl ene] on a fresh-weight basis, with the result shown in Fig. 2, the r value being significant, with P -- .001. The residues of polychlorinated biphenyls (18) have not been studied in these ecosystems, but DDE has been consistently high in the Lake Michigan birds, averaging (fresh weight) 1925 ppm (S.E. 274) in the fat of 12 healthy adults collected in 1963-64 (19). Reproduction in these gull colonies was generally normal in 1967 except perhaps in Wisconsin. At the latter colony where an 11 percent mean de crease in shell thickness occurred, some egg breakage and shell flaking was evi dent in 1967, although not at the fre quency seen in previous years. Excessive reproductive failure occurred at this site in 1964 when about 18 percent of the eggs lost about one-third of the shell due to flaking, when clutch size decreased and embryonic mortality was high, and when DDE residues averaged 202 ppm (S.E. 34) in nine eggs (20). (If linear extrapolation of the 1967 values is carried out to 202 ppm, the shell thickness in 1964 could be esti mated as having decreased by about 32 percent.) The effectiveness of DDE in the enzymatic metabolism of aminopyrine has been reported by Hart and Fouts (21), and our data suggest that this compound, because of its preva lence, has played a major role in induc ing the hepatic microsomal metabolism of steroids that in turn resulted in the eggshell changes we have encountered in museum collections. Without doubt DDE is the commonest insecticide or insecticide analog now being found in avian tissues (22). In 1966, it was found to average 284 ppm (S.E. 62) in the fat of nine arctic-breeding peregrine fal cons (14) and about 414 ppm in four others on a wet-weight basis (15). Con centrations of this compound and other chlorinated hydrocarbons in the pere grine populations that crashed farther south can be assumed to have been as high--and they may have been much higher. From the above evidence and that accumulated by others (2, 4, 6, 8), we have reached these conclusions: (i) many of the recent and spectacular raptor population crashes in both the United States and Western Europe have had a common physiological basis; (ii) eggshell breakage has been widespread but largely overlooked in North Amer ica; (iii) significant decreases in shell thickness and weight are characteristic of the unprecedented reproductive fail ures of raptor populations in certain HARTOLDMON0029845 parts of the United States; (iv) the on set of the calcium change 1 year after the introduction of chlorinated hydro carbons into general usage was not a random circumstance; and (v) these per sisting compounds are having a serious insidious effect on certain species of birds at the tops of contaminated ecosystems. Joseph J. Hickey Daniel W. Anderson Department of Wildlife Ecology, University of Wisconsin, Madison References and Noles 1. E. G. Hunt and A. I. Bischoff, Callj Fish Game 46, 91 (1960); review in R, L. Rudd, Pesticides and the Living Landscape (Univ, of Wisconsin Press, Madison, 1964), pp. 248-264. 2. C. Demandt, Ornithol, Mitt. 7, 5 (1955); P. Linkola, Suomen Luonto 18, 3, 34 (1959); P. Linkola, ibid. 19, 20 (I960); P. Linkola, ibid. 23, 5 (1964); K. Kleinstauber, Falke 10, 80 (1963); C. Kruyfhooft, in Working Con ference on Birds of Prey and Owls (Inter national Council Bird Protection, London, 1964), p. 70; J.-F, Terrasse, ibid., p, 73. 3. D. D. Berger, C. R. Sindelar, Jr., K. E. Gamble, in Peregrine Falcon Populations: Their Biology and Decline, 1. 1. Hickey, Ed, (Univ, of Wisconsin Press, Madison, 1968), p. 165. 4. D. A. Ratclilfe, Bird Study 10, 56 (1963). 5. -----------( Brit. Birds SI, 23 (1958). 6. ---------- , Nature 215, 208 (1967). 7. D. B. Peakall, ibid. 216, 505 (1967); L. G. Hart, R. W. Shulties, J. R. Fouts, Toxicol. Appl. Pharmacol. 5, 371 (1963); A. H. Con- ney, Pharmacol. Bev. 19, 317 (1967); 13. Kupfer, Residue Rev. 19, 11 (1967). ' S. J. J. Hickey, Ed., Peregrine Falcon Popula tions: Their Biology and Decline (Univ, of Wisconsin Press, Madison, 1968)., 9. U.S. Food and Drug Administration, Pesti cide Analytical Manual, vol. 1 [U.S. Dept, of Health, Education, and Welfare, FDA Adm. Publ. (1963, revised 1964 and 1965)]. 10. B. Glading, in Hickey (8), p. 96. 11. J. A. Hagar, In Hickey (8), p. 123. 12. G. H. Hall, Brit. Birds 51, 402 (1958). 13. J. N. Rice, in Hickey (8), p. 155. 14. J. H. Enderson and D. D. Berger, Condor 70, 149 (1968). 15. T. J. Cade, C. M. White, J. R. Haugh, ibid., p. 170. 16. A. L. Romanoff and A. J. Romanoff, The Avion Egg (Wiley, New York, 1949), pp. 154-157. 17. T, G. Taylor and D, A. Stringer, in Avian Physiology, P. D, Sturkie, Ed. (Cornell Univ. Press, Ithaca, N.Y., ed. 2, 1965), p. 486. 18. D. C. Holmes, J. H. Simmons, J. O'G. Tat- ton, Nature 216, 227 (1967). 19. J. J. Hickey, J. A. Keith, F. B. Coon, J. Appl. Ecol. 3 (suppl.), 141 (1966). 20. J. A. Keifh. J. Appl. Ecol 3 (suppi.), 51 (1966). 21. L. G. Hart and J. R. Fouts, Arch. Exp. Pathol. Pharmakol. 249, 486 (1965). 22. E. H. Dustman and L. F. Stickel, Amer. Soc. Agron. Spec. Publ. 8, 109 (1966). 23. J. B. Dixon, B. Glading, W. C. Hanna, E. N. Harrison, S, B. Peyton, personal com munication. . 24a. A. Sprunt, IV, personal communication. 24b. W. A. Stickel, in Hickey (8), p. 337. 25. Except for the California data (23) the data on population trends are given in (8) by F. L. Beebe for British Columbia; by W. R. Spofford for Vermont; by J. A. Hagar for Massachusetts; and by D. D. Berger et al, for New Jersey. 26. Research carried out under contract with the Bureau of Sport Fisheries and Wildlife, Fish and Wildlife Service, U.S. Dept.- of Interior. W. II. Drury, J. T. Emlen, and M. E. Slate provided gull eggs for analysis. E. N. Har rison, W. C. Hanna, and many other oologists greatly facilitated our measurements of egg shells. We thank D. A. Ratcllffe for advice throughout the entire study. HARTOLDMON0029846 TITLE OP THESIS HYDROCARBONS* ON HERRING GULLS AND OTHER SPECIES . Full Name __J5f5il.FJLliiw_Andersp_n Place and Date of Birth__ _.H?derwppd^_J|orthJDakotarrpXebrmr^ 1939............................... Elementary and Secondary Education__________ . Underwood_ School^ Underwood^J|qrth_Dakta Colleges and Universities; Years attended and degrees 1957-61 North Dakota State University, Fargo I96I4.-67 University of Wisconsin, Madison -- 1567-10.. University of Wisconsin. Madison. B.S. M.S. , ______ fm*.________ Membership in Learned or Honorary Societies _ _Phi _Kg.ppa_ Phi^._Sigma_ Xi,__AjUB.ric &0 Ornithologists' Union, Wildlife Society, American Society of Mammalogists, American Assoc, for the Advancement of Science. Publications-----See; Attached-List,-------------------------,---------- ------------ ------------- -----........... Major Department __________________________________ _ Minor (s)__________ ___________ ___*.;-______________ ________ _______ Date__ 13_Augusi._i2Z9______ . Signed ___ Professor in charge of thesis HARTOLDMONOQ29847 ' ! f1 ; | I | | 1 LIST OF PUBLICATIONS 13 August 70 196U Anonymous. The story of F0rt Bliss. Publication of the Fort Bliss, Texas 03 . 061*31*9. 1966 Anderson, D. W. Spring mortality in insectivorous birds. The Loon 37(U)*13U-135. 1966 --------------------------, and J. W, Ellis. Cormorant nesting in northwestern Minnesota. The Loon 38(1) :5-8. 1967 ---- --------------- . An exploration of pesticides in cormorant and pelican populations. Master's Thesis. University of Wisconsin. 39 p. . 1967 ------------------------ -, and Frances Hamerstrom. The recent status of Wisconsin cormorants. Passenger Pigeon 29(l):3-l5. 1967 --------------------------, and J. G, Bartonek. Additional observations on the status of North American White Pelicans. Condor 69(3)i311-313. 1968 Hickey, J. J., and D. W. Anderson. Chlorinated hydrocarbons and eggshell changes in raptorial and fish-eating birds. Science 162(3850):271-273. 1968 -------------------------------.-------- --- -------- . The peregrine falcons life history and population literature, p. 3-1*2, In: Peregrin Falcon Populations, their Biology and Dec line. J. J, Hickey (Ed.). University of Wisconsin Press, Madison. 1968 Evrard, J. 0., and D, W Anderson. Notes on a Pigeon Hawk/ Herring Gull encounter. Passenger Pigeon 30(3): 113-111*. 1969 Anderson, D. W,, J. J. Hickey, R, W, Risebrough, D. L, Hughes, and R. E. Christensen. Significance of chlorinated hydrocarbon residues to breeding pelicans and cormorants. Canadian FieldNaturalist 83(2):91-112, 1970 ------------------------ , and J. J, Hickey. Oological data on egg and breeding characteristics of Brown Pelicans, Wilson Bulletin 82(l) tll*-28. I97O ------ ------- -------Pseudo-sleeping in Lesser Scaup and ling-necked Ducks. Accepted by Condor, HARTOLDMON0029848 1970 2 Nellis, C. H,,, J. J. Zohrer, and D, W. Anderson. Mallard-- Green-winged Teal association in southern Wisconsin. Accepted by Wilson Bulletin. 1970 Rlsebrough, R. W., J, H. Davis, and D, W. Anderson. Effects of various chlorinated hydrocarbons. Oregon State University Symposium, in press. 1970 Anderson, D, W., and J. J, Hickey, Seasonal variation of chlorinated hydrocarbons in Lake Michigan Herring Gulls. Submitted to Journal of Applied Ecology. I97O ------------------------------------------------- ----, Eggshell changes in certain North American birds. Proceedings of the XT' Congressua Internationalis Ornithologicus, The Hague, Netherlands, 30 August in September 1970. I97O ------- ----------------, and J. P, Kreit^er. Thickness of 1967-69 Whooping Crane eggshells compared to that of pre-1910 specimens. Submitted to Auk. I97O --------------------------, H, G. Lumsden, and J, J. Hickey, Geographical variation in the eggshells of Commotm Loons, Accepted by Canadian FieId-Naturalist. I97O -------------------------, J, J, Hickey, and J. A. Keith. Chlorinated . hydrocarbons and eggshell variation in Herring Gulls. Submitted to Auk. 1970 Berger, D. D., D. W. Anderson, and J, A, Keith. Shell thinning in eggs of Ungava Peregrines. Accepted by Canadian PieIdNaturalist. If HARTOLDMON0029849