Document wqdpQZqEYE3ORgyJzwoyrqJGB

SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS A. S. Cooke The Nature Conservancy, Monks Wood Experimental Station, Abbots Ripton, Huntingdon, Great Britain ABSTRACT Literature has been reviewed concerning shell thinning in avian eggs by environmental pollutants. Field evidence indicates that the declines in shell thickness observed in certain species in North America and Great Britain since the Second World War have been largely caused by residues ofpp'-DDE or other compounds or metabolites of the DDT group. In North America, polychlorinated biphenyls and cyclodiene insecticides have played no more Ilian minor roles, although in Britain cyclodienes have probably made a significant contribution. Mercury compounds are not apparently associated with the shell thickness declines. Results from controlled experiments, in which laying birds have been exposed to pollutants, generally support these suggestions. Laboratory investigations indicate an interspecific difference in shell thinning response, gallinaceous species tending to be the most resistant andfalcons the most susceptible. In a laying bird, organochlorine residues affect many biochemical mechanisms known to be essential for proper shell formation and the extent of the contribution of each affected mechanism towards decreasing shell thickness probably depends on variables such as species and environmental conditions. There is no evidence to suggest that one mechanism is always dominant irrespective of the conditions. In North America shell thinning has often been associated with population decreases, but in Britain declines in shell thickness are not thought to be responsible for the population decreases observed in certain raptor species. INTRODUCTION Since Ralcliffe first reported thinner shells in raptor eggs in 1967 and suggested that ' organochlorine insecticides might be responsible, there has been considerable ____________ 85 Environ. Polhil. (4) (1973) pp. 85-152-- Applied Science Publishers Ltd, England, 1973--Printed in Great Britain QSW 031287 STLCOPCB4015249 86 A. S. COOKE interest in shell thinning observed in the field in Europe and North America. More recently, there have been suggestions that polychlorinated biphenyls (PCBs), which are industrial pollutants, and organomercurial fungicides and other mercury compounds might also be implicated in this phenomenon. Concomitant laboratory investigations have assisted in determining the effects of these pollutants on many aspects of avian reproduction, including the production of abnormally thin shells. There have, however, been few reviews on the subject of thin shells. Shell thinning effects in the field in Britain have been discussed by Ratcliffe (1970). In addition, Peakall (1970a) has presented a popular discussion and Talekar (1971) has con sidered some aspects, but the only detailed review of possible mechanisms respon sible for shell thinning is that of Risebrough et al. (1970). It is unfortunate that, being published in symposium proceedings, such an important contribution may be unknown to many interested workers. In the short time since these reviews were written, many relevant research findings have been published, several of which have necessitated considerable revision and re-examination of current theories and concepts. The present paper is an attempt to review the available literature on the effects of pesticides and PCBs on shell thickness, both in the field and in the laboratory, and to assess mechanisms by which thin shells might be formed. THE PRODUCTION OF EGGS WITH THIN SHI LLS BY AVIAN POPULATIONS IN Till: FIELD After noting an increased incidence of egg destruction amongst peregrines Falco peregrimts, sparrowhawks Accipiter uisus and golden eagles Aquila chrysaetos in Britain dating from about 1950, Ratcliffe (1967, 1970) examined egg shells of known age from museums and private collections, and reported that, regionally or nation ally, the shells of all three species significantly declined in thickness during the period 1945-50. The interest aroused by the former paper is reflected by the number of investigations into similar phenomena that have since been reported, particularly from the United States and Canada. Ratcliffe (1967) devised an index, weight of shell (mg) length of shell (mm) x breadth (mm) that facilitates a comparison of the thickness of whole shells without breaking them. This index has been adopted by many other workers. In North America recent decreases in thickness index, thickness, or shell weight, have been reported for populations of the following: peregrine (Hickey & Anderson, 1968; Berger et al., 1970; Cade et al., 1971); bald eagle Haliaeetus leucocephalus (Hickey & Anderson, 1968); prairie falcon Falco mexicanus (Fyfe et al., 1969; Enderson & Berger, 1970); osprey Pandion haliaeiux (Hickey & Anderson, 1968); red-tailed hawk Buteo jamaicensis (Seidensticker & Reynolds, 1971); merlin Falco columbarius (Fox, OSH 031288 STLCOPCB4015250 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 87 1971); white pelican Pelecanus erythrorhynchos (Anderson et al., 1969); brown pelican Pelecanus occidentalis (Anderson & Hickey, 1970; Keith et a!., 1970; Risebrough cl al., 1970; Blus, 1970; Risebrough et a!., 1971); double-creasted cormorant Phalacrocorax auritus (Anderson et a!., 1969; Risebrough et al., 1970); common egret Caswerodius albas (Faber el al., 1972); great blue heron Ardea herodius (Faber et ah, 1972); guillemot Uria aalge (Gress et al., 1971); herring gull Lams argentatus (Hickey & Anderson, 1968); ashy petrel Oceanodroma homochroa (Risebrough et at., 1970). On the other hand, no change has been reported for the following: golden eagle Aquila chrysaetos (Hickey & Anderson, 1968); great horned owl Bubo virginiamts (Hickey & Anderson, 1968; Seidensticker & Reynolds, 1971); gyr falcon Falco rusticolus (Cade el al., 1971); rough-legged hawk Buteo lagopus (Cade et al., 1971); whooping crane Grus americana (Anderson & Kreitzer, 1971); mourning dove Zenaidura macroura (Kreitzer, 1971), or for some regional populations of several of the species in the first list. In Britain other species for which shell thickness has declined are the heron Ardea cinerea (Prestt, 1970) and the kestrel Falco tinnunculus, merlin, hobby Falco subbutco, osprey, rook Corvits fntgilegus, carrion crow Corvus corone and shag Phalacrocorax aristotelis (Ratcliffc, 1970). The last writer reported no change for the raven Corvus corax, golden plover Charadrius apricarius, greenshank Tringa nebularia, black-headed gull Laras ridibundus, kittiwake Rissa tridactyla, razorbill A/co torda, or guillemot. The largest recorded decrease was in a brown pelican colony on Anacapa Island off the Californian coast in 1969 (Risebrough et al., 1971), the average decrease in thickness being 50% when com pared with the pre-1943 data of Anderson & Hickey (1970). Many of these pelicans laid eggs with soft shells (Keith et al., 1970; Risebrough et at., 1971), as did common egrets in a Californian colony (Faber et a!., 1972). Reductions in shell thickness exceeding 20% have been reported for other species by Hickey & Anderson (1968), Berger et al. (1970), Fox (1971) and Cade et al. (1971). In addition to the British populations of the peregrine, sparrowhawk and golden eagle for which shell thickness decreased between 1945 and 1950 (RatclilTe, 1967, 1970), the date of the thickness decrease has been reported for the following: In California: In Britain: peregrine, 1947-52 Hickey & Anderson (1968) In Sweden and Finland: osprey, started 1947-49 Odsjo (1971) Other researchers reporting declines have compared the recently-laid thin shells with shells from eggs laid before 1940 or 1950. Care must be taken when selecting old shells for comparison with recent ones. Blus (1970) noted that before 1947 the eastern subspecies of the brown pelican laid eggs with significantly heavier and 88 A. S. COOKE thicker shells on the Gulf Coast of Florida than on the Atlantic Coast of the same Stale. The cause of these differences remains unknown. Tuck (I960), refeiring to Uspcnski's work, pointed out that shell thickness of eggs of Brunnich's guillemot Uria lomvia can change by a factor of two, depending on the nature of the ground upon which they are laid. Since the initial declines in Britain, thickness of the peregrine and sparrowhawk shells has remained more or less steady at the reduced level, while golden eagle shells have significantly increased in thickness recently but have not yet regained pre-1945 thickness (Ralclifi'e, 1970). Shells of osprey eggs laid in Sweden and Finland have progressively declined in thickness since the late 1940s (Odsjo, 1971). The factor or factors responsible for shell thinning in British raptors were discussed by Ralclifi'e, briefly in 1967 and at length in 1970. He concluded in 1970 that (1) DDT and yBHC used as domestic, veterinary, horticultural and agricul tural insecticides were probably implicated in the initial decrease in shell thickness; (2) cyclodiene insecticides contributed to the thickness decline after their widespread introduction into agriculture in the mid 1950s; and (3) PCBs might also be impli cated. Disease, other forms of pollution and various factors responsible for poultry laying eggs with thin shells were not considered important. The evidence upon which Ratcliffe's assertions were made is presented below, together with more recent evidence that generally supports his views. (1) Organochlorine insecticides and their metabolites occur in much higher residues in the tissues of predatory birds than in granivorous species (sec Moore, 1965), particularly high residues being found in raptors taking birds or freshwater fish. Thus, if organochlorines have had any harmful effects on birds in the field, birds of prey might be the most likely group to be affected. (2) There have been many temporal and geographical associations between insecticide use and birds of prey producing thin shells. F'or instance, the decrease in shell thickness for the British species sufficiently well studied coincided with the introduction of DDT. Egg shells of the golden eagle only became thinner in those parts of Scotland where the birds fed mainly on sheep carrion and so exposed themselves to insecticides in the sheep dip. Since the ban on the use of dieldrin dips (Cook, 1964) was implemented in 1966, shell thickness index in the affected regions has increased significantly (Ratclilfe, 1970). Several other examples of such relationships are quoted by Ratclilfe (1970). Odsjo (1971) reported that shells of osprey eggs were significantly thinner in Sweden and Finland during the period 1939-46 than during 1934-38, a change attributed to PCB pollution. However, this would seem to be an instance of the facts being made to fit the hypothesis, rather than the other way round. Compared with pre-1934 egg shells, those laid during 1934--38 were, on average, thicker by 008 mg/mm2, while those laid during 1939-46 were thinner by only 0 04 mg/mm2. Thus, the question that should have been asked was: Why were the 1934-38 shells thicker? DSW 031290 STLCOPCB4015252 SHELL DUNNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 89 (3) RatclifTe (1970) showed that, for 14 species, those whose eggs contained higher residues had proportionately thinner shells. Amongst brown pelican colonies in America, those birds with the highest DDE residues have been laying eggs with the thinnest shells (Keith et a!., 1970). For field material the decline in egg shell thickness or thickness index has frequently been demonstrated to be related to pollutant content in the egg, either on an individual egg basis or on a colony basis. The pollutant in the egg has not been suggested as the cause of the thin shell, but only as an indicator of the level in the laying bird. Cecil et al. (1972) considered that, in laboratory experiments, DDT-type residues in egg lipid are usually similar to those in body lipid. RatclifTe (1970) reported a relationship between shell thick ness index and total organochlorine content for British peregrine eggs. From North America a flood of papers has recently demonstrated inverse relationships between shell thickness or thickness index and DDE content of the egg for the peregrine (Cade el al., 1971), prairie falcon (Fyfe el al., 1969; Enderson & Berger, 1970; Fimreite el al., 1970), great blue heron (Vermeer & Reynolds, 1970), white pelican (index only, not thickness; Anderson el a!., 1969), brown pelican (Risebrough et al., 1970; Blus el al., 1971, 1972) and double-crested cormorant (Anderson el a!., 1969), but not for the common tern Sterna hirmulo (Switzer el al., 1971). Some of these workers have confidently presented significant correlation coefficients despite the values for egg residues clearly not being normally distributed, but even in these instances the general trend that highly contaminated eggs have thin shells is apparent. Unfortunately, interpretation of the data is hindered by the presence of other contaminants in the eggs. Within a locality, tissue concentrations of PCBs often tend to be positively correlated with DDE concentrations (Risebrough el al., 1968; Peakall & Lincer, 197(D) so it is difficult to determine the significance of the reports of egg PCB content being related to the degree of shell thinning for pelicans (Anderson et al., 1969; Risebrough el al., 1970). A similar relationship was not found for eggs of the double-crested cormorant (Anderson et al., 1969) or the great blue heron (Vermeer & Reynolds, 1970). In prairie falcon eggs both dieldrin and DDE levels were related to the decrease in shell thickness index in Wyoming and Colorado (Enderson & Berger, 1970), but no such relationship existed for mercury levels in Canadian eggs (Fimreite et al., 1970). Many writers who analysed for DDE did not report the concentration of other pollutants. For the brown pelican, Blus cl al. (1971) reported significant negative relationships between shell thickness and egg residues of pp'-DDE, pp'-DDT, pp'-TDE, dieldrin and PCBs, but further statistical analysis revealed pp'-DDE to be the only residue that was consistently associated with significant shell thinning. The contamination pattern of mercury in the egg samples was different to that of the other pollutants, a positive relationship between mercury content and shell thinning being found. A positive correlation between egg content of pp'-DDE and shell thickness has been observed for the eggs of the moorhen (Jallinula chloropus (Fowler el al., 1971). DSW 031291 ? / * STLCOPCB4015253 90 A. S. COOKF. The techniques and conclusions of Blus el al. (1971, 1972) have come under severe attack from Hazcltinc (1972). The first objection was that data from three colonies and two subspecies were grouped together by Blus et al. (1972) and the negative relationships between shell thickness and DDE content only existed on an inter-colony, and not an intra-colony, basis. Hazeltine (1972) seems rather con fused over the word `colony', since Blus el al. (1971, 1972) clearly stated that eggs were collected from twelve (not three) colonies in three different states. Lack of significance for `within-state' data, as calculated by Hazeltine (1972), was perhaps due to the small sample sizes. The Florida data of Blus el al. (1972), which may indeed show a significant negative relationship, were not examined by Hazeltine (1972). Two other objections on statistical grounds included, once again, misuse of the correlation coefficient for data not normally distributed (see above). Hazeltine's fourth objection was that BIus's eggs were apparently often partially incubated, sometimes containing cheeping embryos. Blus el al. (1971, 1972) made no mention of using incubated eggs, referring only to fresh and addled eggs. `Addled' has several meanings, but it is never applied to eggs with living embryos. When working with part-incubated eggs, Hazeltine (1972) recommended that ppm DDE in egg lipid would be a suitable measure of initial pollutant concentration in the egg, and he found a highly significant positive relationship between ppm DDE in the lipid and shell thickness for nine pelican eggs (four part-incubated) collected from Anacapa Island. Certainly, for these eggs, the rate of disappearance of DDE during incubation was similar to the rate of loss of egg lipid. If some of Blus's eggs were partially incubated, the errors should not be too serious since residues in total egg contents (not in the yolk, as stated by Hazelline, 1972) have apparently been determined, and the relative loss in weight of egg contents would be expected to be similar 1o Hazeltine's figures for DDE loss. Nevertheless, until more is known about the dynamics of egg residues in the species under study, it is dangerous to include part-incubated eggs in investigations of this nature. An added complication is that shell thickness changes slightly during incubation (Vanderstoep & Richards, 1971). In contrast to his own Anacapa data, when Hazeltine (1972) examined data supplied for sixty-five Anacapa eggs by R. W. Risebrough, there was a highly significant negative relationship between lipid DDE content and shell thickness. Hazeltine (1972) alleged that conflicting evidence may have been suppressed by some scientists intent on having DDT banned, a misguided action that can never be justified, but for which he has presented no conclusive evidence. (4) Just as field workers were stimulated into action by Ratclifle's first paper (1967), so toxicologists began deliberately exposing captive birds to organochlorines in order to (a) examine eggs for decreases in shell thickness and/or (b) study their tissues to determine possible biochemical mechanisms to explain this phenomenon. These experiments and their implications are discussed at length in the next two sections. As is stressed later, however, the mechanism(s) responsible for shell thinning still require(s) much research. The widely-quoted theory at the end of the DSW 031292 STLCOPCB4015254 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 91 1960s was that organochlorines increased sex hormone metabolism and so de creased medullary bone deposition (Peakall, 1967; Risebrough et a!., 1968), medullary bone being a labile store of egg shell calcium. Although this mechanism is now thought to play no more than a minor role in decreasing shell thickness, it was generally accepted as a satisfactory hypothesis when Ratcliffe presented his conclusions in 1970. Despite the fact that the reasons for shell thinning are now apparently less clearly defined than they seemed to be two or three years ago, during this time there have been several more laboratory experiments that have strengthened the case for organochlorines being to blame by demonstrating that these compounds can cause captive birds to lay eggs with thin shells. Hickey & Anderson (1968) were the first workers in North America to follow up the egg shell effects reported by Ratcliffe (1967) and they concluded, on the strength of the evidence available, that `the onset of the calcium change one year after the introduction of chlorinated hydrocarbons into general usage was not a random circumstance'. More recently, Peakall (1970#) and Risebrough el al. (1970) discussed the evidence for organochlorines being the cause of shell thinning. Sometimes researchers studying thin shells have perhaps not been sufficiently cautious or critical about new evidence blaming organochlorines. Some of the laboratory evidence has been misleading and much of the field evidence is circum stantial. As Robinson (1970) has pointed out, `inferences based on circumstantial evidence may be regarded as doubtful, probable or demonstrated beyond reason able doubt, according to the weight of the evidence'. Placing too much faith in chronological and geographical associations can be dangerous. For example, such associations exist between the decline of the frog Rana temporaria in Britain and cyclodiene insecticide usage, although insecticides have probably only played a minor part in the animal's decline, the relationships being coincidental (Cooke, 1972). Nevertheless, there are now so many well-documented temporal and spatial relationships for decreases in shell thickness and organochlorine contamination of the environment and/or affected species that even a sceptic must admit that organochlorines are the probable cause. The recent reports of high amounts of organochlorines in eggs with thin shells add yet more weight to the argument and, although the presence of more than one contaminant often complicates the issue, the relationship between DDE content and decrease in shell thickness certainly suggests that one or more of the DDT group of compounds is affecting the laying DSW 031293 STLCOPCB4015255 92 A. S. COOKE bird. pp'-DDE is not necessarily the molecule responsible since high DDE concen trations might simply indicate harmful levels ofa water-soluble metaboliteof pp'-DDT that is lost during conventional organochlorine analysis. The recent field evidence, coupled with the many toxicological reports, would seem to put thecausal link'beyond reasonable doubt'. Perhaps the only reason preventing total acceptance of the theory is the failure to elucidate with any degree of certainty the mechanism(s) by which organochlorines cause thin shells. However, taking into account our present lack of knowledge of some of the basic principles of egg shell formation, inability to find a mechanism is perhaps not surprising. While elucidation of the mechanisni(s) would strengthen the case against organochlorines, failure to elucidate does not weaken the case. Armed with the many recent pieces of evidence, there is no reason to doubt any of the concluding remarks of RalclitTe (1970), except perhaps that yBHC was involved at all, and only if a hitherto unexpected and unstudied factor is linked with egg shell thinning is the pesticide theory likely to be displaced. Talekar (1971) is one writer who has expressed concern that uninvesligated industrial pollutants may be at least partially responsible for the shell thinning phenomenon. One should be cautious about blaming pollution every time shell thinning is observed in the field, since other environmental variables can affect shell thickness (Tuck, I960; Anderson et al., 1970; Blus, 1970; Bjerk & Holt, 1971). EXPERIMENTS TO INVESTIGATE THE EEEECTS ON SHELL THICKNESS Ol DELIBERATE EXPOSURE OF LAYING BIRDS TO ORGANOCHLORINES AND OTHER POLLUTANTS Details of thirteen experiments in which organochlorine insecticides were reported as changing shell thickness are given in Table I. Experiments in which no change in shell thickness was observed are referred to later. Methods for determining shell thickness were reviewed by Tyler & Geake (1961). If a micrometer is used, a large number of measurements is required to produce an accurate value of mean shell thickness, since, longitudinally, shell thickness can vary considerably (Tyler, 1961). For chicken eggs, Tyler (1961) recommended twelve measurements per shell in a longitudinal line from broad cap to narrow cap, but the writers listed in Table 1 who used a micrometer made four or less measure ments per shell, usually at the waist. Although thickness at the waist often differs greatly from mean shelf thickness (Tyler 1961), providing an adequate number of shells has been measured, results and conclusions should be satisfactory since information is required on differences between control and treatment groups of shells, rather than on absolute values for individual shells. Both Ratclilfe's index (1967) and a direct measure of shell weight per unit area (Tyler & Geake, 1961) permit an accurate, rapid assessment of shell thickness, overcoming the tedious replication of measurements at different points on the same shell. DSW 031294 STLCOPCB4015256 SHI IX THINNING IN AVIAN EGGS BY ENVIRONMCNTAE POLLUTANTS 93 An increase in thickness was reported in only one of the experiments shown in Table 1 (JetTeries, 1969), but in this experiment egg weight decreased for the dosed birds and there was, in fact, no difference between the weight of shell material laid down by the control and dosed groups, suggesting no effect on shell formation. In eleven of the twelve experiments reporting shell thinning, insecticide was administered in the diet, and of these only in the experiment on wild falcons described by Enderson & Berger (1970) was alternative unconlaminaled food available. Jefferies (1967, 1969) and Prestt et al. (1970) recognised that partial starvation due to dietary rejection after the addition of organochlorincs might mask any direct effects the ingested organochlorines were having on the birds under study. They took precautions against this by presenting untreated seed to Bengalese finches Lonchura striata for 21 h/day and the treated diet for only 2 h. Stephen et al. (1970) fed chickens, Callus domesticus, on diets containing 20 ppm DDT + 20 ppm DDE and considered that any decreases in percentage shell calcium were due to decreased calcium consumption, rather than to a direct effect of the insecticide. In none of the ten experiments in which thin shells were observed after birds had been fed solely on diets with added insecticide were figures for food consumption given. The experimental design is open to criticism (e.g. by Davison & Sell, 1972) when the additive under investigation might lead to partial rejection of the diet and so cause a bird to lay eggs with thin shells because of reduced food intake. The effect of withholding food on shell thickness can be illustrated by an experiment I carried out on twelve chickens. The birds were fed normally for five days, food was then withheld on the sixth and seventh days, and the usual diet was resumed on the eighth day. Mean shell thickness for the group decreased by 16% on day 7 and by 22% on day 8, but recovered to the normal value by day 12. Apart from Stephen et al. (1970), however, writers have only reported decreased consumption of DDT-conlaniinated diets when the DDT level was high. No rejection was reported at 250 ppm by chickens (Noakes & Benfield, 1965), at 300 ppm by chickens (Lillie et al., 1972), at 400 ppm by pheasants Phasianus colchicus (Genelly & Rudd, 1956) or at 500 ppm by bobwhite quail Colinus rirginianus (Linduska & Springer, 1951), but a diet containing 700 ppm DDT was rejected by Japanese quail Coturnix coturnix japonica (Cross et al., 1962). In contrast, a diet containing only 25 ppm dieldrin was rejected by pheasants (Genelly & Rudd, 1956), food consumption being only 86% that of birds on a control diet. 50 ppm dieldrin lowered food consumption to 71 %, and a direct relationship between consumption and egg production was apparent. On the other hand, Gillett & Arscott (1969) noted that young Japanese quail on a diet containing 5 ppm dieldrin consumed more than those on the basal diet. Atkins & Linder (1967) observed decreased food consumption after pheasants had been given dieldrin in capsules for thirteen weeks (6 mg/bird/week, an intake equivalent to 16 ppm in the diet). Here the dieldrin cannot have affected the palatability of the diet, yet food intake STLCOPCB4015257 TABLE 1 EXPERIMENTS IN WHICH EGG SHELL THICKNESS HAS BEEN AFFECTED AFTER EXPOSURE OF THE LAYING BIRD TO ORCANOCHLORINE INSECTICIDES Author(s) Species Poilutanl(s) Exposure Mean egg content (ppm) Shell measurement rjr , , ,, w . , tjfect on shell thickness Bitman et al. (1969) Japanese quail pp'-DDT'l 100 ppm in low 5203 total residues Thickness at 3 Decrease. 6/a\ Compared C. coturnix japonica or V Ca diet for 45 op'-DDTJ days J27 > of DDT type points around Decrease, 4/ t with controls compounds waist of shell ' on low Ca without diet membranes^ Lehner & Egbert Mallard Dieldrin Up to 10 ppm* Up to 44 Thickness at Decrease, up to 4%i (1969) Anas plaryrhyttchos in diet for broad pole and 1-2 years waist of shell plus outer membrane'!' Porter & Wiemeyer American Kestrel pp'-DDT Up to 15 ppm Not given Thickness at 4 Decrease, up to 17%J (1969) Falco sparverius plus pp'-DDT plus points of shell dieldrin 3 ppm dieldrin* plus membranesf in diet for 1-2 years Heath et a!. (1969) Mallard pp'-DDE Up to 40 ppm* Not given Thickness of Decrease, up to 13 %J (see also Heath et al.. or in diet for be shell plus in press) pp'-DDT tween 1 and 2 membrane+ vears o u> Jefferies (1969) Bengalese finch Lonchura striata pp'-DDT Up to 300 pg/ Not given day* in diet for Shell weight/egg Increase by average of 7% weight about 15 weeks N) Enderson & Berger Prairie falcon Dieldrin Wild birds fed on 40 -- vO (1970) Falco mexicanus up to 12 tethered O' starlings pre- Thickness index Decrease of 9% (compared with controls) for 7 most contaminated eggs viously'dosed with dieldrin DSW STLCOPCB4015258 96 A. S. COOKt: was reduced. Unfortunately, these writers did not measure shell thickness. Jefferies & French (1971) noted a similar loss of appetite when pigeons were treated with encapsulated pp'-DDT at the much higher level of 54mg/kg/day. In addition, the food consumption of young cockerels on a diet containing 500 ppm DDT began to drop after the birds had been subjected to the treatment for about six weeks (Srebocan et al., 1971). Body residues of organochiorines must have been respon sible for this partial loss of appetite. The longer a bird is fed on a contaminated diet, so the likelihood is increased of tissue residues building up to an effective level, unless this is above the plateau level. For chickens being fed DDT, this critical effective level seems to be very high. Feeding birds on a diet containing 25 ppm pp'-DDT for twenty-eight weeks followed by 300 ppm for twelve weeks led to small, but non significant, decreases in food consumption (Lillie et a!., 1972) although the mean content in the abdominal fat of the birds at the end of treatment was 2,300 ppm (Cecil et a/., 1972). The chicken is, however, an atypical species, tending to be resistant to DDT {see Noakes & Benfield, 1965; Lillie et al., 1972; and later in this and in other sections). Other species may start to show a loss of appetite at much lower tissue concentrations. Nevertheless, while it is possible that reduced food intake has contributed to shell thinning in some of the experiments in Table 1, the food rejcclion/appelile loss data available at present indicate that reduced food intake cannot account for the great majority of observed decreases in thickness summarised in the table. In the two experiments in which food intake has been controlled or determined, however, the administered pesticide has failed to cause shell thinning (Davison & Sell, 1972; Cecil et al., 1972). These two experiments contrast markedly with those of Smith et al. (1970) and Saulcr & Steele (1972). In each experiment, While Leghorn chickens were treated by adding DDT to the diet. Significant shell thin ning was reported by Smith et at. (1970) after feeding 10 ppm technical DDT for (wo months and by Sauter & Steele (1972) after feeding 01 ppm technical DDT for ten weeks. On the other hand, no shell thinning was found by Davison & Sell (1972) after feeding 200 ppm pp'-DDT for twelve weeks or by Cecil et al. (1972) after feeding 50 ppm pp'-DDT or op'-DDT for twenty-eight weeks, followed by 300 ppm for twelve weeks. It could be argued that the compound responsible for shell thinning is a component (or its metabolite) of technical DDT, other than op'- or pp'-DDT. Since pp'-TDE can probably be ruled out as well (Heath et al.. 1969), the dietary level of this proposed compound in the experiment of Sauter & Steele (1972) must have been less than 0 004 ppm, the concentration of any other con stituent in technical DDT being less than 4% (Metcalf, 1955). The purity of the pp'-DDT used by Davison & Sell (1972) was 99%, so their 200 ppm pp'-DDT diet might be expected to have contained a higher level of the proposed active com pound than the 01 ppm technical DDT diet of Sauter & Steele (1972). Further work is needed before one can even consider ascribing potent shell thinning properties to a minor component of technical DDT. Other differences exist between DSW 031297 STLCOPCB4015259 Peakait (19706) Wiemeyer & Porter (1970) Smith et al. (1970) Stickel & Rhodes (1970) Tucker & Haegele (1970) Ring dove Streptopelia risoria American kestrel pp'-DDE pp'-DDE Chicken Gallus domesticus Japanese quail Technical DDT pp'-DDT Bobwhite quail 3 Colinus virginianus > Mallard J Mallard Technical DDT Technical DDT Longcore et al. (19716) Black duck Anas rubripes Sauter & Steele (1972) Chicken pp'-DDE Technical DDT or lindane Single intraperi- 80 Weight Decrease by 23 " ,, toneal injec tion of 150 mg/kg 10 ppm in diet 32 Thickness at 4 Decrease by 10% for more than 1 points ot shell year plus membranesf Up to 10 ppm* Up to 6 ppm in yolk Not given Decrease, up to 11 %i in diet for 2 months Up to 25 ppm* Not given Not given Decrease, up to 7%J in diet for 26 weeks Up to 30 ppm* Not given Thickness at 4 ( Decrease, up to 4%i in diet for 96 days points around i Decrease, up to 5%t waistt Single oral dose Not given As above Decrease by 9 % over 16 1000 mg/kg. Food days withheld for 2 days after treatment Up to 30 ppm* Up to 135 Thickness at Decrease, up to 38 in diet for 6 waist and both months poles of shell plus membranes^ Up to 10 ppm* Not given Not given Decrease, up to 9%i in diet for 10 after treatment period weeks DSW 0 3 1 2 9 8 * Highest treatment dose t Measured with a micrometer J Mean thickness change for most affected treatment group SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL, POLLUTANTS 97 observations made in these four experiments, which may simply illustrate how similarly-designed studies can sometimes produce totally different results. Several other experiments have been reported in which organochlorines have had no effect on shell thickness. Cecil et al. (1971a) repeated the work of Bitinan et al. (1969) except that (1) quail were exposed to pp'-DDE and not op'-DDT, (2) exposure time was increased from forty-five to seventy-four days and (3) the level of dietary calcium was adequate and not deficient as before. No change in shell thickness was found. Assuming there is a misprint in the shell thickness units in the table of Cecil el al. (1971a), the 0 6% low calcium diet of Bitinan et al. (1969) appears to have contributed to shell thinning to a much greater extent than the 100 ppm DDT added to the diet. Heath cl al. (1969) reported no significant change in shell thickness after 40 ppm TDE was added to the diet of mallards. Bobwhite quail fed on a diet containing 30 ppm DDE throughout a laying season did not produce thin-shelled eggs, unlike mallards on a 10 ppm DDE diet for two seasons (Heath et a!., in press; this being a similar experiment to that of Heath et a/., 1969). DDE had no marked effect on shell thickness in chickens at dietary levels sufficient to produce concentra tions in the abdominal fat of greater than 2000 ppm (Cecil et a!., 1972). Dieldrin has been found to have no effect on shells laid by ring doves Streptopelia risoria (Peakall, 19706), pheasants (Dahlgren & Linder, 1970), chickens (Robinson, 1970; Davison & Sell, 1972) or mallards (Muller, 1971a). Whitehead et at. (1972) admini stered encapsulated lindane to chickens at a level equivalent to about 100 mg/kg diet, but shells were of normal thickness. This is in sharp contrast to the study of Sauter & Steele (1972) in which chickens exposed to dietary levels of 01 mg lindane/kg diet laid eggs with thin shells. Earlier studies on the effects of insecticides on avian reproduction and survival (e.g. Genelly & Rudd, 1956; Cross et al., 1962) were published before the recent upsurge of interest in shell thickness. Since no mention was made in these papers of thickness changes, obvious changes are unlikely to have occurred. A decrease in shell thickness of even 10% would probably have been missed unless the shells were weighed or the thickness measured. One can only speculate at the number of experiments that yielded negative results, which were not considered worth publishing. Single oral doses of one of the commercial PCB mixtures, Aroclor 1254, sufficient to disrupt and then terminate laying in Japanese quail and mallard, produced thin shells in the few eggs laid before complete termination (Tucker's experiment referred to by Peakall & Lincer, 1970a). However, Tucker & Haegele (1970) reported that mallards, similarly treated with Aroclor 1268 and then starved for three days, laid eggs with shells of approximately normal thickness. In addition, ring doves, fed 10 ppm Aroclor 1254 in their diet for six months or injected intrapcritoneally with 160 mg/kg Aroclor 1254, laid eggs with normal shells (Peakall, 1971). Similarly, administration of enough encapsulated Aroclor 1254 to adult pheasants to decrease egg production and hatchabilily had no effect on shell DSW 031299 l STLCOPCB4015261 98 A. S. COOKE thickness (Dahlgren & Linder, 1971). Neither mallards fed on (I) a diet containing 500 ppm Aroclor 1254 for thirty-nine days or (2) 25 ppm for two laying seasons, nor bobwhite quail fed on a diet containing 50 ppm for a single season, laid eggs with thin shells (Heath et al., in press). Chickens maintained on diets containing 100 ppm Aroclor 1242 or 1254 laid eggs with shells thinner by 17 and 24% respec tively, but dietary consumption, reduced by 37 and 33% respectively (Monsanto Company, 1971), could have accounted for this. Mean shell thickness of a group fed 100 ppm Aroclor 1260 was only 2% below that of the control group, food consumption being reduced by 6%. In a second experiment using Aroclor 1242, neither food consumption nor shell thickness was affected (Monsanto Company, 1971). Risebrough et al. (1970) observed that gallinaceous species tend to be more resistant than birds in other orders to decreases in shell thickness after exposure to organochlorine compounds. The trend is still apparent when later evidence is considered. Bobwhite and Japanese quail seem rather resistant, no changes in shell thickness being reported after treatment by Cecil et al. (1971a) and Heath et al. (in press) and small changes or changes only after severe treatment were reported by Bitman et al. (1969), Stickel & Rhodes (1970), Peakall & Lincer (1970a) and Tucker & Haegele (1970). In contrast to these experiments, Japanese quail main tained for 21 days on a diet containing 225 ppm technical DDT laid eggs with shells of greatly reduced thickness both during and soon after treatment, the maximum thickness reduction for a shell being 60% (McFarland et al., 1971). Although chickens have been observed to lay eggs with thin shells following organochlorine treatment (Smith et al., 1970; Sauter & Steele, 1972), they usually show no response or else a response only under severe dietary exposure (Robinson, 1970; Stephen efal., 1970; Cecil et al., 1972; Davison & Sell, 1972; Heath et al., in press). Similarly pheasants are unaffected by dieldrin (Dahlgren & Linder, 1970) or PCBs (Dahlgren & Linder, 1971). On the other hand, deliberately exposed falcons have suffered quite substantial decreases in shell thickness (Porter & Wieineyer, 1969; Enderson & Berger, 1970; Wiemeyer & Porter, 1970). Japanese quail eggs containing 520 ppm total residues of pp'-DDT and its metabolites had shells only 6% thinner than controls (Bitman et al., 1969) but eggs from American kestrels Falco sparverius containing only 32 ppm DDE had shells 10% thinner than controls (Wiemeyer & Porter, 1970). Ring doves and mallards, the other species used in these experi ments, seem intermediate in their response, sometimes producing thin shells (Lehner & Egbert, 1969; Heath et al., 1969; Peakall & Lincer, 1970a; Peakall, 19706; Longcore et al., 19716) and sometimes not (Heath et a!., 1969, in press; Peakall, 19706, 1971; Muller, 1971a). So, if it is possible to generalise, bearing in mind the very diverse sets of conditions that different experimenters have employed, gallinaceous species tend to show less of a response in terms of red need shell thickness than ring doves and mallards arid these, in turn, tend to be more resistant than falcons. Since birds of prey are more susceptible in the laboratory and their tissues OSW 031300 STLCOPCB4015262 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 99 tend to contain higher residues in the field, the fact that they are the species most affected in the field further supports the theory that pesticides are responsible. It is, of course, not surprising that such taxonomic differences should occur when one is considering the response to a group of compounds, the organochlorine insecticides and the PCBs, that are closely related both physically and chemically. In contrast, Tucker & Haegele (1971), who examined the acute oral toxicity to six avian species of sixteen pesticides, including carbamate and organophosphate insecticides, dieldrin and strychnine, reported that the average sensitivity of any one species to this broad spectrum of compounds was not statistically different from that of any of the other five species. Nevertheless, it is interesting that the six species in order of increasing resistance were the house sparrow Passer domesticus, mallard, pigeon Columba livia, pheasant, Japanese quail and ehukar partridge Alectoris graeca, the three gallinaceous species tending to be more resistant. Compounds of the DDT group or their metabolites usually reduced shell thick ness of captive birds to a greater extent than dieldrin or PCBs. For example, doves injected intraperitoneally just before laying with single doses of 160 mg/kg Aroclor 1254 (Peakall, 1971) or 30nig/kg dieldrin (Peakall, 1970>) laid eggs with normal shells, while those similarly injected with 150 mg/kg pp'-DDE laid eggs with shells 23% thinner than normal (Peakall, 19706). Mallards maintained for two seasons on a diet containing 10 ppm DDE laid eggs with thin shells, while those on a diet containing 25 ppm Aroclor 1254 produced normal shelled eggs (Heath et a!., in press). This, coupled with the field evidence presented in the previous section, suggests that DDT-typc compounds have a greater effect on shell thickness than the other organochlorines so far studied. Captive birds have also been used to study the effects of mercury on shell thick ness. Japanese quail fed on diets containing up to 8 ppm mercuric chloride (Stoewsand et al., 1971) laid eggs with thin shells. Feeding chickens (Tejning, 1967) or pheasants (Fimreite, 1971) on methylmercury-treated grain led to a marked increase in soft shelled eggs. Although Fimreite (1971) noted a significant reduction in food intake for birds at the highest dietary mercury levels, significantly more soft-shelled eggs were also laid by pheasants on the lower levels, demonstrating that the ingested mercury must be affecting shell formation. There is, however, evidence indicating that mercury-induced shell thinning does not occur in the field (Fimreite et al., 1970; Blus et a!., 1971). Regarding the organophosphate insecticides, Sauter & Steele (1972) reported that diazinon and malathion caused egg shell thinning in chickens, and Muller (197ln, b) observed that parathion had a similar effect on captive mallard and partridge Perdix perdix. There are now more than twenty reports of organochlorine or organophosphate insecticides, PCBs or mercury compounds causing thin shells in experimental birds. Although several experimental designs or techniques are open to criticism and some nagging doubts about reduced food consumption still remain, the experiments DSW 031301 STLCOPCB4015263 100 A. S. COOKE nevertheless appear to demonstrate positive shell thinning due to treatment. Sometimes severe treatment conditions were required to produce even only slight shell thinning. For example, Stickel & Rhodes (1970) reduced the light period from 14 to 8 h at the end of treatment and several birds died from DDT poisoning. Bitman et ah, (1969) reported that after pp'-DDT treatment, their quail contained on average 1560 ppm pp'-DDT + metabolites in the fat and 240 ppm in the liver. Levels as high as this are only very rarely reported in field samples (e.g. Risebrough et ah, 1968; Prestt, 1970; Keith et ah, 1970). Of the remaining experiments on captive birds referred to in Table I, only in the papers by Peakall (19706) and Smith et al. (1970) are tissue analyses given. In the former paper, an average of 77 ppm DDT in the shell gland tissue of ring doves was associated with shell thinning of 23%, and in the latter 117 ppm DDT plus metabolites in the abdominal fat of chickens was associated with shell thinning of 11 %. No doubt at least some of the experiments yielding negative results failed because conditions were not severe enough. The severity of the treatment should depend on what the experimenter is trying to determine. If he is attempting to elucidate the mechanism for shell thinning then he is justified in using more extreme conditions in order to produce measurable biochemical or physiological changes than if he simply wishes to reproduce field effects. So far these controlled experiments have failed to reach the 50% average reduction in shell thickness reported for the brown pelicans on Anacapa (Risebrough et ah, 1971), although reductions exceeding 20% have been reported (Peakall, 19706; Longcore et ah, 19716). Tucker & Haegele (1970) observed that greater decreases in shell thickness in captive mallard were produced by acute doses of DDT than by chronic doses, and they pointed out that, in the field, fasting and stress during the reproductive period might mobilise sufficient pesticide residues to cause shell thinning. POSSIBLE MECHANISMS FOR THIN EGG SHELL FORMATION The case against organochlorincs would be considerably strengthened if mechanisms by which they cause thin shells could be precisely elucidated. This would then permit a clearer assessment of the importance of other environmental variables and it would also become possible to evaluate the effect of residues at the site of action. Normal egg shell formation Before discussing possible mechanisms for thin egg shell formation after exposure of a bird to organochlorines, it is necessary to outline what is known about the process of shell formation. Research on this topic has been largely carried out on chickens and this atypical species must therefore serve as a model for all species. For most aspects of shell formation, reviews are quoted which discuss the subject in much greater detail than is permitted here. DSW 031302 STLCOPCB4015264 SMELL HIINNINC IN AVIAN EGGS DY ENVIRONMENTAL TOLLUTANTS 101 Egg formation was reviewed by Gilbert (1967). Ovulation is controlled by pituitary gonadotrophin secretion (reviewed by Gilbert, 1971 a) and the egg is formed in the oviduct (Fig. l).Oviducal structure and function have been reviewed by Ailken (1971). In most avian species, including the chicken, only the left ovi duct usually develops and is functional. The ovum passes via the infundibulum to the magnum where albumen is secreted by the epithelial cells; then to the isthmus region where shell membrane material, the first egg shell component, is laid down. Shell formation has recently been reviewed by Sintkiss (1968), Wilbur & Simkiss (1968), Taylor (1970) and Simkiss & Taylor (1971). Recent evidence (Robinson et al., 1966, 1968) indicates that shell calcification is initialed in the isthmus and although this is generally accepted (Cooke & Balch, I970n; Simkiss & Taylor, 1971), Fujii & Tamura (1970) have described shell initiation occurring in the next region, the shell gland. Certainly the main part of the shell is laid down in the shell gland, where an egg usually remains for 16-20 h. About 95 % of the shell is mineral matter (water content <2%) and of this more than 95% is calcium carbonate (Romanoff & Romanoff, 1949) in its most stable polymorphic state, calcite. The calcium metabolism of the laying bird has been reviewed by Taylor & Stringer (1965) and Simkiss (1967) and has been discussed in a popular article by Taylor(1970). Calcium that is eventually incorporated into the shell comes initially, of course, from the diet. Transport of calcium across the intestinal wall has been reviewed by Simkiss (1967) and Wasserman (1968). Plasma calcium is either ionic or protein-bound and the two forms are believed to be in equilibrium. About 25-40% of the calcium incorporated into the shell does not come directly from the diet (Comar & Driggers, 1949; Driggers & Comar, 1949), but is stored in the form of medullary bone (Kyes & Potter, 1934), spicular growth that occurs inside w-ellvasculated bones (Taylor & Moore, 1953, 1956) and is most obvious in the medul lary cavities of the limb bones. Oestrogens and androgens act synergistically to control medullary bone deposition (Taylor & Stringer, 1965). Calcium in the medullary bone is mobilised and carried via the blood to the shell gland. The role of the parathyroid and ultimobranchial secretions in regulating mobilisation has been discussed by Taylor (1966, 1970) and by Simkiss & Dacke (1971) respectively. It is generally accepted that parathormone is responsible for the osteoclastic activity in medullary bone when the shell is being calcified. Probably it is the ionic calcium fraction that is depleted as the blood passes through the shell gland (Simkiss & Taylor, 1971). Calcium is accumulated in the ciliated epithelial cells of the shell gland (Gay & Schraer, 1967) and is actively transported into the shell gland lumen (Schraer el al., 1965; Schraer & Schraer, 1970), perhaps bound to a carrier protein (Corradino et al., 1968). The origin of the shell carbonate fraction is discussed at some length below'. Egg shell nomenclature can be very confusing since various writers use the same word or phase to describe different parts of the shell, and some of the old phrases, such as `spongy layer', can be misleading. A new nomenclature system proposed DSW 031303 STLCOPCB4015265 102 A. S. COOKE Fig. 1. Diagram of a chicken's ovary and oviduct; a: ovary; b: infundibulum; c: magnum; d; isthmus; e: shell gland containing a calcifying egg; f; vagina; g; opening to cloaca. DSW 031304 STLCOPCB4015266 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 103 by Schmidt (1957, 1962a, b) and Tyler (1965) is shown in Fig. 2 and will be used in this paper. The term `mammillary layer' is retained to describe the cone layer plus basal caps. The mineralised shell begins as spherulitic crystal growth around mucoprotein concentrations on the outside of the outer shell membrane (Fig. 2). These organic aggregations with deposited calcite crystals are called mammillary cores, and the nature of the mucoprotein has been investigated histochemically (Robinson & King, 1968) and biochemically (Cooke & Balch, 1970a). The organic cores initiate outer surface of shell Fig. 2. Diagrams of radial sections of shells to show (i) spherulitic growth on the surface of the shell membranes and (ii) the structures that can occur in a fully-formed shell. Nomenclature after Schmidt (1957,1962a, b) and Tyler (1965); a: shell membranes; b: mammillary core; c: spherulitic growth around the organic mammillary core; d: basal cap; c: cone layer; f: palisade layer; g: surface crystal layer; h: cuticle; i: cover; j: pore. The surface crystal layer and the cover are absent from the shells of many species. crystal growth (Fujii & Tamura, 1970), perhaps by providing a convenient lattice for calcite formation or by binding mineral ions either electrostatically or by chelation, thereby exceeding the solubility product of calcium carbonate and so achieving precipitation (Wilbur & Simkiss, 1968; Cooke & Balch, 1970a). Inward growth of the spherulites is inhibited by the presence of the shell membranes, so the crystals tend to develop outwards. They coalesce and the growth is then columnar, forming the palisade layer. In some places coalescence does not occur, possibly because of liquid diffusing through the shell during formation (Tyler, 1965) and pores are formed. Running throughout the calcified shell is a continuous organic phase, the matrix, which again is largely mucoprotein (Baker & Balch, 1962; Cooke & Balch, 19706). This organic matter probably plays an intimate role DSW 031305 / STLCOPCB4015267 104 A. S. COOKE in shell development since electron micrographs show that matrix deposition precedes crystallisation (Simons, 1971). Around the outside of the shell is an organic cuticle. Possible factors controlling oviposilion have been reviewed by Sturkie (1965) and Gilbert (1967, 19716). In a modern breed of chicken, oviposition occurs about 24 h after ovulation. Mechanisms by which pesticides might cause thin shells The poultry industry has been concerned for many years with possible methods of reducing loss due to thin shells. Although many workers have studied shell forma tion and treatments that cause shell thickness to change, our knowledge of these subjects is still sketchy and incomplete. Nevertheless, literature on thin shells, experimentally induced in chickens, is freely referred to in this section in an attempt to find similarities between the action of various treatments known to cause thin shells and the action of pesticides. Basically, possible mechanisms can be divided into four categories: (I) those reducing the availability of the calcium ions, (2) those reducing the availability of the carbonate ions, (3) those affecting other shell constituents upon which proper shell formation depends, and (4) miscellaneous mechanisms producing disruptive changes. Each will be discussed in turn. 1. Reduction of calcium availability Between entering the gut and becoming incorporated in the shell, calcium might be blocked by the action of organochlorines during five different stages: (a) absorption from the gut, (b) deposition of medullary bone, (c) mobilisation of medullary bone, (d) transport in the blood and (e) transport from the blood to the developing shell. 1() Absorption from the gut: Most of the mechanisms involved in the transport of calcium from the intestinal lumen to the blood are dependent on the steriod vitamin D (reviewed by Simkiss, 1967; Wasserman, 1968). An active form was shown by Blunt et al. (1968) to be the 25-hydroxy derivative of vitamin D3 (cholecalciferol), and Anderson el al. (1969) suggested that since organochlorines can induce hydroxylation of steroids (see section 1(6) below), exposure to insecticide might lead to the production of inactive molecules hydroxylated on the wrong carbon atom. Furthermore, Risebrough et al (1970) have argued that organo chlorines might affect egg shell formation by inhibiting ATPases involved in active calcium transport across the intestinal wall or shell gland, pointing out that DDT inhibits the in vitro incorporation of labelled phosphorus into ATP in insect respiratory particles (Avi-Dor & Gonda, 1959). Koch et at. (1969) have reported inhibition in vitro of insect nerve, muscle and brain ATPases by pp'-DDT and by some of its metabolites. DSW 031306 STLCOPCB4015268 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 105 Experiments designed to investigate whether pesticides atTecl calcium uptake from the gut have, however, had conflicting results. Peakall (1969) observed that DDT failed to reduce the uptake of labelled calcium in the intestine of the zebra finch PoephUa gwtata. On the other hand, Nowicki et al (19726) reported that, in rachitic chicks treated with vitamin D3, DDT decreased the normal rise in calcium absorption from the gut. A companion study was carried out to determine whether detrimental effects on vitamin D3 metabolism were associated with this change in calcium absorption (Nowicki cl al, 1972a). Recently, the most active metabolite of vitamin D3 in the intestine was found by Norman and his co-workers to be 1,25-dihydrocycholecalciferbl. In the rachitic chicks, Nowicki et al (I972n) observed no detrimental effects on hepatic or renal conversion of vitamin D3 to its active 25- and 1,25-hydroxylatcd derivatives. Peakall (1969) reported that, in excised liver, the production of polar metabolites from vitamin D is unalTectcd by DDT. Stephen el al. (1970) fed chickens on diets containing DDT plus DDE, some diets being supplemented with sixty-five times the normal requirement of vitamin D3 in order to negate possible increased steroid metabolism. Shell calcium tended to decline for all groups except the controls, and these decreases were attributed to decreased food consumption. Thus, further work is required (1) to explain the conflicting observations of Peakall (1969) and Nowicki el al. (19726) on calcium absorption, and (2) to investi gate other mechanisms that may be involved, such as inhibition of ATPases (Risebrough et al, 1970). 1(6) Deposition of medullary bone: Medullary bone deposition is controlled by the synergistic action of oestrogens and androgens (Taylor & Stringer, 1965). Organochlorine insecticides increase the activity of hepatic microsomal enzymes involved in the oxidative metabolism of drugs and steroids. The earlier work on this subject has been reviewed by Kupfer (1967). In mammals organochlorincs were observed to increase the activity of enzymes responsible for hydroxylating the sex hormones (Kuntzman et al, 1964; Conney et al, 1967), and this led Peakall (1967) to demonstrate with pigeon livers that the rate of production of polar metab olites from testosterone and progesterone increased if the birds had been treated with dieldrin or DDT. A similar experiment was carried out by Risebrough el al, (1968), who found increased breakdown of oestradiol after administration of pp'-DDT, pp'-DDE or PCBs. Although the pigeons used by Peakall (1967) were fed for seven days on a diet contaminated with only 10 ppm DDT, to achieve a significant increase in the production of polar metabolites of oestrogen in chicken liver, birds had to be pre-treated with a diet containing 500 ppm DDT for eighteen days (Britton, 1970). Thus, there is a considerable interspecific difference in re sponse. A PCB level of 250 ppm in the diet was sufficient to suppress comb growth of young cockerels, but this anti-androgenic-type effect may have been due, at least in part, to decreased food consumption (Platonow & Funnell, 1971). Rise brough el al. (1968) were the first to propose that organochlorines might cause thin DSW 03130? n STLCOPCB4015269 106 A. S. COOKE shells by increasing the rate of metabolism of the sex hormones, thereby reducing medullary bone deposition. This idea received wide acceptance. Peakall 0970/!?) subsequently demonstrated that pp'-DDT decreased both blood levels of oestradiol and medullary bone deposition in the ring dove early in the breeding cycle, and Oestreicher et at. (1971) reported that dietary DDE significantly depressed medullary bone formation in pigeons. Other evidence, however, does not support this hypothesis. (i) In three experiments thin shells have been laid without an accompanying anomalous change in medullary bone (Bitman et at., 1969; Peakall, 1970h; Risebrough et at., 1970). In addition, Muller & Lockman (1971) found that feeding young chicks on a diet containing as much as 10 ppm dieldrin for four weeks had no effect on calcium deposition in the tibia. It is perhaps significant that the experiments demonstrating medullary bone depletion have been carried out on pigeons and doves. The X-ray studies ofSchraer & Schraer (1970) show that there is a pronounced rhythmic rise and fall in `bone mass' for the pigeon in phase with egg laying, while, in the chicken, a species that lays continuously instead of in two-egg clutches, such cyclic changes are not apparent. (ii) Several experiments have failed to demonstrate positive microsomal oxidative enzyme induction in birds treated with organochlorines. Stephen et at. (1970) could not produce consistent enzyme induction in chickens with DDT or dieldrin. When young Japanese quail were treated with dieldrin, microsomal aldrin epoxidase activity was increased, but DDT depressed the activity in adult quail (Gillett & Arscolt, 1969). These writers also found that in pheasants caught in an alfalfa-seed growing area with a history of severe treatment with DDT and other insecticides, aldrin epoxidase activity was lower than normal. After treating Japanese quail with the op'- or pp'- isomers of DDT, DDE or TDE, Bitman et at. (1971) reported an increase in pentobarbital sleep time, indicating a slower breakdown of the drug in the liver. Later, Sell et at. (1972) showed that DDT caused a reduction in hepatic oxidase activity in the same species. Cytochrome P45o is thought to be involved in these microsomal oxidative reac tions (see Kupfer, 1967) but attempts in the above experiments to relate changes in the level of this cytochrome with changes in enzyme activity have not been successful (Gillett & Arscott, 1969; Stephen et at., 1970; Sell et al.`, 1972). It is interesting to note that in all these experiments resistant gallinaceous species have been used. (iii) The enzyme induction work of Risebrough et al. (1968) indicates that PCBs are more efficient inducers than DDE and it is thought that PCBs act in the same manner (Lincer & Peakall, 1970; see also Kufper, 1967). However, both in the field (Anderson et at., 1969; Blus el at., 1971) and in experiments on captive birds (see previous section) DDE has apparently had a greater effect on shell thickness. (iv) Eggs collected in the field that contain only small amounts of DDE often display considerable shell thinning (Anderson et at., 1969; Ratcliffe, 1970; Peakall, DSW 031308 SHELL THINNING IN AVJAN EGGS BY ENVIRONMENTAL POLLUTANTS 107 1970o; Blus et al., 1972). If thin shells were caused by increased hormone metabo lism, at these low DDE levels a feedback mechanism would be expected to maintain sex hormone levels via increased gonadotrophin secretion. (v) Lower levels of circulatory oestrogen might not have been caused by increased formation of hydroxylated derivatives. In the laboratory organochlorines have been found to delay ovulation (Jefferies, 1967; Bitman et al., 1969; Peakall, 1970a; Cecil et al., 1971a), and in the field spermatogenesis in wood-pigeons Cohtmba palumbus was retarded in Cambridgeshire in 1961, perhaps because of sublethal insecticide residues (Lofts & Murton, 1966). Peakall (1970a) stated that the delay in ovulation, `evidently was caused by the depression of the estrogen level resulting from the induction of liver enzymes by the pesticide'. Reduced secretion of gonado trophins from the pituitary, however, seems a more plausible explanation of delayed ovulation and reduced oestrogen levels. In order to explain how increased metabo lism of sex hormones could result in low oestrogen levels being maintained, one has to assume disruption of the feedback mechanism (see Sturkie, 1965) that would normally correct low oestrogen levels by increasing gonadotrophin secretion. Thus, it seems likely that organochlorines have a direct or indirect effect on the pituitary. The evidence suggests that although impairment of medullary bone deposition has been reported for pigeons and doves, this mechanism is unlikely to play any part in shell thinning observed in gallinaceous species or perhaps in other species. 1 (r), (d) Mobilisationfrom the medullary bone and transport in the blood: Medullary bone mobilisation is probably controlled by parathormone (Taylor, 1966, 1970), while calcitonin, a hormone secreted by the ultimobranchial glands, is probably responsible for limiting the extent of the bone resorption (Simkiss & Dacke, 1971). Hypostimulation of the parathyroids or hyperslimulatiou of the uliiniobranchials might lead to hypocalcaemia but the effect of pesticides on these glands remains unknown. Nowicki et al. (1972b) demonstrated that, in rachitic chicks, the vitamin Dj-mediated rise in serum calcium due to increased bone resorption could be inhibited by pre-treatment with pp'-DDT. Inhibition was not caused by impair ment of the production of hydroxylated vitamin D3 metabolites in the liver or the kidney (Nowicki et a!., 1972a) and an alternative mechanism has not yet been suggested. Since serum oestrogen levels can be decreased either directly or indirectly by pp'-DDT (Peakall, 19706) and since this sex hormone regulates the synthesis in the liver of phosvitin, a protein concerned with the transport of much of the calcium in the blood (Taylor, 1970), DDT might restrict transport of scrum calcium. Although phosvitin is directly concerned with the provision of calcium for the egg yolk (Taylor, 1970), a reduction in protein-bound calcium will reduce the amount of calcium available to the shell gland. However, Bitman et al. (1969) reported that neither op'- nor pp'-DDT caused a reduction in the serum calcium of laying quail on calcium-deficient diets. Further studies are needed to decide whether pesticides have significant effects on boneresorptionorcalcium transport in the blood. DSW 031309 STLCOPCB4015271 108 A. S. COOKE 1(e) Transport from the blood to the shell: The ciliated epithelial cells of the shell gland are thought to be able to transport calcium ions from the blood to the shell gland lumen (Hohman & Schraer, 1966). These workers studied the distribution of labelled calcium amongst cell organelles and their results indicated that calcium ions are stored in mitochondria and are transported via endoplasmic reticulum. More recently, Corradino et al. (1968) demonstrated that vitamin D3 increased calcium-binding by the shell gland of laying chickens and these workers isolated a calcium-binding protein that was electrophoretically indistinguishable from the protein involved in calcium transport in chicks' intestines (Wasserman & Taylor, 1966). Since this protein is capable of removing calcium from mitochondria (Hamil ton & Holdsworlh, 1969), it is likely to be implicated in shell formation. Just as with absorption mechanisms in the gut, it has been proposed that DDT may reduce calcium transport by affecting hydroxylation of vitamin D or by inhibiting ATPases concerned with active transport. As Risebrough et al. (1970) point out, in the Anacapa pelicans (a) enzyme inhibition is more likely to be the shell thinning mechanism than increased breakdown of hormones because of lack of a rectifying feedback at low pollutant concentrations (see objection (iv) to the decreased deposition of medullary bone theory); and (b) the site of action is probably in the shell gland since apparently healthy pelicans laid eggs with soft shells (Risebrough et al., 1971). There is some evidence that organochlorines can selectively reduce calcium transport to the shell gland lumen. Bitman et al. (1969) reported that quail on a low calcium diet laid eggs with thinner shells and with a lower `percentage of shell calcium' when fed op'- or pp'-DDT. `Percentage shell calcium' was the weight of calcium in the shell (x 100) divided by the weight of the fresh egg, and, since pesticides also decreased egg weight, interpretation of the results of these workers in terms of percentage of calcium in the shell is rather difficult. However, it can be inferred that pesticides decreased the weight of calcium per unit area of shell. More recently, Cecil et al. (1971a) reported that both pp'-DDE and pp'-DDT in a diet containing adequate calcium decreased `percentage shell calcium', but had no effect on shell thickness. No details of egg weights were given. Comparing the results from the two experiments, the effect of the low calcium diet on `percentage shell calcium' was between three and six times greater than that of the pesticide treatments. Longcore et at. (1971a) reported a decrease in calcium content from 35-2 to 340% in the shells of mallards that had been fed on a diet containing 10 ppm DDE, a diet known to cause a significant drop in shell thickness (Heath et al., 1969). Longcore et al. (1971a) also observed a significant increase in magnesium content for the shells of both the mallards and black duck Anas rnbripes (fed up to 30 ppm pp'-DDE), as well as a reduction in the strontium content of the black duck shells. A shift in the proportions of cations present in the shell may point to the partial inhibition of the passage of some metal cations from the blood to the shell gland lumen, so allowing other metals to take their places in the crystal lattice of DSW 031310 STLCOPCB4015272 SMELL THINNING IN AVIAN EGGS HY ENVIRONMENTAL POLLUTANTS 109 the developing shell. ]t is, however, possible that this simply reflects a minor disturbance in calcium metabolism outside the shell gland, and might indeed be a consequence of reduced deposition of medullary bone. Snetsinger et al. (1966) reported that increasing dietary magnesium or strontium resulted in higher levels of these elements in the shell. The fact that Drori et al. (1964) reported that the barrier between the blood and the developing shell has a Sr89/Ca45 discrimination factor of 0-93 led Simkiss (1968) and Simkiss & Taylor (1971) to doubt the existence of a specific calcium pump in the shell gland. In support of a partial blockage of calcium ion movement from the blood to the shell, a reduction was noted by Peakall & Lincer (1970h) in the active transport of calcium ions across sections of shell gland (not magnum; Peakall, personal communication) removed from the oviducts of ring doves after the birds had been given either PCBs orally or DDE by injection. Later experiments with PCBs were inconclusive, but reduced transport with DDE was again observed (Peakall, personal communication). However, Stephen et al. (1970) reported that 20 ppm pp'-DDT + 20 ppm pp'DDE in the diets of chickens did not significantly affect the percentage of egg shell calcium (compared with egg weight). Similarly, Davison & Sell (1972) found no change in shell thickness or shell calcium (compared with shell weight) when chickens were exposed to dietary levels of 20 ppm dieldrin or 200 ppm DDT. Thus, although small changes in egg shell calcium have been observed and DDE has been shown to reduce calcium transport, changes have only been slight, and available evidence suggests that inhibition of calcium transport from the blood to the shell gland lumen is not the major mechanism by which organochlorines cause thin shells. 2. Reduction in carbonate availability The origin of shell carbonate has received the attention of many workers and has recently been reviewed by Simkiss (1968). In some suggested mechanisms, it has been proposed that blood bicarbonate is transported either to the shell gland epithelium (Gutowska& Mitchell, 1945), or directly to the sites of shell development (Robinson & King, 1963). However, Hodges & Lorcher (1967) demonstrated that very little labelled bicarbonate injected into the blood becomes incorporated into the shell, and the theory of Simkiss (1961) that the source of the shell carbonate fraction is metabolic C02 is now widely accepted. In all proposed schemes the enzyme carbonic anhydrase (CA) is thought to play an important part by main taining the equilibrium between C02 and carbonic acid, so ensuring a supply of anions for shell formation. Sulphanilamide added to a chicken's diet immediately reduces shell thickness (Scott et al., 1944) and since sulphanilamide is a powerful CA inhibitor (Mann & Keilin, 1940), Gutowska & Mitchell (1945) suggested that it produces thin shells by enzyme inhibition. These last workers reported that the activity of CA in the DSW 031311 * STLCOPCB4015273 1)0 A. S. COOKE shell gland epithelium is greater in good laying birds than in poor ones, but Mueller (1962) was unable to repeat this observation and suggested that sulphauilamide may reduce shell thickness because of its diuretic properties. However, Opel (1965) and Cooke (1968) have since demonstrated that diuresis does not cause thin shells. More recently, Mueller et al. (1969) showed that acetazolamide, another sulphonatnide drug known to inhibit CA, caused a significant drop in bicarbonate and C02 levels in the shell gland lumen, and CA is still generally believed to play an intimate part in shell formation. According to the current theory of carbonate ion production, protons are released by the dissociation of water molecules in shell gland tissue, so that observation of a fall in intracellular pH during egg shell formation (Simkiss, 1969) supports the theory. Birds fed acetazolamide do not exhibit this drop in pH (Simkiss, 1970). No correlation between shell strength and shell gland CA content was found by Heald et at. (1968), who concluded that supplies of enzyme are normally in excess. CA is found in the tubular glands of the shell gland (Diamantstein & Schluens, 1964) while calcium tends to be accumulated in the ciliated epithelial cells (Gay & Schraer, 1967; Schraer & Schraer, 1970), so it is likely that the two main ionic components of the shell are secreted by different sets of cells (Simkiss & Taylor, 1971). The site of action of CA is not necessarily restricted to shell gland tissue since it has been reported in lumen fluid (Bernstein & Schraer, 1969) and in the mammillary cores (Robinson & King, 1963, see 3(a)). Theories for organochlorines affecting the availability of the carbonate moiety of the shell have centred on possible CA inhibition. It is difficult to envisage how the activity of water soluble proteins can be inhibited by fat soluble organochlorines, since they would be expected to be concentrated in different parts of the cell. In the chicken, CA is found in the soluble fraction of shell gland homogenates (Bernstein et al., 1968) and organochlorines would be expected to be associated with intracellular membranes. This objection was considered by Risebrough et al. (1970) who pointed out that both pp'-DDE and pp'-DDT can bind with soluble proteins (Brunnert & Malsumura, 1969). Risebrough et al. (1970) also suggested that the enzyme might possibly be membrane-bound in sensitive species. Literature on whether DDT can inhibit CA is conflicting. While Torda & Wolff (1949) and Keller (1952) reported apparent inhibition, Anderson & March (1956) and Wistrand (quoted in Maren, 1967) found no inhibition. Risebrough et al. (1970) carried out a series of in vitro experiments and reported that dieldrin, Aroclor 1254 and pp'-DDE all inhibited CA activity. pp'-DDE was the most potent inhibitor, being nearly as powerful as sulphanilamide. These workers, however, considered inhibition might be due to non-specific adsorption and expressed some doubt as to whether inhibition could occur in vivo. Two experiments have been reported in which organochlorines have been admini stered to birds and then reductions in carbonic anhydrase activity have been found. Peakall (19706) injected ring doves with pp'-DDE within a day of the first egg of DSW 031312 STLCOPCB4015274 SMELL THINNING IN AVIAN EGGS BY ENV1RONMENTAL POLLUTANIS III the two-egg clutch being laid. The second eggs had thin shells (by 23%) and the birds were killed after laying, CA activity of the shell gland (not magnum; Peakall, personal communication) was reduced by 60%. Similar injections of dieldrin had no effect on shell thickness or CA activity (Peakall, 1970b). In the second experi ment, Bitman et a!. (1970) added pp'-DDE or pp'-DDT to low and normal calcium diets being fed to Japanese quail over a period of three months. Birds were killed with an egg in the shell gland and lower CA activity was found both in the blood (22-44%) and in the shell gland (16-19%). These workers recognised that their in vitro demonstration did not `preclude normal functioning of the CA enzymatic machinery in the intact tissue in vivo'. However, results for the two different calcium diets were not separated and it was not stated whether the eggs laid by the quail did, in fact, have thin shells. In similar experiments, Bitman et al. (1969) reported thinner shells after feeding a low calcium diet containing pp'-DDT, but Cecil el al. (1971n) reported no change with pp'-DDE or pp'-DDT in normal diets. Thus it is not clear whether the 16 to 19% decrease in CA activity in the quail shell glands, as recorded by Bitman et al. (1970), was associated with thin shells being laid. Indeed, it was pointed out by Dvorchik et al. (1971) that the degree of reduc tion in enzyme activity reported by Peakall (1970b) and Bitman et al. (1970) was probably insufficient to cause physiological inhibition. It is thought that under normal conditions CA activity in the shell gland exceeds the physiological require ment and is not rate limiting (Heald et al., 1968). In chickens, intramuscular injection of sufficient acetazolamide to reduce shell weight by 41 % caused CA activity in shell gland scrapings to be zero or `extremely low' (Bernstein et al., 1968). Thus, CA inhibition would not be expected to cause shell thinning after slight contamination with pesticide (Risebrough et a!., 1970) as has often been reported. When, however, chickens are exposed to different levels of sulphanilamide in the diet, there is a linear relationship between the decrease in shell thick ness and the logarithm of the dietary concentration of sulphanilamide (Scott et al., 1944), strikingly reminiscent of the relationship between shell thickness and DDE content of pelicans' eggs from dillerent colonies as observed by Blus et al. (1972). Dvorchik et al. (1971) criticised the experimental technique of Peakall (1970b) and Bitman et al. (1970), and gave an account of their own in vitro experiments, which suggested that DDT should not affect CA activity at concentrations found in tissues in the field. Inhibition was only noted when high DDT concentrations resulted in precipitation. Since then, Pocker et al. (1971) have demonstrated that pesticides are not true inhibitors of CA.. Apparent inhibition occurs in in vitro tests only when a pesticide precipitate forms and occludes CA from solution. Thus, inhibition is unlikely to be caused directly by pp'-DDT, pp'-DDE or dieldrin (Dvorchik et al., 1971; Pocker et al., 1971), but inhibition by metabolites of these compounds, or a reduction in enzyme synthesis (suggested by Kenny &. Dacke, personal communication) cannot be ruled out. Inhibition by a water soluble DSW 031313 STLCOPCB4015275 112 A. S. COOKE metabolite is an attractive theory, since interaction with the soluble enzyme (Bernstein et ah, 1968) then becomes more plausible. Before leaving CA inhibition, the `salt gland theory' of Risebrough et ah (1970) should be mentioned. CA is found in many places in the body including, in birds, the salt gland, and effects noted in the shell gland might possibly only be a conse quence of primary effects elsewhere. The early work on the avian salt gland was carried out by Schmidt-Nielsen and his co-workers in the late 1950s. They found that the salt (or nasal) glands of marine species excrete a highly concentrated salt solution, thereby ridding the bird exlrarenally of unwanted salt. Brown pelicans were injected intravenously with salt solution by Schmidt-Nielsen & Tange (1958) who found that the salt gland excreted about four times as much salt as the kidney. CA is essential for proper functioning of the salt gland and sulphonamides can prevent fluid excretion by the gland (see Maren, 1967). Since an excess of chloride in the diet causes chickens to lay eggs with thin shells (Hall & Hclbacka, 1959; Hunt & Aitken, 1962a, b; Lorcher et ah, 1964), Risebrough et ah (1970) suggested that pesticides may inhibit proper functioning of the salt gland in marine species, so reducing chloride excretion and causing a decrease in shell thickness. They also pointed out that this would explain why some species, such as the brown pelican, which largely rely on extrarenal means for chloride excretion, should be particularly affected. This is an interesting theory and affords yet another site of action for a CA inhibitor. It is, however, still only a tentative theory. One objection is that chloride salts in the diet do not consistently reduce shell thickness. Although Lorcher et ah (1964) reported shell thinning when sodium chloride and potassium chloride were added together to the rations of chickens, Hunt & Aitken (19626) reported that neither calcium nor potassium chloride reduced shell thickness. The latter writers considered that the common characteristics of salts causing reduced shell thickness were (a) an acidic anion, not readily removed from the blood, and (b) a readily-metabolisable cation. Thus, it is uncertain whether circulatory excess of sodium chlorine, the main chloride to which the marine species will be exposed, can cause significant shell thinning. It is well proven that an alteration in the acid-base balance of the blood can lead to changes in shell thickness. For instance, supplementing the diet of laying chickens with 1-3% ammonium chloride leads to metabolic acidosis, characterised by a decrease in blood pH and bicarbonate concentration and by the production of eggs with thin shells (Hall & Helbacka, 1959; Hunt & Aitken, 19626). That high environmental temperatures cause birds to lay eggs with thin shells has long been known. This effect is probably due to the birds hyperventilating, so that blood pC02 is decreased and the bicarbonate level drops (Mongin & Lacassagne, 1966; Simkiss, 1968). Such birds are in a state of respiratory alkalosis, blood pH being raised. The opposite, respiratory acidosis, can be achieved by exposing a bird to high levels of atmospheric C02. Helbacka et ah (1963) and Hunt & Simkiss (1967) reported that, after acute exposure, birds in a stale of respiratory acidosis laid eggs DSW 031314 STLCOPCB4015276 SMELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS ] 13 with thin shells, but after a longer study, Frank & Burger (1965) reported that thicker shelled eggs were laid. These discrepancies may be due to renal compensa tion, leading to an increase in blood bicarbonate and carbonate levels during chronic exposure (Simkiss, 1968). Studies on the effect of organochlorines on blood pH, pC02, etc. would seem to be particularly worthwhile. 3. Effects on other shell constituents Most workers studying mechanisms by which pesticides cause thin shells have devoted their attentions to the laying birds' tissues. The thin shells, apart from having their thickness measured, have apparently been discarded. There have, however, been two scanning electron microscope studies of thin shells caused by pesticides. Erben & Krampitz (1971) examined the fine structure of two very thin shells collected from the brown pelican colony on Anacapa Island. They reported that the palisade layer decreased in thickness and the number of globular inclusions in this layer increased. Pelican shells have a calcareous cover (Tyler, 1969; see Fig. 2) and Erben & Krampitz (1971) noted that in the cover of the thin shells were numerous abnormal granules. Although the shells were much thinner than a control shell (0-20 and 0-36 mm compared with 0 59 mm), there was apparently no loss of matrix protein. There were, however, marked differences in the amino acid patterns of both the membranes and the matrix. In the second investigation, quail were dosed with technical DDT by McFarland et at. (1971). Thin shells laid by the quail showed changes in mineral deposition and porosity in the mammillary layer, but the overall decreases in thickness could usually be accounted for by decreases in the palisade layer. One shell that was 40 % thinner than normal had a palisade layer reduced in thickness by 60%. The surface crystal layer changed considerably, sometimes being thicker and sometimes being virtually absent. Shells often had a rippled surface with enlarged pore channels. Such gross defects in shell structure suggest that changes in other shell constituents, apart from the calcium and carbonate moieties, may be markedly impairing normal shell formation. Factors to be considered are (a) disruption of the sites of initiation of crystal growth, (b) deleterious effects on shell constituents essential for proper shell formation, (c) enhancement of the action of constituents responsible for inhibition of shell growth, and (d) premature induction of the shell constituents or factors controlling termination of shell growth. 3(n) Disruption of initiation sites: The presence of CA in the mammillary cores, the initiation sites, was reported by Robinson & King (1963). Although this has since been refuted by Diamantstein (1966), it is worth remembering that if pesticides do reduce the activity of this enzyme, the effect is not necessarily restricted to shell gland tissue (or the salt gland). Calcification is initiated by the deposition of calcite on preformed organic mammillary cores (Fujii &Tamura, 1970). In a collection of chicken shells exam ined by Robinson & King (1970), while the cores in shells of average strength DSW 031315 114 A. S. COOKE were always found to be normal, more than half of the weak and thin shells had abnormal mammillary cores. The abnormally was characterised by the organic cores being irregular and diffuse, and sometimes small dense organic particles were present on the membrane surface. Shells with surface concretions were found to be completely devoid of organic mammillary cores beneath the concretions. Overlying the membrane fibres in such shells were aggregations of organic material, considered by Robinson & King (1970) to be an extreme form of the mammillary core abnormality. Thus, here is an association between weak, thin and malformed shells and abnormal mammillary cores. The organic cores are probably composed of modified shell membrane material (Robinson & King, 1968; Robinson et al., 1968). There are modifications in structure (Simons & Wiertz, 1963) and in carbohydrate content (Cooke & Balch, 1970a), but no marked differences in amino acid content have so far been found (Cooke & Balch, 1970a). Erben & Krampitz (1971) reported a change in the amino acid content of shell membranes on the thin, pelican shells from Anacapa. If this is indicative of a similar change in amino acids in the mammillary cores, then initiation of crystallisation could be affected. This is only speculation, but in the shells of quail treated with DDT by McFarland et al. (1971), there were gaps between normal basal caps occupied by calcile spherulites about lO/i in diameter. These may have been sites where crystallisation failed soon after initiation. The densely staining particles reported by Robinson & King (1970) on the membranes of chicken shells with abnormal cores may have been the organic material in such failed centres that was left attached to membrane fibres after the preparations were decalcified. These particles were up to 15p in diameter. The calcite spherulites are unlikely to have been caused by a calcium deficiency, since thin shells from chickens on a low calcium diet have structurally normal mammillary layers (Cooke, 1968). Irregularities have been observed in the mammillary layers of shells laid by chickens exposed to high temperatures (El-Boushy et al., 1968). In the shells examined by Erben & Krampitz (1971) and McFarland el al. (1971), however, the palisade layer seems to have been affected more than the mammillary layer, indicating less disruption during initiation of shell growth than during later stages of calcification. 3(b) Deleterious effects on shell matrix: Nothing is known about essential in organic constituents (apart from calcium carbonate) so this discussion is limited to possible effects on shell matrix. Since no researchers into pesticide effects on egg shells have so far discussed the importance of matrix, the subject is reviewed in detail here. The matrix is a continuous organic phase found throughout the calcified shell and was included in an electron microscope study of shell components by Simons & Wiertz (1963). It is mainly composed of protein (Baker & Balch, 1962), with 10-11 % carbohydrate (dry weight; Baker & Balch, 1962; Cooke, 1968) and fat has been detected histochemically (Simkiss & Tyler, 1957). Analytical studies DSW 031316 STLCOPCB4015278 SMELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS ] 15 (Baker & Balch, 1962; Cooke, 1968; Cooke & Balch, 1970ft) reveal that the muco polysaccharide moieties bound to the protein are chondroitin sulphates A and B (see Brimacombe & Webber, 1964) plus at least two neutral polysaccharides. Degradative acid hydrolysis (Cooke, 1968) indicates that the neutral polysaccharides have the typical structure (Dische, 1965) of a backbone of hexosamine and hexose units with short side chains terminating in fucose or sialic acid. Matrices front thin shells have already been examined in detail (Cooke, 1968). The treatment regimes of one experiment involving nine chickens are shown in Table 2. Treatments selected to cause thin shells were high temperature (34C); TABLE 2 TREATMENT REGIMES USED IN AN EXPERIMENT TO STUDY THE ORGANIC MATTER IN THICK AND THIN shells laid by chickens Galltts domestkus (COOKE, 1968) Control Room temperature treatment (13-I9C) Recovery High temperature treatment (34 C) Recovery 3 birds 3 birds 3 birds 2 weeks B.D. B.D. B.D. 3 weeks B.D. Supplement, 0 03% suiphanilainide Supplement, 2% ammonium chloride 3 weeks B.D, B.D. B.D. 3 weeks B.D. Supplement, 0 03% sulphanilamide Supplement, 2% ammonium chloride 3 weeks B.D. B.D. B.D. B.D. - basal diet. 0 03 % sulphanilantide in the diet at room temperature (13-19C) or high tempera ture; and 2% ammonium chloride in the diet at room or high temperature. After removal of the membranes and the cuticle, shell thickness was determined (mg/cm2) and bulked samples of shells from the last two weeks of each period were decalcified. The matrix suspension was centrifuged. Then the matrix was dried, weighed and analysed for sialic acid and uronic acid, which were chosen because they contain acidic groups and are constituent units of matrix mucopolysaccharides. These are believed to be involved in stabilising the matrix (Baker, 1959; Simkiss & Tyler, 1959) and/or in binding calcium to the developing shell lattice either ionically (Baker, 1959; Cooke & Balch, 1970ft) or by chelation (Simkiss & Tyler, 1958, 1959). The experiment yielded thirty samples from birds on a control diet and fourteen samples of abnormal thin shells laid under the various treatment conditions (there should have been fifteen, but one bird on the ammonium chloride treatment at room temperature ceased laying for three weeks). Birds on a control diet will, of course, produce shells with a wide range of thicknesses, and to find out whether differences exist between the organic matter in `normal thick' and `normal thin' shells, the bulked samples from control and recovery periods have been divided into two groups: (i) `normal thick' shells, samples with mean shell thickness >70mg/cm2 ( = 16); (ii)`normal thin'shells, samples with mean shell thickness < 70mg/cm2(u = 14) DSW 031317 STLCOPCB4015279 116 A. S. COOKE Results are shown in Table 3. Jn these and in other normal shell samples, the amount of matrix present was significantly correlated with shell thickness (r = 0-560, d.f. = 37, P < 0 001; Cooke, 1968). In Table 3, there is significantly less matrix (mg/cm3) in the `normal thin' shells than in the `normal thick' shells and there is a tendency for thinner shells to have a lower matrix content (mg/g). Although the difi'erence in thickness between the `abnormal thin' shells and the `normal thin' shells shown in Table 3 is twice that between the two normal groups, the `abnormal thin' shells do, in fact, contain as much matrix (mg/cm2) as the `normal thin' shells. Because of this, their matrix content (mg/g) is significantly higher. The two types of thin shells contained significantly less sialic acid (/ig/m2), but their matrix sialic acid content (pg/mg) is similar to that of `normal thick' shell matrix. The matrix in `abnormal thin' shells is, however, considerably deficient in uronic acid (pg/mg) and, furthermore, a shift in the ratio of the constituent uronic acids has often been found (Cooke, 1968). `Abnormal thin' shell matrix is relatively richer in glucuronic acid (present in chondroilin sulphate A) and poorer in iduronic acid (present in chondroitin sulphate B). In order to make a convenient comparison of matrix composition, all the `abnormal thin' shells have been grouped together. There were, in fact, some differences in shell matrix composition from the various treatments, but an increase in matrix content and a decrease in uronic acid were consistently found for these `abnormal thin' shell samples, irrespective of treatment conditions (Cooke, 1968). Similarly, Longstaff & Hill (1972) recently reported that in thin-shelled eggs laid by pullets on manganese deficient diets, shell matrix content increased and the uronic acid content of the matrix decreased. Nevertheless, this cannot be taken as a general rule for the matrices of all `abnormal thin' shells, since a bird main tained on a strontium-rich diet laid eggs with thin shells, but they contained less matrix (mg/g) and the matrix contained unusually high amounts of uronic acid (Cooke, 1968). The shift towards glucuronic acid was, however, still evident. Frank et a!. (1965) reported that in a group of twenty-nine shells the weaker shells contained a higher percentage of protein. Deciding the relevance of changes in matrix composition in thin shells is rendered considerably more difficult by the fact that matrix distribution throughout the depth of a normal shell is not uniform (Cooke & Balch, 19706). Thus it is not possible to draw up a simple formula for calculating the matrix requirement for proper shell formation. The observed changes in the matrix of `abnormal thin' shells are not, of course, necessarily the cause of shell thinning. The usually-accepted theories for explaining why high temperature, ammonium chloride or sulphanilamide treatment cause thin shells revolve around reducing the availability of the carbonate fraction. These treatments all produce significant decreases in bicarbonate and C02 levels in shell gland lumen fluid (Mueller et al., 1969), but the reasons for shell thinning are liable to be more complex than simple reductions in the available carbonate. Sulphonamides, for instance, depress calcium retention from the food (Tyler, 1950), OSW 031318 STLCOPCB4015280 1(8 A. S. COOKE reduce blood calcium levels (Siegmund & Dulce, 1960) and can reduce egg size (Cooke, 1970u). High temperature reduces food consumption and, therefore, calcium intake, but retention is increased (Mueller, 1959). Despite this, serum calcium levels are reduced (Mueller, 1959) and incorporation into the shell of labelled calcium, given orally, is significantly decreased at elevated temperatures (Bragg cl al., 1971), indicating impairment of calcium metabolism. Ammonium chloride may reduce blood calcium (Siegmund & Dulce, 1960). Questions that need answering are: (a) how essential is the matrix for proper shell formation and (b) can subtle changes in composition or structure contribute to shell thinning after a bird has been exposed to sulphonantidcs, ammonium chloride, high temperature or pesticides? Electron micrographs of developing shells taken by Simons (1971) clearly demonstrate that deposition of organic matter precedes crystallisation. Like egg shell matrix, molluscan shell matrix is largely protein in nature with smaller amounts of neutral and acidic mucopolysaccharides and traces of fat (Wilbur & Simkiss, 1968). Mollusc shell matrices have the ability to exert some degree of control over which polymorph of calcium carbonate is formed (Watabe & Wilbur, 1960; Simkiss, 1965). These writers found that a matrix, removed by decalcification from a calcite mollusc shell and placed in a solution that normally precipitated calcium carbonate as aragonite, often caused calcite to be deposited instead. Wilbur & Watabe (1967) concluded that molluscan matrix probably `provides a substratum on which crystals nucleate, orient and grow'. Since the matrix is apparently vital for proper shell formation, deficiencies or changes in matrix composition might indeed result in crystal disruption or termination in both mollusc shells and avian egg shells. Whilst on the subject of molluscan shell deposition, E. Pollard and I recently exposed snails, Helix pomatia, to pp'-DDT during their first growing season. Preliminary analysis of the results indicates a significant retardation of shell growth at exposure levels that had no effect on body growth. The structure of thin shells caused by subjecting chickens to high temperatures was examined by El-Boushy el a!. (1968), who reported that a rise in temperature from 13 to 29C resulted in normal columnar crystal orientation being severely disrupted, less secure attachment of the membranes to the mammillary layer and cavities in the palisade layer. Electron micrographs of decalcified sections showed that the vesicles in the matrix in the palisade layer were abnormally large and cavities in the matrix were also recorded. The organic matrix showed physical deformations corresponding to those in the mineralised shell (Wiertz & Simons, personal communication), mineral matter disruption apparently being caused by an abnormal matrix. There are similarities between the high temperature thin shells examined by El-Boushy et al. (1968) and the DDT-induced thin shells examined by McFarland el al. (1971), who reported defects where the membrane fibres joined the shell and cavities penetrating the palisade layer. Crystal structure was not so disrupted in SW 031319 STLCOPCB4015281 TABLE 3 THE AMOUNT OF MATRIX, SIALIC ACID AND URONIC ACID IN THICK AND THIN SHELLS LAID BY CHICKENS Gallus domeSticUS. FOR DETAILS SEE TEXT (ADAPTED FROM COOKE, 1968) No. of True shell bulked thickness samples* (mg!cm1) Matrix (mg!cm1) (mgjg shell) Sialic acid (ftg/cm1) (pglrng matrix) Uronic acid (pgicm2) (pgjmg matrix) `Normal thick' shells, thickness > 70 mg/cm 2 `Normal thin' shells, thickness < 70 mg/cmt `Abnormal thin' shells 16 73-7 0-5 1 09 0 05a 14-8 0*5a 9-3 0-2a 14 66.2 0-5 0-90 0-04= 13-5 0'5a 81 i 0-2d 14 52-7 1-3 0-90 004c 17-2 l-0i 7-7 i 0-3d 8-6 0*3a 9-2 0-3a 8-7 0-3a 14-4 0-3a 12-5 0-6c 10-6 0-4d 13-3 0-4a 13-6 i 0-3a 11 -9 0-48J S.E. * Samples contained segments from 3-13 shells a- b- c- d For a pair of means differing by one letter, P < 0-05; by two letters, P < 0-01: by three letters, P < 0 001. J Compared with `normal thin' shells. P < 0 01 /DSW 0 3 1 3 2 0 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 119 the slid Is invesligated by Erben & Krampitz (1971), but ail abnormal number of globular inclusions were noted in the palisade layer. In addition, Erben & Krampitz (1971) found a threefold increase in organic matter content of the shell, expressed in /unoles of amino acid/lOOmg shell. Nothing was stated about the amount of matrix per shell, although Erben and Krampitz's figures suggest that little, if any, was missing compared with normal shells. At this point it is worthwhile remember ing that the `abnormal thin' shells (including the high temperature shells) shown in Table 3 contained as much matrix per shell as the `normal thin' shells although they were, on average, more than 20% thinner. Changes in matrix composition have been noted for both pesticide-induced and high temperature thin shells, Erben & Krampitz (1971) reporting amino acid changes in the former and Cooke (1968) uronic acid changes in the latter. Thus, if shell disruption in high temperature thin shells is caused by changes in the physical and chemical structure of the matrix, as has been proposed, shell disruption in pesticide-induced thin shells may be similarly caused. Having reached this tentative conclusion, it should be stressed that matrix disruption is only part of the story. The chemically and physically changed matrix can only affect the deposition of those lattice ions that are available in the lumen during shell formation, so that, if a blockage in either calcium or carbonate ions occurs, the shell will be thin. If the matrix is abnormal, then the mineralised shell is also likely to have an abnormal structure, but can essential shell organic material be so changed that, even though sufficient mineral ions are present in the lumen, the shell is thin, due to low' incorporation of minerals into the shell? Since high temperature, sulphonamides and ammonium chloride, which all produce changes in matrix composition (Cooke, 1968), also all lower the bicarbonate concentration of shell gland lumen fluid, and since this parameter is known to be significantly correlated with shell thickness (Mueller el ah, 1969), organic matter disruption is most unlikely to be the primary cause of thin shells. Longstafif & Hill (1972) reached a similar conclusion after studying shells produced by manganese-deficient birds. However, an enhancement of shell thinning due to changes in organic matter cannot be ruled out, particularly since Robinson & King (1970) reported an associa tion between physical and histochemical abnormalities in mammillary cores and thin, weak shells. Matrix may influence shell strength. Simons (1971) examined tw'O of the shells studied by Robinson & King (1970) and found that, in a shell that was unusually weak for its thickness, the number of vesicles in the matrix was greatly increased. Treatment of shells with sodium sulphide solution at 95C does not affect the mineral matter, but shells become weaker because of destruction of matrix (Simons el ah, 1966). Frank el ah (1965) were, however, unable to detect significant changes in the amino acid patterns of matrices in abnormally weak or strong shells. Why matrix composition should be changed by certain treatments is not known. It may be essentially a secondary effect due to reduced incorporation of minerals DSW 031321 f y STLCOPCB4015283 120 A. s. cooki: into the shell or there may be some direct effect on the matrix. A dietary lack of manganese is thought to have a direct effect on glucosamine metabolism (see Longstaff & Hill, 1971). Glucosamine, a hexosamine, is present in egg shell mem branes, mammillary cores, matrix and cuticle (Baker & Balch, 1962; Cooke & Balch, 1970 a, b). A deficiency of manganese in the diets of pullets just beginning to lay leads to a reduction in the glucosamine content of the isthmus and shell gland and can also cause the production of weak egg shells (Longstaff & Hill, 1971). Matrix in these shells contains decreased amounts of hexosamine and uronic acid, apparently because of direct action on mucopolysaccharide metabolism (Longstaff& Hill, 1972). In the shell gland both the epithelial cells and the tubular glands secrete acid mucin fractions, the more strongly acidic mucins being in the epithelial cells (Robinson et al., 1968), Some shifts in the matrix composition in thin shells might be explained by relative changes in activity in the two types of secretory tissue. From the observed changes in matrix composition in `abnormal thin' shells (Cooke, 1968), it would appear that the epithelial cells, which are thought to be involved in secreting calcium ions (Gay & Schraer, 1967; Schraer & Schraer, 1970) are affected more than the tubular glands. 3(c) Enhancement of shell growth inhibitors: The results obtained by Brooks & Hale (1955), Itoh & Hatano (1964) and Snetsinger et al. (1966) show that, in shells produced by chickens, the higher levels of phosphorus, magnesium and strontium occur towards the outside, i.e. in those parts of the shell that are the last to be formed. Simkiss (1964) reported that phosphate ions can inhibit calcite crystal growth, and Sobel & Hanok(1952) observed that strontium ions inhibit the calcifica tion of rachitic bone in vitro, while magnesium ions enhance this effect. These ions, therefore, may play some part in the inhibition of shell growth. The only experiment designed to study changes in shell mineral constituents other than calcium after administration of organochlorines is that of Longcore et al. (1971a). In shells laid by ducks being fed a diet containing DDE, there was a decrease in strontium content and an increase in magnesium, but no precise data were given for phos phorus. Thus there is no evidence to suggest that the effect of these known inhibitors may be enhanced in shells from treated birds. 3(d) Premature termination: Since shell growth is normally terminated quite suddenly in the chicken oviduct (Cooke, 1968), it is unlikely that increasing amounts of the inhibitors significantly contribute to the process of termination. The concentration of calcium ions in the shell gland lumen falls during the last 2-4 h of calcification, but the level when mineralisation ceases is still higher than during the early stages of calcification (Mongin & Saveur, 1970). Termination may be brought about by secretion of cuticular material, which, if secreted quickly, could blanket the shell and prevent inorganic ions reaching the lattice. The basic amino acids, arginine and lysine, are present in high concentrations in the cuticle (Baker & Balch, 1962) and may compete with calcium ions for negative sites on the matrix during cuticle secretion. Thus, premature secretion of the cuticle might DSN 031322 STLCOPCB4015284 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 121 produce shells with the outer layers missing. Gross analysis of shells laid by hens treated with sulphanilamide at room temperature suggests they may, indeed, be lacking (he outer layers (Cooke, 1968). Whether one is justified in comparing thin shells with the inner levels of normal, naturally-laid shells is doubtful. Although (a) matrix content is non-uniform, increasing from the inside of the shell, then decreasing over the outer third of the shell's thickness (Cooke & Balch, 19706) and (b) Robinson et al. (1968) noted a decrease in staining for acid mucins in shell gland tissue during the final stages of calcification, nevertheless, the rate of incorpor ation of organic material into the shell seems to be linearly related to shell growth (Cooke, 1968; Longstaff& Hill, 1972). These apparently contradictory observations can largely be reconciled by assuming that incorporation of fresh matrix into the shell occurs not only on the developing surface but also in channels radiating out from the pores. A matrix fraction readily attacked by alkali is thought to occur in such a location (Tyler & Simkiss, 1958). This means that a shell developing in the oviduct might contain less organic matter than the corresponding inner layers of a completed shell. The thin shells structurally investigated by McFarland et al. (1971) certainly do not seem to have been caused by premature termination, since the surface crystal layer just below the cuticle (see Fig. 2) was sometimes of increased thickness in thin shells. A pelican shell does not have a surface crystal layer (Tyler, 1969), and it is not possible to rule out premature termination as the cause of the thin shells examined by Erben & Krampitz (1971). Indeed, if premature termination by cuticular secretion produced these thin shells, the deficient palisade layer, reported by Erben & Krampitz (1971), would be expected. Moreover, these workers also observed irregular calcareous granules in the cover (see Fig. 2) which might be shell material that would have normally been deposited before (and not after) the cuticle. On the surface of some chicken shells, Tyler & Simkiss (1959) have found abnormal calcified globules that apparently contained normal shell matrix. On the other hand, aberrant spheres of shell material were also occasionally noted on the surface of thin shells examined by McFarland et al. (1971) and these shells were apparently not caused by premature termination. Pigment is present in the outer shell layers and in the cuticle, and lack of pigment could be indicative of premature termination, premature oviposition or reduced incorporation of pigment into the developing shell. Ratcliffe (1970) pointed out that since 1947 peregrine shells have become less richly marked, and, similarly, Finireite (1971) noted that pheasants fed on organomercury-treated grain often laid eggs with abnormally light-coloured shells. Premature oviposition is unlikely to be involved in shell thinning (see below) so these poorly pigmented shells are likely to have been caused by premature termination, by a reduction in available pigment or by reduced binding of pigment to shell components. Poorly pigmented shells are also produced by other treatments that cause thin shells, such as sulphanil amide (Scott et al., 1944). DSW 031323 7 STLCOPCB4015285 122 A. S. COOKK 4. Changes in other factors involved in egg shed formation 4(a) Premature oviposition: The possibilities of neural or hormonal control of oviposition have been discussed by Sturkie (1965) and Gilbert (1967, 1971 h). Eggs laid prematurely will lack the outer levels of shell and will also be without a cuticle. Chickens in a state of shock will sometimes lay prematurely. Premature oviposition has been tentatively suggested as the possible cause of thin shells induced by pesticides (e.g. Bitman et al., 1969; Ratcliffe, 1970), but the evidence now points towards this being very improbable. In the pesticidc-induccd thin shells that have been studied (Erbeu & Krampitz, 1971; McFarland et a!., 1971), the shell components deposited during the final stages of formation, the cuticle and the cover, were present. Sulphanilamide, ammonium chloride or high temperature conditions are also unlikely to act in this manner, since none of these treatments affects either the time the egg remains in the shell gland or the amount of cuticular material deposited (Cooke, 1968). Similarly, cuticles are present on thin-shelled eggs laid by pullets on manganese-deficient diets (Longstaff & Hill, 1972). 4(6) Decreased food consumption: It is quite feasible that the residues in the laying bird could (a) have a detrimental effect on the bird's hunting ability (as suggested by Ratclilfe, 1970), perhaps by impairing vision (suspected in pigeons sublethally poisoned with endrin; Revzin, 1966) or (b) cause a loss of appetite. In this way pesticides might have an indirect effect on shell thickness via reduced food intake. Administration in capsules of either dieldrin to pheasants (Atkins & Linder, 1967) or pp'-DDT to pigeons (Jefferies & French, 1971) resulted in loss of appetite. Jefferies & French (1971) reported significantly lower hepatic levels of vitamin A in their birds and suggested that a reduction in vitamin A might be responsible for appetite loss (see Moore, 1967). Against a theory of decreased food consumption, laboratory birds have consumed, at a normal rate over periods of more than one month, diets containing 250 500 ppm DDT (Linduska & Springer, 1951; GeneJly & Rudd, 1956; Noakes & Benfield, 1965; Davison & Sell, 1972; Lillie et al., 1972). 4(c) Thyroid effects: Thyroidectomy in chickens causes severe effects on growih and reproduction (see Sturkie, 1965), so it is perhaps not surprising to find that it also causes shell thinning. Taylor & Burmester (1940) reported a 9% reduction in percentage shell after the operation, although they expressed concern that uninten tional removal of the parathyroids may have been at least partially responsible for the effect. Other reproductive changes were a 3 % reduction in egg weight and a decrease in egg production of about 70%. Egg production and egg weight wrere again decreased by the addition of 01 %thiouracil to the diet (Berg & Bearse, 1951), shell thickness being either unchanged (Berg & Bearse, 1951) or reduced (Gahuten & Shaffner, 1954). Thiouracil causes a bird to become hypothyroid by blocking thyroxine synthesis (see Sturkie, 1965). In thyroidectomised drakes, the rate of induction of medullary bone by oestrogen was retarded (Benoit & Clavert, 1947). OSW 031324 STLCOPCB4015286 SHELL THINNING IN AVIAN EGGS nV ENVIRONMENTAL POLLUTANTS 12.1 This bone was found to be osteoporotic, attempts to repeat this observation using thyroid inhibitors failed (Taylor & Stringer, 1965). The oestrogen-induced rise in blood calcium can be inhibited by both thyroxine and thiouracil (Hohn, 1961). Jefferies & French (1971) and Jefferies et at. (1971) observed that pigeons became hyperthyroid after being fed low doses of DDT in the diet, while those fed higher doses were hypothyroid. This led these workers to suggest that in the field birds of prey laying thin-shelled eggs were hypothyroid. Loss of colloid from the follicles of the thyroid, indicating either hyper- or hypothyroidism, was also noted after pigeons were treated with pp'-DDE or dieldrin (Jefferies & French, 1972). Richert & Prahlad (1972) reported that histological sections of thyroids taken from Japanese quail previously treated with DDE or DDT indicated hypo functioning of the glands. In addition, DDE doubled the weight of the thyroids and halved iodine uptake per unit weight (the absolute uptake of iodine therefore apparently being unaffected). DDA had no significant effect on the gland. Hyperthyroidism resulting from eating a diet containing iodinated casein or thyroprotein caused chickens to lay eggs with thicker shells (Berg & Bearse, 1951; Gabuten & Shaffncr, 1954). In a similar experiment, Wilson (1949) also reported that thicker shells were sometimes laid, but since egg weight decreased, shell weight was probably unaffected. Asmundson & Pinsky (1935) found that hyper thyroidism caused a slight decrease in egg weight and a slight increase in shell weight. Conditions used by these workers were sufficient to reduce or terminate egg laying. Bengalese finches subjected to DDT treatment by Jefferies (1969) were apparently in a hyperthyroid state. Egg weight was reduced, but shell weight was unchanged, so that the shells were thicker. To explain hyper- and hypothyroidism in their pigeons, Jefferies et al. (1971) suggested that pesticides might have tw'O effects, stimulation of the thyroid, w'hich is probably mediated via the pituitary, and competition with thyroxine for binding sites on the transport proteins in the serum. There is evidence that op'-TDE can lower serum protein-bound iodine in humans by competing for sites on thyroxine binding globulin (Marshall & Tomkins, 1968) although op'-TDE administration did not cause hypothyroid symptoms and the effect could not be demonstrated in vitro. The last fact may indicate that a metabolite was the active compound rather than op'-TDE itself. In birds treated with DDT, the thyroid gland is considered by Jefferies (personal communication) to be permanently hyperactive, but the metabolic activity of the bird, as reflected by pulse rate, body temperature, etc., might be indicative of hypo- or hyperthyroidism, depending on whether competition for binding sites overrides stimulation. Variations in response between mammals and birds, and between different species of birds could be due to differences in serum binding proteins (Jefferies et a!., 1971). Certainly, thyroxine is bound more strongly in human serum than in duck or chicken serum (Sturkie, 1965). Jefferies el al. (1971) have suggested that a thyroid defect with an accompanying change in vitamin A metabolism might be responsible for many of the observed OSk 031325 STLCOPCB4015287 124 A. S. COOKE sublet hal effects of organochlorines, e.g. changes in liver and gonad weight. Can hypothyroidism, however, explain the thin shells observed in the field and in the laboratory? Effects on egg shells in the experiments quoted above have been slight, if indeed they have occurred at all, and other reproductive effects, such as reduced egg production, were often much more marked, It should be remembered, however, that most of these tests were carried out on chickens, a species that is probably more resistant to shell thickness changes than those species affected in the field (ice the previous section). Although, with caged pigeons, hypothyroid birds cannot be distinguished from normal birds by their external appearance (Jefferies, personal communication), if the pelicans on Anacapa were in such a hypothyroid condition as to reduce egg shell thickness by 50%, other changes in reproductive processes or in behaviour might have been expected. On the island in 1969, a minimum of 1272 nests were built and eggs were probably laid in at least 75% of these nests, but shells collapsed during incubation and not more than four chicks were raised (Risebrough et al., 1971). Nest-building, mating, ovulation and brooding were apparently normal, but shell thickness was, on average, halved. The birds might have been in a hypothyroid state but it seems likely that thyroid malfunction reflects a general pesticide effect on the bird, particularly if the activity of the pituitary is altered (see section 1(6)) rather than the binding of pesticide on to thyroxine transport proteins being the primary lesion responsible for egg shell thinning. Possible effects of hypothyroidism on reproduction have concerned poultry keepers for many years, since a decreased thyroid secretion rate has been suggested as a possible cause of the usual decline in egg production and shell thickness during the summer months (Thornton & Moreng, 1959; Tyler & Geake, 1960; Sturkie, 1965). This is of particular interest to the present discussion, since it offers a further connection between the effects of pesticides and the effects of high temperature on shell formation. Attempts to increase egg production in chickens by feeding thyroprotein or iodinated casein have met with some (Turner el al., 1945) or no success (Wilson, 1949). The last writer also reported that iodinated casein failed to correct shell weight at high temperatures. Vitamin C has also been proposed as a dietary additive to correct shell thinning due to high temperature, but attempts to achieve corrective shell thickening with (his vitamin have usually been unsuccessful (see El-Boushy et a/., 1968). 4(d) Adrenal effects: The adrenals of pigeons dosed with pp'-DDT increased in weight (Jefferies et al., 1971) and these writers suggested that the thyroid may directly control such changes, since both hypothyroidism and marked hyper thyroidism cause adrenal hypertrophy in the chicken (see Sturkie, 1965). pp'-DDE, but not dieldrin, causes similar increases in adrenal weight (Jefferies & French, 1972). Srebocan et al. (1971) reported reduced concentrations of corticosteroids in the adrenals and in the plasma after young cockerels had been treated with technical DDT or pp'-DDT. DSW 031326 STLCOPCB4015288 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 125 Administration of adrenal corticosteroids to a laying bird can affect shell deposition. Decreases in shell weight after cortisone treatment have been noted by Urist & Deutsch (1960) and Tyler (personal communication), although intra muscular injections of 10-30 mg cortisone acetate/day cause chickens to lay eggs with thicker shells before the treatment terminates laying (Cooke, 1968). Cortisone causes a type of osteoporosis and hypercalcaemia (Urist & Deutsch, 1960) and marked decreases in body weight (Cooke, 1968), so it would appear that its action is to mobilise skeletal calcium. Thicker shells might be formed because of a short-term rise in available calcium before a calcium deficiency terminates ovulation (Taylor et al., 1962). Alternatively, Simkiss (1961) has suggested that cortisone might inhibit the 3-phosphoadenosine 5-phosphosuIphate system (Whitehouse & Lash, 1961) which is involved in the sulphation of acid mucopolysaccharides (Suzuki & Slrominger, 1959, 1960a, b, c), so reducing the synthesis of shell organic matter essential for proper shell development. It was proposed that this reaction would occur in the isthmus region of the oviduct, but since shell organic material secreted by isthmal cells does not apparently contain acid mucopolysaccharide (Robinson et al., 1968; Cooke & Balch, 1970a), this explanation seems unlikely. Nevertheless, an effect on acid mucopolysaccharide availability in the oviduct certainly cannot be dismissed, as daily intramuscular injections of 10 mg cortisone acetate cause a considerable shift in the ratio of glucuronic and iduronic acid in the shell (Cooke, 1968). Glucuronic acid is a constituent unit of chondroitin sulphate A (see Brimacombe & Webber, 1964), the main acid mucopolysaccharide in egg shell matrix (Baker & Balch, 1962; Cooke, 1968), while iduronic acid, an epiiner of glucuronic acid, is a constituent unit of chondroitin sulphate B, which is also found in the matrix (Baker & Balch, 1962; Cooke, 1968). Three birds subjected to this cortisone acetate treatment for three weeks showed increases in glucuronic acid in the shell of 52, 44 and 73% and decreases in iduronic acid of 37,62 and 59% respectively (Cooke, 1968). Thus cortisone, and to a lesser extent other treatments (see section 3(b)), can somehow reduce the relative availability of chondroitin sulphate B to the developing shell. There has been one experiment linking pesticides, corticosteroids and carbo hydrate metabolism. After treating cockerels with DDT, Srebocan et al. (1971) found a decrease in liver glycogen synthesis associated with a decrease in adrenal and blood corticosteroid levels. Decreased corticosteroid levels might result in thin shells being laid, but supporting evidence is lacking. 4(e) Hormone-type effects: Levin et al. (1968) reported that technical DDT increased uterine wet weight in rats, the active constituent being op'-DDT, which, they found, competed with labelled oeslradiol for binding sites on the uterus. At about the same time, oestrogen-like responses after op'-DDT treatment were reported in weight and glycogen content of chicken and quail oviducts and rat uteri (Bitman et al, 1968). op'-DDT given orally to quail produced, however, no marked changes in serum calcium levels (Bitman et al., 1969; Cooke, 19706) or DSW 031327 STLCOPCB4015289 126 A. S. COOKK in Ihc mating behaviour of males (Cooke, 19706). Since op'-DDT is readily lost by birds (French & Jefferies, 1969; Bitman et al., 1971; Cecil et a!., 1972) and is only very rarely detected in wildlife samples, the intake of this isomer by predatory birds is probably negligible. Even if birds in the field are exposed to technical DDT, they are likely to be killed by pp'-DDT before op'-DDT has any significant sublethal efTect (Cooke, 19706). Nevertheless, that op'-DDT should exhibit oestrogenic-type properties is of great interest and researchers have turned their attentions to its effects on mammals (Welch et al., 1969; Singhal et al., 1970; Cecil et al., 19716; Fleinrichs et al., 1971; Wrenn et a/., 1971). The relative oeslrogcnicity of op'-DDT is about one ten thousandth (Cecil et al., 19716) or one hundred thousandth (Singhal et al., 1970) that of oestradiol. Although op'-DDT has certain oestrogenic properties, it docs not inhibit pregnancy in the rat, unlike oestrone (Duby et al., 1971), and at a dietary level of 10 ppm has no oestrogenic action on ewes (Wrenn et a!., 1971). Bitman & Cecil (1970) extended the study to include other organochlorines and their results indicated that halide or alkyl groups in the para positions on the basic diphenylalkane structure rendered the compounds inactive, while -H, -OH or -OCH3 in these positions rendered them active. Some PCBs were also active. Objections on steric grounds as to why DDT and similar molecules should be oestrogenic (Fisher et al., 1952) have been overcome by Bitman & Cecil (1970). Apparent hormone mimicry is not limited to sex hormone-type effects. It is believed that op'-TDE can compete with thyroxine for protein binding sites in human serum (see section 4(e)) and thereby lower protein bound iodine (Marshall & Tomkins, 1968). With so many egg formation mechanisms under hormonal control, it is not possible to rule out hormone mimicry having effects on egg shell formation. An overall assessment of possible mechanisms So far each mechanism has been discussed individually and an assessment, based on available information, is presented in Table 4. There could be several reasons for the lack of an obvious mechanism. (1) Most of the discussions in this section have been based on knowledge derived from experiments on the chicken, a species relatively resistant to shell thinning by organochlorines. It is quite likely that the physiological and biochemical processes involved in egg shell formation in species that have been observed to lay thin shells in the field dilfer considerably in degree or nature from those in the chicken. For an example of an interspecific difference see the hepatic metabolism data for the chicken and pigeon referred to in section 1(6). There is, however, little information on non-domestic species, and even for the chicken our knowledge of shell formation is very incomplete. (2) The list of factors discussed is unlikely to be exhaustive and shell thinning may be due to a factor not considered. D5W 031328 STLCOPCB4015290 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 127 (3) Pesticides produce such a broad spectrum of effects in an organism, perhaps due to action on the endocrine system, that deciding whether changes in a physio logical process merely reflect this broad effect on the bird or whether they are responsible for reducing shell thickness is difficult. (4) Pesticides may cause thin shells because of effects on several (probably inter related) mechanisms and various factors such as species and environmental conditions might affect which mechanism or mechanisms are dominant in any given situation. TABLE 4 AN ASSESSMENT Or POSSIBLE MECHANISMS BY WHICH ORGANOC lll ORINES C AUSE F.GG SIIKL.I. THINNING Mechanism Evidence Ear Against 1. Reduced availability of calcium ions (a) Absorption from gul (b) Deposition of medullary bone (c) Mobilisation of medullary bone (d) Transport in blood (c) Transport from blood to shell 2. Reduced availability of carbonate ions 3. Effects on other shell constituents (a) Initiation sites (b) Matrix (c) Shell growth inhibitors (d) Premature termination 4. Other factors (a) Premature oviposition (b) Decreased food consumption (c) Thyroid defects (d) Stimulation of adrenals (e) Hormone-type effects i II i ! II I - -! I* i* I -j -! -i I -I or -- indicates a little evidence available; I | or -.... , more evidence; H -I -i or ........ , good evidence. * There is reasonable evidence for these factors being responsible for crystal disruption rather than shell thinning. Some researchers probably already have information that would assist in deciding on likely mechanisms. For instance, it would help to know the pattern ofshell thickness plotted against time when birds are subjected to continuous exposure of pesticide. For the chicken, a low calcium diet, sulphanilamide in the diet and perhaps high temperature all produce characteristic patterns (Fig. 3). The first two patterns are well known and the compensatory rise in thickness at high temperature has been observed before (B. A. Morris, personal communication). Unfortunately, there is no definite information on the pattern produced by pesticides. Information from Bitman et al. (1969) suggests it may be of the low calcium type, but their controls in this experiment were fed a calcium deficient diet and did not lay to this pattern. The results of Stickel & Rhodes (1970) indicate it may be a sulphanilamidetype pattern. On the other hand, shell thickness, relative to egg residues, of eggs laid by North American cormorants in 1965 was greater in the second clutch DSW 031329 STLCOPCB4015291 128 80 KQ> 60 ac> V) 40 l A. S. COOKE (a) Calcium deficient diet (0-8%Ca) Egg production : reduced Shel I thickness : erratic 11 little, if any, reduction steady at reduced level Fig. 3. Type of response in shell thickness shown by chickens to (a) a calcium deficient diet, (b) sulphanilamide in the diet, or (c) elevated environmental temperature. Birds were subjected to the treatment from day 8 to day 28. These curves, which show the salient characteristics of (he responses, have been constructed from the results of Cooke (1968). DSW 031330 STLCOPCB4015292 SHELL thinning in avian lgcs by fnvironmhntal pollutants 129 than in the first (Anderson el al., 1969), and this may indicate operation of a compensating mechanism (Risebrough el al., 1970) as has been observed at elevated temperatures (Fig. 3). The effects on the bird and on the egg of other treatments known to cause thin shells have been frequently referred to in an attempt to find corresponding effects to those produced by pesticides. These effects are summarised in Table 5. The one recurring theme throughout this section has been the similarity in many respects between the effects of high temperature and the effects of pesticides. High tempera ture is believed to reduce shell thickness because blood pC02 is depressed when the birds hyperventilate (Mongin & Lacassagne, 1966; Simkiss, 1968) and it would be interesting to know whether organochlorines have a similar effect on blood pC02. There could be an effect via thyroid-mediated changes in breathing. In pigeons treated with high doses of DDT, Jefferies & French (1971) noted reduced oxygen consumption and body temperature, which may indicate hypoventilation and an accompanying elevation of blood pCOz. This state of respiratory acidosis is, however, the reverse of the alkalosis produced by high environmental temperatures, and the effect of an increased level of atmospheric C02 (to increase blood pC02) on shell thickness is not entirely clear. Thinner shells have been observed after acute exposure (Helbacka et al., 1963; Hunt & Simkiss, 1967) and thicker shells after chronic exposure (Frank & Burger, 1965), the difference apparently being due to renal compensation raising blood bicarbonate levels during long-term studies (Simkiss, 1968). While it is possible that inhibition of thyroid function can contribute indirectly to some observed changes in shell thickness, blockage of sites on thyroxine-binding globulin seems unlikely to be the primary lesion involved in thin shell formation by pesticides. Undoubtedly changes in the thyroids and the adrenals can occur when birds are exposed to organochlorine insecticides, and since DDT can virtually completely inhibit thyroid function (Jefferies, personal communication), the effects on the well-being of the bird must be considerable. Such changes might indeed be responsible for many of the frequently-observed sublethal effects, such as reduced fertility, hatchability, egg weight, chick weight, delay in ovulation, etc. (Jefferies el al., 1971) and they would therefore be expected to exert some influence on essential mechanisms in shelf formation. It is, however, hard to reconcile this broad-based effect due to direct or indirect action on thyroid function with the apparently relatively specific effect of pesticides on shell formation sometimes observed in the field and in the laboratory. Although several mechanisms are probably affected by pesticides when thin shells are produced (see below), present evidence from laboratory experiments suggests that thyroid malfunction has relatively little effect on shell deposition. With respect to thyroid changes, pesticides may again be similar to high environmental temperature, which is known to cause the thyroid secretion rate to fall (jee Sturkie, 1965), but this is seemingly unconnected with the mechanism causing the shells to become thin (see Simkiss, 1968). DSW 031331 STLCOPCB4015293 TABLE 5 A COMPARISON OF THE EFFECTS ON THE BIRD AND ON ITS EGGS OF PESTICIDES, SULPHONAMIDES, AMMONIUM CHLORIDE OR A DEFICIENCY OF CALCIUM IN THE DIET OR HIGH ENVIRONMENTAL TEMPERATURE Pesticides and PCBs High temperature Sulphonamide Ammonium chloride Calcium deficient diet Effects on the bird (a) Blood calcium No change: Bitman et al. Decrease: Mueller Decrease: Siegmund Decrease: Siegmund Decrease: Taylor & (1969), Cooke (19706), (1959) & Du Ice (1960) & Dulce (1960) Hertelendv (1961) Whitehead et al. (1972) No change: Scott No change: Hunt & et al. (1944) Aitken (19626) (b) Other changes in Reduction in medullary Retention from food Retention decreased: Disrupted: e.g. calcium metabolism bone: Peakall (19706), increased: Muller Tyler (1950,1954) Hurwitz (1970), Oestreicher et al. (1971) (1959) Taylor (1970), Bragg Dieldrin no change on bone Incorporation into et at. (1971) deposition: Mueller & shell reduced: Bragg Lockman (1971) Effects et al. (1971) on vitamin Dj mediated changes: Nowicki et at. (19726) (c) Possible connection Jefferies (1969), Jefferies & Reviewed by Sturkie with thyroid malfunction French (1969, 1971), (1965) Jefferies et al. (1971) Richert & Prahlad (1972) (d) Carbonic anhydrase Reduction in shell gland: e.g. Mann & Keilin Unconvincing: see inhibition Peakall (19706), Bitman (1940) Hunt & Aitken (19626) et at. (1970). No in vitro inhibition: Dvorchik et al. tO/0 at Effects on the egg (a) Egg production (1971), Pocker et al. (1971) DDT decrease: Cross et al. Decrease: Sturkie Decrease at high dose Decrease: Hunt & Decrease: e.g. Cooke (1962), Stickel & Rhodes (1965), Cooke (1968), rate: Scott et al. Aitken (1962a) (1968) Bragg et al. o (1970), Sauter & Steele Bragg et al. (1971) (1944) No change: Hunt & (1971) D (1972). Lillie et al. (1972). No change: Tyler Aitken (19626), Cooke DDT no change: Genelly (1950, 1954), Cooke (1968) OJ & Rudd (1956), Tucker & (1968) i\j Haegele (1970), Hurst et al. (1971), Davison & Sell. (1972) Dieldrin decrease: Genelly & Rudd (1956), Atkins & Linder (1967) Dieldrin no change: STLCOPCB4015294 TAB1.F. 1 EXPERIMENTS IN WHICH EOG SHELL THICKNESS HAS BEEN ATEECTED AFTER EXPOSURE OF THE LAYING BIRD TO ORGANOCHLORINE INSECTICIOES Authors) Species Pollutant (s) Exposure Mean egg content (ppm) Shellmmeenafsure- rEffec,t on sh, e.l.l th.i.ck.ness Bilman et at. (1969) Japanese quail C. coturnix japomV Eehner &. Egbert 0969) Mallard Anas plafyrhynchos Porter Sc Wiemeyer (1969) American Kestrel Falco sparverius Heath ct al. (1969) Mallard (see also Heath el al., in press) Jefferies (1969) Bengalese finch Lonchura striata Enderson Sc Berger (1970) Prairie falcon Falco mexicanus pp'-DDT) or > op'-DDTj Dieldrin pp'-DDT plus dieldrin pp'-DDE or pp'-DDT pp'-DDT Dieldrin 100 ppm in low 520'| total residues Cr diet for 45 27 >- of DD r type days ) compounds Up to 10 ppm* Up to 44 in diet for 1-2 years Up to 15 ppm Not given pp'-DDT plus 3 ppm dieldrin* in diet for 1-2 years Up to 40 ppm* Not given in diet for be tween 1 and 2 years Up to 300 pgl Not given day* in diet for about 15 weeks Wild birds fed on 40 + up to 12 tethered starlings previously'dosed with dieldrin Thickness at 3 Decrease. 6%) Compared points around Decrease 4%) wi,h controls waist of shell ' on low Ca without diet membranesf Thickness at Decrease, up to 4%{ broad pole and waist of shell plus outer membranet Thickness al 4 Decrease, up to 17 %t points of shell plus membranesf Thickness of shell plus membranet Decrease, up to 13 %t Shell weight/egg Increase by average of 7 % weight Thickness index Decrease of 9% (compared with controls) for 7 most contaminated eggs i DSW 031333 STLCOPCB4015295 Pcakall (19706) Wiemeyer A Porter (1970) Smith ei at. (1970) Stickel & Rhodes (1970) Tucker & Haegele (1970) Ring dove StrcptopeUa risoria American kestrel pp'-DDE pp'-DDE Chicken Callus dnmesticus Japanese quail Technical DDT pp'-DDT Bobwhite quail 3 Colituts vlrginiam/s > Mallard ) Mallard Technical DDT Technical DDT Longcorerial. (19716) Black duck Anas rubripts Sauter & Steele (1972) Chicken pp'-DDE Technical DDT or lindane Single intraperi- BO Weigh! Decrease by 23 % toneal injec tion of 150 mg/kg 10 ppm in diet 32 Thickness at 4 Decrease by 10% for more than 1 points of shell year plus membranes \ Up to 10 ppm* Up to 6 ppm in yolk Not given Decrease, up to 11 % in diet for 2 months Up to 25 ppm* Not given Not given Decrease, up to 7 in die! for 26 weeks Up to 30 ppm* N1 ot given Thickness at 4 (Decrease, up to 4%t in diet for 96 days points around j Decrease, up to 5 %i waist t Single oral dose Not given As above Decrease by 9% over 16 1000 mg/kg. Food days withheld for 2 days after treatment Up to 30 ppm* Up to 135 Thickness at Decrease, up to 38 in diet for 6 waist and both months poles of shell plus membranes Up to 10 ppm* Not given Not given Decrease, up to 9 in diet for 10 after treatment period weeks Highest treatment dose J Measured with a micrometer Mean thickness change for most affected treatment group ! SW 031334 STLCOPCB4015296 TABLE 3 THE AMOUNT OF MATRJX SIALIC ACIO AND URONIC ACID IN THICK AND THIN SHELLS LAID BY CHICKENS GallltS domestiCUS. FOR DETAILS SEE TEXT (ADAPTED ' FROM COOKE, 1968) "Normal thick' shells, thickness > 70 nig/cmJ `Normal thin' shells, thickness < 70 mg/cm2 'Abnormal thin' shells No. of True shell bulked thickness samples* (mg/cm*) 16 73-7 0-5 Matrix (mglcm2) (mglg shell) 1-09 d: 005* 14-8 d; 0-5* Sialic acid (pglcm1) (pg/mg matrix) 9-3 dt 0 2 8 6 0 3* Uronic acid {pglcm*) (pglmg matrix) 14 4 0 3 13 3 d- 0-4 14 66.2 O S 0 90 0 04c 13 5 0 5 81 i o 2<> 9 2 0 3 12-5 0 6* 13 6 0 3 14 52-7 db 1-3 0-90 d; 0-04c 17 2 db 1-06J 7-7 d; 0-3d 8-7 0-3* 10 6 d: 0 4^ 119 d: 0 4b* d- S.F.. Samples contained segments from 3-13 shells ^ .. , _ b. C d For a pair of means differing by one letter, P < 0 05; by two letters. P < 0 0); by three letters. P < 0 001. } Compared with `normal thin' shells, P < 0 01 STLCOPCB4015297 TABLE 5 A COMPARISON Ot 1 HE ErFECTS ON THE BIND AND ON ITS EGOS OF PESTICIDES, SULPHONAMIDES. AMMONIUM CHLORIDE OR A DErtCIENCV OF CALCIUM IN THE DIET OR HIGH ENVIRONMENTAL TEMPFRATURE Pesticides and PCBs High temperature Sulphonamide Ammonium chloride Calcium deficient diet Effects on the bird (a) Blood calcium No change: Bitman et a!. Decrease: Mueller (1969). Cooke (1970ft). (1959) Whitehead et al. (1972) Decrease: Siegmund & Dulce (1960) No change: Scott et al. (1944) Decrease: Siegmund Sc Dulce (1960) No change: Hunt & Ailken (19626) Decrease; Taylor !c Hertelendy (1961) (b) Other changes in calcium metabolism Reduction in medullary Retention from food bone: Pcakall (19706), increased: Muller Oesi reicher et a!. (1971) (1959) Dicldrin no change on bone Incorporation into deposition: Mueller St shell reduced: Bragg Retention decreased: Tyler (1950,1954) Disrupted: e.g. Hurwitz (1970), Taylor (1970), Bragg et al. (1971) Lockman (1971) Effects et al. (1971) on vitamin Dj mediated changes: Nowicki et al. (19726) (c) Possible connection Jefferies (1969), Jefferies St Reviewed by Sturkie with thyroid malfunction French (1969, 1971). (1965) Jefferies et al. (1971) Richer! & Prahlad (1972) (d) Carbonic anhydrase Reduction in shell gland: -- e.g. Mann St Keilin Unconvincing: see inhibition Pcakall (19706). Bitman (1940) Hunt & Aitken (19626) et al. (1970). No in vitro inhibition: Dvorchik et al. (1971), Pocker et c/. (1971) Effects on the egg (a) Egg production DDT decrease: Cross et al Decrease: Sturkie Decrease at high dose Decrease: Hunt St (1962), Shekel & Rhodes (1965), Cooke (1968). rate: Scott et al Aitken (1962a) Decrease: e.g Cooke (1968) Bragg et al. (1970), Sauter St Steele Bragg et al (1971) (1944) (1972), Lillie er al. (1972). No change: Tyler No change: Hunt & (1971) Aitken (19626), Cooke DDT no change: Genelly & Rudd (1956). Tucker & (1950, 1954), Cooke (1968) (1968) Haegele (1970), Hurst et al. (1971), Davison & Sell. (1972) Dieldrin decrease: Genelly & Rudd (1956), Atkins St Linder J1967) Dieldrin no change: __ueu. viyic) PCB decrease: Hurst el al. (1971) Dahlgren & Linder (1971) Mercury decrease: Fimreite (1971) (b) Egg size or weight Mercury no change: Stoewsand et al (1971) DDT decrease: Jefferies (1969), Bitman el al (1969) DDT no change: Davison St Sell (1972), Cecil et al. (1972) Decrease: Cooke (1968) Sometimes decrease* Cooke (1968. I970n) Decrease: Hall & Helbacka (1959) N^change: Cooke Dieldrin increase: Atkins & Linder (1967) Dieldrin no change: Davison A Sell (1972) Hcxachlorobenzene decrease: Vos et al. (1971) Mercury decrease: DSW 031336 (c) Yolk size Fimreite (1971) DDT decrease: suggested by Jefferies (1971) Lindane no change: Decrease: Sturkie (1965) (d) Apparent ability to cause soft shelled eggs Whitehead el al. (1972) Risebrough el al. (J971). Keith et al. (1970). Faber etal (1972) (c) Compensatory rise in shell thickness (see text) (f) Defects in shell structure in mammillary or palisade layer or on surface Suggested by Risebrough et al. (1970) to explain results or Anderson ft al (1969) Various defects: Erben & Krampitz (1971) McFarland et al (1971) (g) Calcium: percentage in shell Decrease: Longcore et al (l97tol nrvr .................. U968)'nal,y: CC,ke e.g. Tyler (1950. 1954) Not recorded by Cooke. (1968). B.A. Morns (personal No: Cooke (1968) comm.) authors who have used this treatment No: Cooke (1968) Widespread defects: El-Boushy et at (1968) Granular surface* Scott et al (1944) Structure not examined Occasionally, but cut out mechanisms sin,,, Tay,or ef al. ( 96_), Taylor (1970) No: Cooke (1968) Normal mammillary layer but palisade layer deficient - C'~..u . STLCOPCB4015298 (b) Egg size or weight (c) Yolk size (d) Apparent ability to cause soft shelled eggs Davison & Sell. (1972) PCR decrease: Hurst et at. (1971) Dahlgren & Linder (1971) Mercury decrease: Finueite (1971) Mercury no change: Stoewsand al. (1971) DDT decrease: Jefferies Decrease: Cooke (1969). Bitman et at. (1968) (1969) DDT no change: Davison & Sell (1972). Cecil et at. (1972) Dieldrin increase: Atkins' A Linder (1967) Dieldrin no change: Davison &. Sell (1972) Hexachlorobenzene decrease: Vos et al. (1971) Mercury decrease: Fimreite (1971) DDT decrease: suggested Decrease: Sturkie by Jefferies (1971) (1965) Lindane no change: Whitehead et at. (1972) Risebrough et al. (1971), Occasionally: Cooke Keith et at. (1970), Faber (1968) et al. (1972) Sometimes decrease: Cooke (1968, I970n) Decrease: Hall & Helbacka (1959) No change: Cooke (1968) e.g. Tyler (1950. 1954) Not recorded by authors who have used this treatment (e) Compensatory rise in shell thickness (see text) (f) Defects in shell structure in mammillary or palisade layer or on surface (g) Calcium: percentage in shell (h) Matrix: increase in percentage in shell or change in composition Suggested by Risebrough et at. (1970) to explain results of Anderson et al. (1969) Various defects: Erben & Krampitz (1971) McFarland et at. (1971) Cooke, (1968), B.A. Morris (personal comm.) Widespread defects: El-Boushy et al. (1968) Decrease: Longcore et al. (197(o). DDT and dieldrin no change: Davison & Sell (1972) Erben & Krampitz (1971) -- Cooke (1968) No: Cooke (1968) Granular surface: Scott et at. (1944) Structure not examined No change: Tyler (1950) Cooke (1968) No: Cooke (1968) Cooke (1968) Decrease: Bragg et al. (1971) Occasionally, but cut out mechanisms should operate: Taylor et al. (1962), Taylor (1970) No: Cooke (1968) Normal mammillary layer but palisade layer deficient: Cooke (1968) No change: Bragg et al. (1971) DSW 031337 STLCOPCB4015299 132 A. S. COOKE Failure to find the primary cause of shell thinning is hardly surprising since, in the past, investigations into mechanisms for thin shell formation have rarely led to unquestionable conclusions and often seemingly well-proven and widely-accepted theories have fallen from favour. The mode of action of sulphanilamide on chickens seemed so simple in the 1940s. Gutowska & Mitchell (1945) observed that the CA content of the shell glands of good layers was greater than that of shell glands of poor layers, and knowing that sulphanilamide, a CA inhibitor (Mann & Keilin, 1940), caused thin shells (Scott et al., 1944), their theory that enzyme inhibition produced the thin shells seemed completely sound. The first seeds of doubt were sown when Tyler (1950,1954) showed that sulphanilamide caused changes in calcium retention that could not be explained as secondary effects arising from inhibition in the shell gland. Then reduction of blood calcium levels by acetazolamide was reported by Siegmund & Dulce (1960), although Scott el al. (1944) had previously found that sulphanilamide did not affect blood calcium. In 1962 Mueller proposed that sulphanilamide caused thin shells because it was a diuretic. Even though Heald et al. (1968) demonstrated that CA content of the shell gland is unrelated to shell thickness, sulphonamides are now once again thought to cause thin shells by CA inhibition in the shell gland (Bernstein et al., 1968), although it is generally appreciated that enzyme inhibition cannot be the complete answer since both metabolic and respiratory acidosis occur after treatment (Mongin, 1970). This and other information given in this section suggests that sulphanilamide, ammon ium chloride and high temperature affect, in chickens, a variety of mechanisms involved in shell formation, but shell thinning is largely brought about by a change in a single vital mechanism. Treatment of birds with organochlorines has been shown to bring about detri mental changes in many processes involved in egg shell formation (Table 4). None of these changes can alone account for the observed decreases in shell thickness, and it is likely that each suggested mechanism plays an appreciable part in some situations but a negligible part in others. For instance, if medullary bone deposition is reduced by increased hepatic metabolism of the sex hormones, then this should have a lesser role in shell thinning in gallinaceous species than in pigeons and doves (see section 1(6)). The evidence suggests that, in shell thinning by organochlorines, no one mechanism is dominant, irrespective of conditions. However, when very thin shells are formed after exposure to organochlorines, only one mechanism is likely to have been markedly affected or else each major affected mechanism must be very close to the site of shell formation, since a general effect of all mechanisms would be expected to produce other symptoms such as prevention of ovulation or changes in behaviour. The brown pelican colony present on Anacapa Island in 1969 can be taken as an example, since, although the birds were atypical as regards the extent of shell thinning, their reproductive failure has been well documented. Their egg shells had relatively normal mammillary layers but deficient palisade layers (Erben & DSW 031338 STLCOPCB4015300 DSW 0 3 1 3 3 9 (b) Egg size or weight (c) Yolk size (d) Apparent ability to cause soft shelled eggs Davison & Sell. (1972) PCB decrease: Hurst et al. (1971) Dahlgren & Linder (1971) Mercury decrease: Fimreite (1971) Mercury no change: Stoewsand et al. (1971) DDT decrease: Jefferies Decrease: Cooke (1969), Bitman et al. (1968) (1969) DDT no change: Davison & Sell (1972), Cecil et al. (1972) Dieldrin increase: Atkins & Linder (1967) Dieldrin no change: Davison & Sell (1972) Hexachlorobenzene decrease: Vos et al. (1971) Mercury decrease: Fimreite (1971) DDT decrease: suggested Decrease: Sturkie by Jefferies (1971) (1965) Lindane no change: Whitehead et al. (1972) Risebrough et al. (1971), Occasionally: Cooke Keith et al. (1970), Faber (1968) etal. (1972) Sometimes decrease: Cooke (1968, 1970a) Decrease: Hall & Helbacka (1959) No change: Cooke (1968) e.g. Tyler (1950, 1954) Not recorded by authors who have used this treatment (e) Compensatory rise in shell thickness (see text) (f) Defects in shell structure in mammillary or palisade layer or on surface (g) Calcium: percentage in shell (h) Matrix: increase in percentage in shell or change in composition Suggested by Risebrough et al. (1970) to explain results of Anderson et al. (1969) Various defects: Erben & Krampitz (1971) McFarland et al. (1971) Cooke. (1968), B.A. Morris (personal comm.) Widespread defects: El-Boushy et al. (1968) Decrease: Longcore et al. (1971a). DDT and dieldrin no change: Davison & Sell (1972) Erben & Krampitz (1971) Cooke (1968) No: Cooke (1968) Granular surface: Scott et at. (1944) Structure not examined No change: Tyler (1950) Cooke (1968) No: Cooke (1968) Cooke (1968) Decrease: Bragg et al. (1971) Occasionally, but cut out mechanisms should operate: Taylor et at. (1962), Taylor (1970) No: Cooke (1968) Normal mammillary layer but palisade laver deficient: Cooke (1968) No change: Bragg et al. (1971) SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 133 Krampitz, 1971), evidence consistent with a general calcium deficiency being responsible (Cooke, 1968). This is not supported by other evidence, however. The average decrease in shell thickness was 50% and many birds laid eggs with soft shells (Risebrough et a!., 1971). Calcium deficient poultry do not usually lay eggs with soft shells, they stop laying instead. This is thought to be due to low blood calcium levels triggering a cut-out mechanism that prevents further secretion of gonadotrophins from the pituitary (Taylor elal., 1962).In thiswaybirds are protected from utilising too much skeletal calcium. That the pelicans laid eggs with soft shells suggests that a general calcium deficiency was not to blame (assuming that pelicans also have a cut-out mechanism) and mechanisms 1 (d)-(d) can be ruled out. Although evidence implicating reduced transport of calcium from the blood to the shell gland (1(e)) is not sufficiently convincing to suggest that here is the cause of a 50 % reduction in shell material, it should be remembered that again the support ing experiments have been carried out on laboratory species. Since movement of calcium ions across the shell gland wall can be reduced by prior treatment with DDE (Pcakall & Lincer, 1970b; Peakall, personal communication), a reduction of available calcium because of reduced transport may have been responsible, at least in part, for the thin shells. Mechanisms such as effects on thyroid or adrenal function or decreases in food consumption that would be expected to cause other obvious symptoms can probably be discounted. There is no evidence to suggest that the change in the amino acid composition of shell matrix noted by Erben & Krampitz (1971) can have played more than a very minor role in producing the thin shells. This leaves premature termination and carbonate deficiency to be assessed. The former is feasible, at least for these pelican eggs {see section 3(d)), but, as with all the suspected mechanisms, further research is needed. A reduction in carbonate availability can lead to the laying of eggs with soft shells. It has long been known that sulphanilaniide treatment can reduce shell thickness to such an extent, as can changes in the acid-base balance of the blood, since the maximum reduction in thickness for a shell laid by any of the three control birds at high temperature (Table 2) was 75%. Attempts to explain how pesticides might reduce carbonate availability have mainly rested on CA inhibition. Failure to inhibit CA in vitro (Dvorchik et al., 1971; Pocker et al., 1971) can be overcome by assuming a metabo lite to be the active compound. There is a marked similarity between the shell thinning response of chickens to various levels of dietary sulphanilaniide (Scott el al., 1944) and the response of brown pelicans from twelve American colonies to environmental DDE levels, as rellected by egg residues (Blus et al., 1972). Inhibition of CA or possible change in the acid-base balance of the blood must at present, however, remain speculation. Nevertheless, the evidence points towards enzyme inhibition close to the site of shell formation being the dominant affected mechanism, at least in this particular instance (Risebrough et al., 1970). As suggested by these writers, mechanisms that might be involved are (a) inhibition of CA leading to a reduction in carbonate availability and (b) inhibition of ATPases in DSW 031340 r STLCOPCB4015302 134 A. S. COOKK (lie shell gland resulting in reduced calcium ion concentration in the lumen. If the action of the factor rcponsible for premature termination is confined to the shell gland, then premature termination could be at least a contributory mechanism. Premature termination certainly cannot be implicated in all situations, since this was not responsible for the thin shells examined by McFarland et al. (1971). Finally, something should be said about the accusations (e.g. by Rogers, 1972) that disturbances by visiting biologists during 1969 and 1970 contributed to shell thinning and breeding failure on Anacapa. Disturbance of the breeding birds sufficient to increase shell breakage during incubation or adversely affect breeding in any other way is to be deplored. However, shells of eggs laid prematurely by poultry in a slate of shock are characterised by a lack of the outer shell components, and the two thin shells, produced by Anacapa pelicans, that were examined by Erben & Krampilz (1971) both possessed the outermost component, the calcareous cover. There is no reason to doubt that these shells were typical of the thin shells laid by this pelican colony. Shell thinning appears to have been caused by a biochemical, rather than an emotional, disturbance. Tin; SIGNIFICANCE OP THIN SIIIT.I.S TO AVIAN POPULATIONS IN TUP. FIELD That a decrease in shell thickness has occurred for certain populations of birds of prey since the Second World War has seldom, unlike its cause, been disputed. It is important to determine what effect this has had on population levels. At the beginning of the first section of this paper it was pointed out that Ratcliffe initially studied shell thickness because of an increased incidence of egg breakage in the nest. He suspected that thin shells were responsible for increased shell breakage (Ratcliffe, 1967, 1970), and Prestt (1970) reported that, in the three heron colonies he studied in Eastern England, virtually every female laid eggs with abnormally thin shells and egg shell breakage was practised by about one third of the pairs. In North America an association between decreased shell thickness and increased shell breakage has been reported by Hickey & Anderson (1968) for herring gulls in Wisconsin, by Berger et al. (1970) for peregrines in Quebec, and by Keith cl al. (1970) and Risebrough et at. (1970) for the Anacapa Island pelicans. Similarly, an association between a decrease in shell thickness and a decrease in population has been observed for the prairie falcon (Fyfe et al., 1969), merlin (Fox, 1971), doublecrested cormorant (Anderson et al., 1969), brown pelican (Keith et al., 1970; Risebrough et al., 1970, 1971 ;Blus, 1970), bald eagle, osprey and peregrine (Hickey & Anderson, 1968). Although common egrets in the Audubon Canyon Ranch colony in California did not decline between 1967 and 1970, a breeding failure associated with thin shells was observed (Faber et al., 1972). Some of the birds produced eggs with soft shells, and, in 1970, 54% of their attempts to nest suffered shell breakage during incubation. The fledging success of prairie falcons in Wyoming DSW 031341 STLCOPCB4015303 SIILLL THINNING IN AVIAN TOGS BY LNVIRONMLNTAL POLLUTANTS 135 and Colorado in 1967 and 1968 was directly related to shell thickness (Enderson & Berger, 1970). Twenty-one pairs laying eggs with relatively thick shells had a fledging rate of 2-6 young per pair, while ten pairs producing thin shells fledged only three young between them. No declines occurred for regional populations of several raptor species that had not suffered shell thinning (Hickey & Anderson, 1968). In Alaska, however, where peregrines have laid eggs with thin shells for about two decades, breeding populations have remained steady although breeding failures have been observed recently (Cade et al., 1971). This is similar to the situation in Britain, where the golden eagle (Ratcliffe, 1970) and the heron (Prestt, 1970) have been observed to lay eggs with thin shells in some regions, but breeding populations have not declined. Theoretically a population of a long-lived, slowbreeding bird is much more quickly reduced by the death of adults than by the breeding failure of those adults (Young, 1968). Thus, although thin shells due to organochlorine insecticides have probably often caused breeding failure amongst British raptors, adult mortality due to cyclodienes is a more likely reason for the population crashes of the peregrine (Ratcliffe, 1970) and sparrowhawk (Prestt, 1965; Jefferies & Prestt, 1966). Indeed, during the first decade after the onset of shell thinning, the peregrine increased in several regions after being reduced by controlled killing during the war (Ratcliffe, 1970). It is interesting to note that although in the Californian egret colony referred to above (Faber et al., 1972) DDE levels in adults' tissues and in eggs were comparable with those in other species laying eggs with thin shells, dead and dying adults were found apparently poisoned by the dieldrin also present in their tissues. Although reduced breeding success may not have caused population crashes in Britain, it has probably retarded subsequent recoveries (Jefferies & Prestt, 1966; Ratcliffe, 1970). An egg laid by a bird previously exposed to organochlorine insecticides might be of reduced viability for several reasons. If the shell is thin it is more likely to be cracked when laid (McNally, 1965). When captive ducks were exposed to DDE (Heath et al., 1969; Longcore et al., 19716) shell thickness was reduced and there was an increased incidence of cracked shells. Cracks elevate the risk of bacterial infection (Romanolf & Romanoff, 1949) so causing embryonic mortality. Heath et al. (1969) reported that hairline cracks in the shell halved the hatchability of mallard eggs while all eggs with shell fractures failed to hatch. Dead embryos were found by Berger et al. (1970) in six falcon eggs with cracked shells collected from the field. The eggs contained about 13 ppin DDE (wet weight). Ratcliffe (1970) proposed that brooding adults might respond to a damaged shell by destroying the egg, often eating it. He suggested that egg eating may be a normal adaptive response to certain conditions that reduce breeding success, and insecticide residues in the bird might elicit a similar response. Parental destruction of an egg with a damaged shell or destruction of an infertile or addled egg will not reduce reproductive success, whereas deliberate destruction of an egg with a thin but uncracked shell might kill an embryo that would otherwise have survived. DSW 031342 136 A. S. COOKB It has been tentatively suggested that egg shell breakage amongst herring gulls in a Lake Michigan colony (Ludwig & Tomoff, 1966) and in a Wisconsin colony (Keith, 1966) may have been due to sublethal doses of organochlorines causing aberrant behaviour in the brooding adults. Recently, Kenny & Dacke (personal communica tion) have demonstrated that Japanese quail maintained on a calcium deficient diet tend to destroy and eat many eggs immediately after oviposition, and addition of commercial DDT to the diet usually significantly increases the proportion lost. This could account, at least partially, for an observed decrease in egg production by captive birds being dosed with pp'-DDT (Stickel & Rhodes, 1970; Lillie et a!., 1972) or PCBs (Dahlgren & Linder, 1971). The likelihood of shells being eaten in the field because of `calcium hunger' was discussed by Ratcliffe (1970), but he concluded that, since birds of prey have some times been observed to throw eggs out of the nest or cat the contents and leave the shell, `it is therefore possible that egg eating is simply a behavioural response to the internal state of the bird, with no nutritional significance'. In the peregrine, egg loss has been most frequently observed during the later stages of incubation, although it can occur at any time after the first egg has been laid (RatclifTe, 1970). This contrasts with the experiment of Kenny & Dacke (personal communication) in which many eggs known by palpation to be calcifying were never seen and were presumed to have been eaten soon after being laid. If this occurred in the field it is improbable that the birds would even be suspected of having laid. After inducing birds to lay thin-shelled eggs on a diet contaminated with insecti cide, several workers have carried their experiments one stage further, either by allowing natural incubation (Porter & Wiemeyer, 1969) or else by artificially incubating the eggs (Heath et a!., 1969; Stickel & Rhodes, 1970; Sauter & Steele, 1972). Porter & Wiemeyer (1969) reported that the factor mainly responsible for reproductive failure amongst their dosed American kestrels was the disappearance of eggs during incubation plus, perhaps, some disappearance of newly hatched young. This was attributed to the egg or young bird being eaten by the adults after (accidental?) shell breakage, due to abnormally thin shells. Data on artificial incubation of eggs with thin shells are valuable because such eggs are no longer vulnerable to fits of aberrant behaviour by the adult bird. Reduced breeding success from eggs with thin, but sound, shells need not solely be due to parental breakage. The thin shells themselves might not be such a hazard to the developing embryo as the pesticide in the egg contents and indeed reduced shell thickness might sometimes merely be an indicator of harmful pesticide residues in the yolk. Mortality would be expected mainly (a) during the later stages of incubation due to increased rate of uptake (Guthrie & Donaldson, 1970; Cooke, 1971) or (b) during the first few days of fife due to absorption of yolk sac material (Koeman et a!., 1967) coupled with mobilisation of fat reserves (Cooke, 1971). Increased mortality at these stages, probably due to pesticide, has been observed both in the laboratory and in the field. DSW 031343 STLCOPCB4015305 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 137 (a) After treating mallards with DDE, Heath et al. (1969) artificially incubated the eggs with sound shells and found that eggs with thin shells from dosed birds had reduced hatchability, mainly because of increased embryo mortality during the final stages of incubation. Eggs with thin shells laid by chickens treated with technical DDT or lindane by Sauter & Steele (1972) also had reduced hatchability, but increased embryonic mortality occurred throughout the incubation period. Stickel & Rhodes (1970) incubated eggs with thin shells laid by groups of quail on a series of pp'-DDT-treated diets. There was a tendency for eggs from the highest dosed group to have reduced hatchability, but the difference was not significant. Dahlgren & Linder (1971) reported increased mortality amongst pheasant chicks during the hatching stage after adults had been treated with Aroclor 1254. In this experiment PCBs did not cause thin shells and this is presumably an example of mobilisation of egg residues rather than fragile shells having a detrimental effect on chick survival. In the field dead, almost fully-developed, embryos have been found in contaminated falcon eggs (Fyfe et a!., 1969; Berger et al., 1970), but since there will, of course, also be some embryonic mortality amongst uncontaminated eggs, dead-in-shell embryos in the field cannot be regarded as more than supporting evidence. A report by Peakall (1970a) of an interesting experiment carried out by Spitzer suggests that breeding failure amongst ospreys in the United States might be due to egg residues causing embryonic mortality. Eggs were exchanged between nests in a failing New England colony and a flourishing colony in Chesapeake Bay. The New England eggs still suffered a low hatching success in the Chesapeake Bay nests, while the Chesapeake Bay eggs hatched quite normally in the ailing New England colony. (b) Young Sandwich terns, Sterna sandvicensis, in a Dutch colony have been observed dying in convulsions after hatching from eggs with high organochlorine residues (Koeman et al., 1967). Feeding pheasants on diets containing DDT or dieldrin resulted in decreased survival of young during the first two weeks of life, particularly during the first day after hatching (Genelly & Rudd, 1956). Injection of eggs with DDT or dieldrin solutions insufficient to cause embryonic mortality can kill young chicks soon after hatching if food is withheld (Koeman et al., 1967; Dunachie & Fletcher, 1969). The disappearance of American kestrel chicks after hatching (Porter & Wiemeyer, 1969) could be accounted for by death due to an internal post-hatch release of pesticide followed by being eaten by the adult. Fimreite (1971) fed pheasants with niethylmercury-treatcd grain and noted a significant increase in unfertilised eggs and/or in embryonic mortality early in the (artificial) incubation period. As yet no one has investigated the rate at which mercury in the egg becomes incorporated in the embryo's tissues. Thus, to summarise, there are probably three main reasons why a raptor popula tion may suffer loss of embryos or young chicks as a consequence of prior exposure to organochlorine insecticides. DSW 031344 1 STLCOPCB4015306 138 A. S. COOKK. (1) Thin shells may be cracked during laying or become cracked later and the embryos die. Parental destruction of such eggs may occur before or after death of the embryos. (2) Eggs, embryos or chicks may be destroyed or eaten because of aberrant behaviour by the adults. (3) Embryos or young chicks might die because of residues mobilised from the egg contents and again this may be followed by being eaten. glnlral conclusions For many species of birds, particularly raptors, in Europe and North America, egg shell thickness has decreased since the Second World War. The species most affected are usually those with the highest levels of organochlorine insecticide residues in their tissues. Often, spatial and temporal associations have been noted between shell thickness declines and organochlorine insecticide usage, and rela tionships have frequently been observed between the decrease in shell thickness and levels of pp'-DDE in the parent bird or in the egg. The DDE in the laying hen may be the residue directly responsible for shell thinning, but it may simply be indicative of a more harmful compound (or compounds). If this proposed com pound is not in the DDT group, its physical and chemical properties must closely resemble those of pp'-DDE in order to account for the above relationships. PCBs and other organochlorine insecticides have similar properties, and within a locality residue levels of PCBs and organochlorine insecticides in living organisms are related. Present field evidence suggests that in North America these compounds have probably played no more than a minor role, although in Britain the cyclodienes have made a significant contribution towards the reduction of egg shell thickness in certain species exposed to relatively high environmental levels. The possibility cannot be dismissed of an unstudied pollutant with similar properties being responsible. There is at present no evidence to implicate mercury in these widespread declines in shell thickness. Findings from experiments on captive birds generally support the conclusions from field evidence. There are now more than 20 reports of organochlorine or organophosphate insecticides, PCBs or mercury compounds causing thin shells in experimental birds. Unfortunately, in the great majority of these experiments the pollutant was administered via the diet and food consumption was neither controlled nor monitored. Decreased food consumption, either by partial rejection of a diet of reduced palatability or by loss of appetite through the action of tissue residues, could lead to the production of thin shells. Available evidence on the consumption of treated diets does, however, indicate that decreased food consumption is unlikely to have been serious in most of the experiments. There are many reports of pesticides and PCBs having no effect on shell thickness, and some marked DSW 031345 STLCOPCB4015307 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS I 39 differences between very similarly-designed experiments, some reporting no change and some a significant decrease, cannot be satisfactorily explained. The main conclusions from the controlled experiments are as follows. (1) Some experiments demonstrate that environmental pollutants can cause shell thinning in captive birds. (2) DDT-type compounds generally produce a greater response than other organochlorine insecticides or PCBs. (3) Interspecific differences in response exist, gallinaceous species tending to be the most resistant. Falcons seem to be the most susceptible birds so far studied. These conclusions support the field evidence set out above. Several experiments in which shell thickness lias been reduced by organophosphatc insecticides or mercury compounds suggest that, should birds come into contact with high levels of these compounds in the field, then shell thickness could be reduced. Much work has been carried out in an attempt to elucidate the mechanisms involved in the reduction of shell thickness by organochlorines. Effects have been observed on many mechanisms known to be essential for proper shell formation. Birds laying eggs with thin shells after treatment with an organochlorine insecticide might suffer from (a) a general impairment of calcium metabolism, (b) a reduction in available carbonate in the shell gland lumen, (c) a deleterious effect on both the thyroids and the adrenals, and (d) an alteration in organic matter being incorpor ated into developing shells. Each of these detrimental changes is probably potentially capable of thinning the shell or disrupting shell structure, and the extent of the contribution of each mechanism is likely to be governed by variables such as species, condition of the bird and environmental conditions. When very thin shells are formed after exposure to organochlorines, only one mechanism is likely to have been markedly affected, or, alternatively, each major mechanism affected must be close to, or in, the shell gland, since a general effect on all mechanisms would be expected to produce other responses such as prevention of ovulation or a change in behaviour. Other factors that cause shell thinning in chickens, such as sulphanilamide, ammonium chloride or high environmental temperature, also affect a variety of mechanisms involved in shell formation, but shell thinning is thought to be brought about mainly by a change in a single mechanism. There is no evidence to suggest that in shell thinning mediated by organochlorines the same mechanism is always dominant, irrespective of the circumstances. Similarities have been noticed between the effects of organochlorines and high environmental temperature on the bird and on the egg. In the case of the brown pelicans breeding on Anacapa Island, evidence points towards shell thinning being caused by one or more of the following: 0) a reduction in available carbonate, perhaps by CA inhibition; (2) reduced transport of calcium from the blood to the shell gland lumen; and (3) premature termination of shell growth. DSW Q3L346 STLCOPCB4015308 140 A. S. COOKE In both Britain and North America, eggs with thin shells have led to breeding failure amongst predatory birds. In North America thin shells have often been associated with population declines, but in Britain population declines are thought to have been due to increased adult mortality because of cyclodiene poisoning. An egg with a thin shell is more liable to be cracked and this may lead to death of the embryo because of bacterial infection or by creating a stimulus for destruc tion by the brooding parent. An egg with a thin but sound shell might fail to produce a healthy chick because of (a) aberrant behaviour by the parents suffering from sublethal poisoning, resulting in destruction of the egg or chick, or (b) lethal residues mobilised from the yolk lipid. ACKNOWLEDGEMENTS ] thank Dr D. J. Jefferies, Dr N. W. Moore, Dr F. Moriarty, Dr D. A. Ratcliffe and Professor K. Simkiss for their helpful comments during the preparation of the manuscript. REFERENCES Aitkf.n, R. N. C. (1971). The oviduct. In Physiology and biochemistry of the domestic fowl, ed. by D. J. Bel! & B. M. Freeman, 1237-89. London, Academic Press. Anderson, A. D. & March, R. (1956). Inhibitors of carbonic anhydrase in the American cock roach, Periplaneta americana (L.). Can. J. Zoo!., 34, 68-74. Anderson, D. W. & Hickey, J. J. (1970). Oological data on egg and breeding characteristics of brown pelicans. Wilson Bull., 82, 14-28. Anderson, D. W., Hickey, J. J., Risebrough, R. W., Hughes, D. F, & Christensen, R. E. (1969), Significance of chlorinated hydrocarbon residues to breeding pelicans and cormorants. Can. Fid Nat., S3, 91-112. Anderson, D. W. & Kreitzer. J. F. (1971). Thickness of 1967-1969 whooping crane eggshells compared to that of pre-1910 specimens. Auk, 88, 433-4. Anderson, D. W., I.umsden, H. G. & Hickey, J. J. (1970). Geographical variation in the eggshells of common loons. Can. Fid Nat., 84, 351-6. Asmundson, V. S. & Pinsky. P. (1935). The effect of the thyroid on the formation of the hen's egg. Poult. Sci., 14, 99-104. Atkins, T. D. & Linder, R. L. (1967), Effects of dieldrin on reproduction of penned pheasants. J. Wild!. Mgmt, 31, 746 -53. A vi-Dor, Y. & Gonda, O. (1959). Studies on the adenosine triphosphate-phosphate exchange and the hydrolysis of adenosine triphosphate catalysed by a particulate fraction from the mosquito. Biochem. J., 72, 8-14. Baker, J. R. (1959). The organic constituents of eggshells. Ph.D. thesis, University of Reading. Baker, J. R. & Balch, D. A. (1962). A study of the organic material of hen's-egg shell. Biochem. J., 82, 352-61. Benoit, J. & Clavert, J. (1947). Action de la thyroidcclomie sur l'ossification folliculinique clicz le canard. C. r. Sfanc. Soc. Biol., 141, 1256-7. Berg, L. R. & Bearse, G. E. (1951). Effect of iodinated casein and thiouracil on the performance of laying birds. Poult. Sci., 30, 21-8. Berger, D. D., Anderson, D. W., Weaver, J. D. & Risebrough, R. W. (1970). Shell thinning in eggs of Ungava peregrines. Can. Fid Nat., 84, 265-7. Bernstein, R. S., Nevai.ainen, T., Scitraer, R. & Schraer, H. (1968). Intracellular distribution and role of carbonic anhydrase in the avian (Callus domesticus) shell gland mucosa. Biochim. hiophys, Acta, 159, 367-76. DSW 031347 STLCOPCB4015309 SIIHLL THINNING IN AVIAN EGGS BY ENVIRONMENTAL ROLLU'I AN IS 141 Bernstein, R. S. & Schraer, R. (1969). Carbonic anhydrasc and other proteins in the secretions of the avian shell gland mucosa. Fedn Froc. Fedn Am. Socs exp. Biol., 28, 837. Bitman, J. & Cecil, H. C. (1970). Estrogenic activity of DDT analogs and polychlorinated biphenyls. J. agric. Fd Chem., 18, 1)08-12. Hitman, J., Cecil, H. C. & Fries, G. F. (1970). DDT-induccd inhibition of avian shell gland carbonic anhydrase: a mechanism for thin eggshells. Science, Ar.Y., 168, 594 -6. Bitman, J., Cfcmi., H. C., Harris. S. J. & Fries, G. F. (1968). Estrogenic activity of o.p'-DDT in the mammalian uterus and avian oviduct. Science, N.Y., 162, 371-2. Bitman, J., Cecil, H. C., Harris, S. J. & Fries, G. F. (1969). DDT induces a decrease in eggshell calcium. Nature, Land., 224, 44-6. Bitman, J. Cecil, H. C, Harris, S. J. & Fries, G. F. (1971). Comparison of DDT effect of pentobarbital metabolism in rats and quail. J. agric. Fd Chem., 19, 333-8. Bjerk, J. E. & Holt, G. (1971). Residues of DDE and PCB in eggs from herring gull (Lams argcnlatus) and common gull (Larin conus) in Norway. Acta vet. seand., 12, 429-41. Blunt, J. W., Deluca, H. F. &Schnoes, H. K. (1968). 25-hydroxycholccalciferol. A biologically active metabolite of Vitamin Dj. Biochemistry, 7, 3317-22, Blus, L. J. (1970), Measurements of brown pelican eggshells from Florida and South Carolina. BioScience, 20, 867-9. Blus, L. J.. Gish. C. D.. Bf.lisle. A. A. & Prouty. R. M. (1972). Logarithmic relationship of DDE residues to eggshell thinning. Nature, Lond., 235, 376-7. Blus, L. J., Heath, R, G., Gish, C. D., Belisle, A. A. & Prouty, R. M. (1971). Eggshell thinning in the brown pelican: implication of DDE. BioScience, 21, 1213-5. Bragg, D. B., Floyd, J. & Stephenson, E. L. (1971). Factors affecting the transfer of calcium (4?Ca) from the hen's diet to the egg shell. Poult. Sci., 50, 167 -73. Brimacombe, J. S. & Webber, J. M. (1964). Mucopolysaccharides. Amsterdam, Elsevier. Britton. W. M. (1970). Effect of DDT on liver microsomal estrogen metabolism in the laying hen. Poult. Sci., 49, 1371-2. Brooks, J. & Hale, H. P. (1955). Strength of (he shell of the hen's egg. Nature, Lond., 175, 848-9. Brunnert, H. & Matsumura, F. (1969). Binding of l,l,l-trichIoro-2,2-di-p-chlorophenylethane (DDT) with subcellular fractions of rat brain. Bioehem. J., 114, 135-9. Cade, T. J,, Lincer, J. L., White, C. M., Roseneau, D, G. & Swartz, L. G. (1971). DDE residues and eggshell changes in Alaskan falcons and hawks. Science, N.Y., 172, 955-7. Cecil. H. C., Bitman, J. & Harris, S. J. (1971a). Effects of dietary p,p'-DDT and p.p'-DDE on egg production and egg shell characteristics of Japanese quail receiving an adequate calcium diet. Poult. Sci., 50, 657-9. Cecil, H. C., Bitman, J. & Harris, S. J. (19716). Estrogenicity of o.p'-DDT in rats. J. agric. Fd Chem., 19, 61-5. Cecil, H. C,, Fries, G. F,, Bitman, J., Harris, S. J., Lillie, R. J. & Df.nton, C. A. (1972). Dietary p,p'-DDT, o.p'-DDT or pp'-DDE and changes in egg shell characteristics and pesticide accumulation in egg contents and body fat of caged white leghorns. Poult. Sci., 51, 130-8. Comar, C. L. & Driggers, J. C. (1949). Secretion of radioactive calcium in the hen's egg. Science, N.Y., 109, 282. Conney, A. H., Welch, R. M.. Kuntzman, R. & Burns, J. J. (1967). Effects of pesticides on drug and steriod metabolism. Clin. Pharmacol. & Therap., 8, 2-10. Cook, J. W. (1964). Review of the persistent organochlorine pesticides. Report by the Advisory Committee on poisonous substances used in agriculture and food storage. London, HMSO. Cooke. A. S. (1968). The composition and distribution of organic matter in the hen's egg shell. Ph.D. thesis, University of Reading. Cooke, A. S. (1970a). Some observations on corrugated egg shells. Br. Poult. Sci., 11, 291-7. Cooke, A. S. (19706). The effect of op'-DDT on Japanese quail. Bull, environ. Contain. & Toxicol. 5, 152-7. Cooke, A. S. (1971). Uptakeof DDT and DDE by the quail embryo and chick. Pestic. Sci.,2,144 -7. Cooke, A. S. (1972). Indications of recent changes in status in the British Isles of the frog (Raita temporaria) and the toad (Bufo bufo). J. Zool, Lond., 167, 161-78. Cooke, A. S. & Balch, D. A. (1970a). Studies of membrane, mammillary cores and cuticle of the hen egg shell. Br. Poult. Sci., 11, 345-52. Cooke, A. S. & Balch, D. A. (19706). The distribution and carbohydrate composition of the organic matrix in hen egg shell. Br. Poult. Sci., 11, 353-65. Corradino, R. A., WassermaN. R. H., Pubols, M. H. & Chang. S. I. (1968). Vitamin Dj induc tion of a calcium-binding protein in the uterus of the laying hen. Archs. Bioehem. Biophys., 125, 378-80. Cross, D. L., King, H. L. & Haynes, D. L. (1962). The effects of DDT in the diet of Japanese quail, Q. Bail. Mich. St. Univ. agric. Exp. Stn, 44, 688-96. DSW 031348 STLCOPCB4015310 142 A. S. COOKE Dahlgrfn, R. B. & Linder, R. L. (1970). Eggshell thickness in pheasants given dieldrin J Wild!. Mgmr, 34, 226 -8. Dahlgrfn, R. B. & Linder, R. L. (1971). Effects of polychlorinated biphenyls on pheasant reproduction, behaviour and survival. J. Wild!. Mgmr, 35, 315-9. Davison, K. L. Sell, J. L. (1972). Dieldrin and p.p'-DD f effects on egg production and eggshell thickness of chickens. Bull, environ. Contam. & Toxicol., 7, 9-18. DrAMANTSTFiN, T. (1966). Uber die lokalc Rolle der Karboanhydratasc ini Hinblick auf die Eischalcnverkalkung. Arch. Geftugeik., 30, 309-20. Diamanistein, T. & Schluens, J. (1964). Lokalisation und Bedeutung der Karboanhydrase ini Uterus von Lcgehennen. Acta histochem., 19, 296-302. Dischf. Z. (1965). The informational potentials of conjugated proteins. In Protides ofthe biological fluids, cd. by H. Peelers, 13, 1-20. Amsterdam. Elsevier. Driggers, J. C. & Comar, C. L. (1949). The secretion of radioactive calcium (Ca45) in the hen's egg. Ponlt. Sci., 28, 420-4, Drori, D., Voi.cani, R., Efige, Y., Shalmon, E. & Passy, N. (1964). Factors affecting the C'a45 and Sr*9 transfer in the laying hen. Ponlt. Sci., 43, 486-91. Duijy, R. T., Travis, H. F. & Terrill, C. E. (1971). Uterotropic activity of DDT in rats and mink and its influence on reproduction in (he rat. Toxic, appl. Pharmac., 18, 348-55. Dunachik, J. F. & Fli;khi:k, W. W. (1969). An investigation of the toxicity of insecticides to birds' eggs using the egg-injection technique. Ann. appl. Biol., 64, 409-23. Dvorchik, B. H., Istin, M. & Marf.n, T. H. (1971), Does DDT inhibit carbonic anhydrase? Science, N. Y,, 172, 728-9. Fl-Boushy, A. R.. Simons, P. C. M. & Wiertz, G. (1968). Structure and ultra-structure of the hen's egg shell as influenced by environmental temperature, humidity and vitamin C additions. Ponlt. Sci., 47, 456-67. Hndfrson, J. H. & Berger, D. D. (1970). Pesticides: eggshell thinning and lowered production of young in prairie falcons. BioScicnce, 20, 355-6. F.rbe:n, II. K. & Krampitz, G. (1971). Eischalen DDT-verseuchtcr Vogel: Ultrastruktur und organische Substanz. Abh. math-naturw. KL Akad. IViss. Mainz, 1971(2), 31-64. Faiifr. R. A., Risfiirougii, R. W. & Pratt, H. M. (1972). Organochlorines and mercury in common egrets and great blue herons. Environ. Pollnt., 3, 111-22, Fimreite, N. (1971). Effects of dietary methytmercury on ring-necked pheasants. Occasional Paper Number 9, Canadian Wildlife Service, Ottawa. Fimreite, N., Fyfe, R. W. & Keith, J. A. (1970). Mercury contamination of Canadian prairie seed-eaters and their avian predators. Can. Eld Nat., 84, 269-76. Fisher, A. L., Kf.asi.ino, H. M, & Schuelfr, F. N. (1952). Estrogenic action of some DDT analogues. Proc. Soc. exp. Biol. Med., 81,439 41. Fowi.fr, J. F., Newsom. L. D., Graves, J. B., Bonner, F. L. & Schilling. P. E. (1971). Effect of dieldrin on egg hatchabilily, chick survival and egg-shell thickness in purple and common gallinulcs. Bnll. environ. Contam. & Toxicol., 6, 495-501. Fox G. A. (1971). Recent changes in the reproductive success of the pigeon hawk. J. H'ildl. Mgml, 35, 122-8. Frank, F. R. & Burger, R. E. (1965). lhc effect of carbon dioxide inhalation and sodium bicarbonate ingestion on egg shell deposition. Poult. Sci., 44, 1604-6. Frank, F. R., Burger, R. E. & Swanson. M. H. (1965). The relationships among shell mem branes, selected chemical properties and the resistance to shell failure of Gallns domestiens eggs. Poult. Sci., 44, 63-9. French, M. C. & Jefferies, D. J. (1969). Degradation and disappearance of ortho, para isomer of technical DDT in living and dead avian tissues. Science, N. Y,, 165, 914-6. Fujii, S. & Tamura, T. (1970). Scanning electron microscopy of shell formation in hen's eggs. J. Ear. Fish. Anim. Husb. Hiroshima University., 9, 65-81. Fyee, R. W,, Campbell, J., IIayson, B. & Hodson, K. (1969). Regional population declines and organochlorine insecticides in Canadian prairie falcons. Can. Fid Nat., 83, 191-200. Gabuten. A. R. & Shaftnfr, C. S. (1954). A study of the physiological mechanisms affecting specific gravity of chicken eggs. Poult. Sci., 33, 47-70. Gay, C. V. & Schraer, H. (1967). Autoradiographic localization of calcium4* in the shell gland mucosa of the hen. Eedn Proc. Ecdn Am. Socs exp. Bio!., 26, 660. Genelly, R. E. & Rudd, R. L. (1956). Effects of DDT, toxaphene and dieldrin on pheasant reproduction. Auk, 73, 529-39. Gilbert, A. B. (1967). Formation of the egg in the domestic chicken. In Advances in reproductive physiology, ed. by A. MacLaren, 2, 111-80. London, Logos Press. Gilbert, A. B. (1971c/). Control of ovulation. In Physiology and biochemistry of the domestic font, ed. by D. J. Bell and B. M. Freeman, 1225-35. London, Academic Press. DSW 031349 STLCOPCB4015311 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 143 Gilbert, A. B, (19716). Transport of the egg through (he oviduct and oviposition. In Physiology and biochemistry of the domestic fowl, ed. by D. J. Bell and B. M. Freeman. 1345--52. London. Academic Press. Gil l-El r, J. W. & Arscott, G. H. (1969). Microsomal epoxidation in Japanese quail: induction by dietary dieldrin. Comp. Biochem. Physiol., 30, 589-600. Gitrss, F. Rist.brough, R. W. & Sibley, F. C. (1971). Shell thinning in eggs of the common murre Uria aalgac from the Farallon Islands, California. Condor, 73, 368-9. Guthrie, F. E. & Donaldson, W. E. (1970). Distribution of DDT and dieldrin in the avian embryo. Toxic, appl Pharmac.. 16, 475-81. Gutowska, M. S. & Mitchell, C. A. (1945). Carbonic anhydrasc in the calcification of the egg shell. Poult. Sci., 24, 159-67. Hah., K. N. & Hei.backa, N. V. (1959). Improving albumen quality. Poult. Sci., 38, 111-4, Hamilton, J. W. & Holdsworth, E. S. (1969). Calcium transport by the chick small intestine. The role of vitamin Dj. the mitochondria and binding protein. Calcif. Tissue Res , 4, 284. Hazeltine, W. (1972). Disagreements on why brown pelican eggs are thin. Nature, Loud., 239,410-1. Heald, P. J., Pohlman, D. & Martin, E. G. (1968). Shell strength and carbonic anhydrasc activity of the shell gland of the domestic fowl. Poult. Sci., 47, 858 -62. Heath, R. G., Seann, J. W. & Kreieze.k, J. F. (1969). Marked DDE impairment of mallard reproduction in controlled studies. Nature, Loud., 224, 47 -8. Heath, R. G., Spann, J. W., Kreitzer, J. F. & Vance, C. (In press). Effects of polychlorinated biphenyls on birds. Proc. inf. Orn. Congr., 15th. Heinrichs, W. L., Gellert, R. J., Bakke, J. L. & Lawrence, N. L. (1971). DDT administered to neonatal rats induces persistent eslrus syndrome. Science, N.Y., 173, 642-3. Helbi.cka, N. V., Casierune, J. L. & Smith, C. J. (1963). The effect of high carbon dioxide atmosphere on the laying hen. Poult. Sci., 42, 1082-4. Hickey, J. J. & Anderson, D. W. (1968). Chlorinated hydrocarbons and eggshell changes in raptorial and fish-eating birds. Science, N. Y., 162, 271-3. Hodges, R. D. & LOrcher, K. (1967). Possible sources of the carbonate fractions of egg shell calcium carbonate. Nature, Land., 216, 609-10. Hoiiman, W. & Schraer, H. (1966). The intracellular distribution of calcium in the mucosa of the avian shell gland. J. Cell Bio!., 30, 317-31. H6hn, E. O. (1961). Endocrine glands, thymus and pineal body. In Biology and comparative physiology of birds, ed, by A. J. Marshall. 2, 87-114. New York, Academic Press. Hunt, J. R. & Aitken, J. R, (1962a). Studies on the influence of ascorbic acid on shell quality. Poult. Sci., 41, 219-26. Hunt, J. R. & Aitken. J, R. (19626). The elTecl of ammonium and chloride ions in the diet of hens on egg shell quality. Poult. Sci., 41, 434-8. Hunt, J. R. & Simkiss, K. (1967). Acute respiratory acidosis in the domestic fowl. Comp. Biochem. Physio!., 21, 223-30. Hurst, J. G.. Newcomer, W, S. & Morrison, J. A. (1971). The effects of DDT, toxaphene and polychlorinated biphenyl on thyroid and reproductive function in bobwhite quail Colinus virginiattus. Oklahoma Coop. IVildl. Res. Unit Q. Rep., 24, 28-31. Hurwitz, S. (1970). The role of the intestine in calcium homeostasis in the laying hen. Annls Biol. auim. Biochim. Biophys., 10, (hors-serie 2), 69-76. Itoh, H. & Hatano, T. (1964). Variation of magnesium and phosphorus deposition rates during egg shell formation. Poult. Sci., 43, 77-80. Jr:Fitries, D. J. (1967). The delay in ovulation produced by pp'-DDT and its possible significance in the field, ibis, 109, 266-72. Jefferies, D. J. (1969). Induction of apparent hyperthyroidism in birds fed DDT. Nature, Land., 222, 578-9. Jefferies, D. J. (1971). Some sublethal effects of pp'-DDT and its metabolite pp'-DDE on breeding passerine birds. Meded. Fakult. Landhouwwetenschappen Gent, 36, 34-42. Jefferies, D. J, & French, M. C. (1971). Hyper- and hypothyroidism in pigeons fed DDT: an explanation for the `thin eggshell phenomenon'. Environ. Pollut., 1, 235-42. Jefferies, D. J. & French, M. C. (1972). Changes induced in the pigeon thyroid by pp'-DDE and dieldrin. J. Wildl. Mgmt, 36, 24-30. Jefferies. D. J., French, M. C. & Osborne. B. E. (1971). The effect of pp'-DDT on the rate, amplitude and weight of the heart of (he pigeon and Bengalese finch. Br. Poult. Sci., 12, 387-99. Jefferies, D. J. & Prf.stt, I. (1966). Post-mortems of peregrines and lanncrs with particular reference to organochlorine residues. Br. Birds, 59, 49-64. Keith, J. A. (1966). Reproduction in a population of herring gulls (Lotus argentatus) contami nated by DDT. In Pesticides in tlse environment and their effects on wildlife, cd. by N. W. Moore, J. appl. Ecol., 3 (Suppl.). 57-70. OSW 031350 STLCOPCB4015312 144 A. S. COOKE Keith, J. O., Woods, L. A. & Hunt, E. G. (1970). Reproductive failure in brown pelicans on the Pacific Coast. Trans. N. Artier. Wildl. Conf., 35, 56-64. Keller, H. (1952). Die Bestimmung kleinstcr Mengen DDT auf enzymanalytischem Wege. Natunvissciischaften, 39, 109. Koch. R. IT, Cutkomp, L. K. & Do, F. M. (1969). Chlorinated hydrocarbon insecticide inhibition of cockroach and honey bee ATPascs. Life Sci.. 8, 289-97. Koeman, J. 11., Oskamp, A. A. G., Veen, J., Brouwer, E., Rooth, .1., Zw'art, P., van oer Broek, E. & van Genderen, H. (1967). Insecticides as a factor in the mortality of the sandwich tern {Sterna sandvicensis). A preliminary communication. Meded. Fakult. Landbouwweten- schappen Cent, 12, 841-54. Krmtzer, J. F. (1971). Eggshell thickness in mourning dove populations. J. Wild!. Mgmt, 35, 563-64. . ,_ Kuntzman, R., Jacobson, M., Schnfidman, K. & Conney, A, II. (1964). Similarities between oxidative drug-metabolizing enzymes and steroid hydroxylases in liver microsomcs. J. Pharmae. exp. Titer., 146, 280-5. Kupier, D. (1967). Effects of sonic pesticides and related compounds on steroid function and metabolism. Residue Rev., 19, 11-30. Kyes, P. & Potter, T. S. (1934). Physiological marrow ossification in female pigeons. Anal. Rcc., 60, 377-9. Lehner, P. N. & Egbert, A. (1969). Dicldrin and eggshell thickness in ducks. Science, N. K, 224, 121R-9. Lf.vin, W., Welch, R. M. & Conney, A. II. (1968). Estrogenic action of DDT and its analogs. Fedn Proc. Fedn Am. Sacs exp. Biol., 27, 649. Lillie, R. J., Denton, C. A., Cecil, H. C., Bitman, J. & Fries, G. F. (1972). Effect of p,p'-DDT, op'-DDT and pp'-DDE on the reproductive performance of caged white leghorns. Poult. Sci., 51, 122-9. Lincer, J. L. & Peakall, D. B. (1970). Metabolic effects of polychlorinated biphenyls in the American kestrel. Nature, Land., 228, 783. Linduska, J. P. & Springer, P. F. (1951). Chronic toxicity of some new insecticides to bob-white quail. U.S. Dept Interior, Spec. Sci. Rep. IVildl., 9. (Mimeograph), Lofts, B. & Murton, R. K. (1966). The role of weather, food and biological factors in timing the sexual cycle of woodpigeons. Br. Birds, 59, 261-80. Lonocore, J. R., Samson, F. B., Kreitzer, J. F. & Spann, J. W. (1971a). Changes in mineral composition of eggshells from black ducks and mallards fed DDE in the diet. Bull, environ. Contain. & Toxicol., 6, 345-50. Longcore, J. R., Samson, F. B. & Whitiendale, T. W. (19716). DDE thins eggshells and lowers reproductive success of captive black ducks. Bull, environ. Contain. & Toxicol., 6, 485-90. Lonostaff, M. & Hill, R. (1971). The influence of dietary manganese and stage of egg production on the hexosamine and hydroxyproline contents of parts of the oviduct. Br. Poult. Sci., 12, 401-11. Lonostaff, M. & Hill, R. (1972). The hexosamine and uronic acid contents of the matrix of shells of eggs from pullets fed on diets of different manganese content. Br. Poult. Sci., 13, 377-85. LOrcher, K., Diamantstein, T. & Kobow, J. (1964). Dcr Einfluss von Chloridioncn auf die Eischalenstabililat. Z. Tierphysiol. Ticrcrnohr. Futtermittelk., 19, 228-36. Ludwig, J. P. & Tomoff, C. S. (1966), Reproductive success and insecticide residues in Lake Michigan herring gulls. Jack Pine Warbler, 44, 77-84. Mann, T. & Keii.in, D. (1940). Sulphanilamide as a specific inhibitor of carbonic anhydrase. Nature, Land., 146, 164-5. Maren, T. H. (1967). Carbonic anhydrase: chemistry, physiology and inhibition. Physiol. Rev., 47, 595-781. Marshall, J. S. & Tompkins, L. S. (1968). Effect of o.p'-DDD and similar compounds on thyroxine binding globulin. J. din. Endocr. Metal)., 28, 386-92. McFarland, L. Z., Garrett, R. L. & Nowell, J. A. (1971). Normal eggshells and thin eggshells caused by organochlorine insecticides viewed by the scanning electron microscope. Proc. Ann. Scan. Elect. Micr. Symp., 4, 377-84. McNally, E, H. (1965). The relationship of eggshell weight to cracked eggs. Poult. Sci., 44,1513-8. Metcalf, R. L. (1955). Organic insecticides. Their chemistry and mode of action. New York, Interscience Publishers Inc. Mongin. P. (1970). Etude de 1'anhydrase carbonique chex les oiseaux. Aunts Biol. anim. Biocliim. Biophys., 10, (hors-s6rie 2), 119-30. Mongin, P. & Lacassagnf., L. (1966). Rythmc respirntoire et physiologie de la formation de la coquillc de j'oeuf. Annls Biol. anim. Biocliim. Biophys., 6, 101-11. DSW 031351 STLCOPCB4015313 SMELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 145 Monoin. P. & Saveur, B. (1970). Composition du fluidc utcrin c dc I'albumcn durant le sejonr dc 1'oeuf dans I'uterus chez la poule domeslique. C. r. Acad. Sc. Paris, 270 (Serie D), 1715-8. Monsanto Company (1971). Toxicity, reproduction and residue study on: Aroclor 1242, Aroclor 1254, Aroclor 1260 and supplementary study on Aroclor 1242 in white leghorn chickens. Two reports by Industrial Bio-Test Laboratories, Inc. (Mimeograph.) Moore, N. W. (1965). Pesticides in birds--a review of the situation in Great Britain in 1965. Bird Study, 12, 222-52. Moore, T. (1967). Effects of vitamin A deficiency in animals. In The vitamins, ed. by W. M, Scbrell & R. S. Harris. 1,245-66. New York, Academic Press. Mueller, W. J. (1959). The effect of environmental temperature and humidity on the calcium balance and serum calcium of laying pullets. Poult. Sci., 38, 1296-1301. Muei.lf.r, W. J. (1962). Carbonic anhydrase, diuretics and egg shell formation. Poult. Sci., 41, 1792-6. Mueller. W. J., Brubaker, R. L. & Caplan. M. D. (1969). Egg shell formation and bone resorption in laying hens. Fedn Proc. Fedn Am. Socs exp. Biol., 28, 1851-6. Muller, H, D. (1971a). Reproductive responses of the mallard duck to subloxic pesticide inges tion. Poult. Sci., 50, 1610. Muller, H. D. (19716). Responses of three generations of gray partridge to low level pesticide ingestion. Poult. Sci., 50, 1610 -1. Muller, H. D. & Lockman, D. C. (1971). The influence of dicldrin on bone calcium deposition in domestic chicks. Poult. Sci., 50. 1611. Noakes, D. N. & Benfield, C. A. (1965). Tissue accumulation of DDT and its metabolites in the domestic fowl. J. Sci. Fd Agric., 16, 693-7. Nowicki, H. G., Myrtle, J. F. & Norman, A. W. (1972a). EITects of organochlorine insecticides on metabolism of cholecalciferol (vitamin D3) in rachitic cockerel. J. agric. Fd Chem., 20, 380-4. Nowicki, H. G., Wong, R. G., Myrtle, J. F, & Norman, A. W. (19726). Inhibition of biological activity of cholecalciferol (vitamin Dj) by op'-DDT or pp'-DDT in rachitic cockerel. J. agric. Fd Chem., 20, 376-80. Odsjo, T. (1971). Klorerade Kolvaten och aggskalsfortunning hos fiskgjuse. Fauna Flora, Upps., 66, 90-100. Oestrficher. M. I., Shuman, D. H. & Wurster, C. F. (1971). DDE reduces medullary bone formation in birds. Nature, Load., 229, 571. Opel, H. (1965). Failure of neurohypophyseclomy to reduce egg shell thickness in chickens. Poult. Sci., 44, 1135. Peakall, D. B. (1967). Pesticide-induced enzyme breakdown of steroids in birds. Nature, Land., 216, 505-6. Pf.akall, D. B. (1969). Effect of DDT on calcium uptake and vitamin D metabolism in birds. Nature, Lond., 224, 1219-20. Peakall, D. B. (1970a). Pesticides and the reproduction of birds. Scient. Am., 222, 73-8. Peakall, D. B. (19706). p,p'-DDT: effect on calcium metabolism and concentration of estradiol in the blood. Science, N.Y., 168, 592-4. Peakall, D. B. (1971). Effect of polychlorinated biphenyls (PCB's) on the eggshells of ring doves. Bull, environ. Contam. & Toxicol., 6, 100--1. Peakall, D. B. & Linger, J. L. (1970a). Polychlorinated biphenyls. Another long-life widespread of chemical in the environment. BioSciencc, 20, 958-64. Peakall, D. B. & Lincf.r, J. L. (19706). Effect of chlorinated hydrocarbons on active transport of calcium in the avian oviduct. Am. Zoo!., 10, 515. Platonow, N. S. & Funnell, H. S. (1971). Anti-androgenic-like effect of polychlorinated biphenyls in cockerels. Vet. Rec., 88, 109-10. Pocker, Y., Beug, W. M. & Ainardi, V. R. (1971), Carbonic anhydrase interaction with DDT, DDE and dieldrin. Science, N.Y., 174, 1336-9. Porter, R. D. & Wiemeyer, S. N. (1969). Dieldrin and DDT: effects on sparrow hawk eggshells and reproduction. Science, N.Y., 165, 199-200. Prestt, I. (1965). An enquiry into the recent breeding status of some of the smaller birds of prey and crows in Britain. Bird Study, 12, 196-221. Prestt, I. (1970). Organochlorine pollution of rivers and the heron (Ardea cinerea L.). Tech. Meet. int. Un. Cons. Nat. nat. Resour., 11th, New Delhi, 1, 95-102. Prestt. 1., Jefferies, D. J. & Moore, N. W. (1970). Polychlorinated biphenyls in wild birds in Britain and their avian toxicity. Environ. Pollut., 1, 3-26. Ratcliffe, D. A. (1967). Decrease in eggshell weight in certain birds of prey. Nature, Lond., 215, 208-10. OSW 031352 STLCOPCB4015314 146 A. S. COOKE Rate lime, D. A. (J970). Changes attributable to pesticides in egg breakage frequency and eggshell thickness in some British birds. J. appl. Ecol., 7, 67-115. Rf.vz.in, A. M. (1966). Effects of endrin on tclcncephalic function in the pigeon. Toxic, appl. Pharmac., 9, 75--83. Richfri, E. P. & Prahlad, K. V. (1972). Effects of DDT and its metabolites on thyroid of the Japanese quail, Commix cotwnix japonica. Poult. Sci., 51, 196-200. Risfiirough, R. W.. Davis, J. & Andlrson, D. W. (1970). Effects of various chlorinated hydro carbons. Oregon State Univ. Environ. Health Sci., Scries No. 1, 40-53. Risehrough. R. W,, Riechf, P., Peakall. D. B., Herman, S. G. & Kirven, M. N. (1968). Poly chlorinated biphenyls in the global ecosystem. Nature, Loud., 220, 1098 -102. Risfbrough, R. W., Sibley, F. C. & Kirven, M. N. (1971). Reproductive failure of the brown pelican on Anacapa Island in 1969. Am. Birds, 25, 8-9. RoniNSON, D. S. & King, N. R. (1963). Carbonic anhydrase and formation of the hen's egg shell. Nature. Load., 199, 497-8. Rohinson, D. S. & KtNG, N. R. (1968). Mucopolysaccharides of an avian egg shell membrane. J! R. micro.sc. Soc., 88, 13-22. Robinson, D. S. & King, N. R. (1970). The structure of the organic mammillary cores in some weak egg shells. Br. Poult. Sci., 11, 39-44. Robinson, D. S., King, N. R. & Bowen, D. J. (1966). Neutral and acidic mucins in the reproductive tract of the laying hen. Proc. R. micro.sc. Soc., I, 213-4. Robinson. D. S., King, N. R. & Bowen, D. J. (1968). The occurrence of neutral and acidic mucins in the reproductive tract of the laying hen. Histochemie, 13, 97-104. Robinson, J. (1970). Birds and pest control chemicals. Bird Study, 17, 195-228. Rogers, A. J. (1972). Eagles, affluence and pesticides. Mosquito Ncsvs, 32, 151-7. Romanoff, A. L. & Romanoff, A. J. (1949). The avian egg. New York, John Wiley & Sons, Inc. Sauter, E. A. & Steele, E. E. (1972). The effect of low- level pesticide feeding on the fertility and halchability of chicken eggs. Poult. Sci., 51, 71-6. Scihmidt-nielsen. K. & Fangf, R. (1958). The function of the salt gland in the brown pelican. Auk, 75, 282-9. . Schmidt, W. J. (1957). Ctbcr den Aufbau der Schale des VogeleiesncbstBemerkungen fiber kalkige F.ischalen andercr Ticre. Ber. oberhess. Gcs. Nat. u. Heilk. (N.F.), 28, 82-108. Schmidt, W. J. (1962(7). Obcr die Basalkalotlcn der Vogeleischale. J. Orn. Lpz., 103, 28-37. Schmidt, W. J. (19626). Liegt der F.ischalenkalk der Vogel als submikroskopischc Kristallitc vor? Z. Zellforsch, microsk. Anal., 57, 848-80. Schraer, H., Hohman, W.. Ehrenseeck, G. & Schraer, R. (1965). Subcellular distribution of calcium in the avian shell gland in relation to calcium transport. J. Cell Biol., 27, 96A. Schraer, H. & Schraer, R. (1970). Calcium translocation in the avian shell gland. Annls Biol, anitn. Biochim. Biophys., 10 (hors-serie 2). 229-44. Scott, El. M., Junghfrr. E. & Matierson. L. D. (1944). The effect of feeding sulphanilamidc to the laying fowl. Poult. Sci., 23, 446-53. Seidenstickf.r, J. C. & Reynolds, H. V. (1971). The nesting, reproductive performance, and chlorinated hydrocarbon residues in the red-tailed hawk and great horned owl in south-central Montana. Wilson Bull., 83, 408-18. Sell, J. L., Davison, K. L. & Poonacha, K. B. (1972). Decreased aniline hydroxylase activity in Japanese quail due to dietary DDT. J. agric. Ed Chem., 20, 553-7. Siegmund, P. & Dulce, H.-J. (I960). Zur Biochemic der Knochenauflosung, I. Einfluss des Carboanhydralase-Inhibitors 2-Acetamino-1.3.4-thiodiazolsulfonamid-(5)(Diamox) auf den Calciumstoffwechse! von Legehcnnen. Hoppe-Seyler's Z. physiol. Chem., 320, 149-59. Simkiss, K. (1961). Calcium metabolism and avian reproduction. Biol. Rev., 36, 321-67. Simkiss, K. (1964). Phosphates as crystal poisons of calcification. Biol. Rev., 39, 487-505. Simkiss. K. (1965) The organic matrix of (he oyster shell. Comp. Biochem. Physio!., 16, 427-35. Simkiss, K. (1967). Calcium in reproductive physiology. London, Chapman and Hall. Simkiss, K. (1968). The structure and formation of the shell and shell membranes. In Egg quality: A study of the hen's egg, ed. by T. C. Carter, 3-25. Edinburgh, Oliver and Boyd. Simkiss, K. (1969). Intracellular pH during calcification. A study of the avian shell gland. Biochem. J., Ill, 647-52. Simkiss, K. (1970). Effect of acctazolamide on intracellular pH of avian shell gland. J. Physiol., Load., 207, 63-4P. Simkiss. K. & Dacke, C. G. (1971). Ultimobranchial glands and calcitonin. In Physiology and biochemistry ofthe domestic fowl, ed. by D. J. Bell & B. M. Freeman, 481-8. London, Academic Press. Simkiss, K. & Taylor, T. G. (1971), Shell formation. In Physiology and biochemistry ofthe domestic fowl, ed. by D. J. Bell & B. M, Freeman, 1331-43. London, Academic Press. GSW 031353 STLCOPCB4015315 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 147 Simkiss, K. & Tyi.fr, C. (1957). A hisiochcmical study of the organic matrix of lien egg-shells. Q. Jl microsc. Sci., 98, 19 -28. Simkiss, K. & Tylf.r, C. (1958). Reactions between eggshell matrix and metallic cations. Q. Jl microsc. Sci., 99, 5-13. Simkiss, K. & Tyler, C. (1959). The possible calcification mechanisms in sonic reptilian egg shells Q. Jl microsc. Sci.. 100, 529 -38. Simons, P. C. M. (1971). Ullrastruclure of the hen egg shell anil its physiological interpretation. Wagcningcn, Centre for Agricultural Publishing and Documentation. Simons, P. C. M., Tyler, C. & Thomas, H. P. (1966). The effect of sodium hydroxide and sodium sulphide on the snapping strength of egg shells. Br. Poult. Sci., 7, 309--14. Simons, P. C. M. & Wie.riz, G, (1963). Notes on the structure of membranes and shell in the hen's egg. Z. ZcUforsch. microsk. Anal., 59, 555-67. Singhai., R. L., Valadares, J. R. E. & Schwark, W. S. (1970). Metabolic control mechanisms in mammalian systems-lX. Fstrogcn-likc stimulation of uterine enzymes by op'-DDT. Biochem. Pharmac., 19, 2145-55. Smith, S. I., Weber, C. W. & Reid, IT L. (1970). Dietary pesticides and contamination of yolks and abdominal fat of laying hens. Poult. Sci., 49, 233-7. Snetsinger, D. C., Paul, H. S. & Mraz, F. R. (1966). Variations in location and quantity of selected mineral elements in egg shells. Poult. Sci., 45, 1125. Soiifi , A. E. & FIanok, A. (1952). Calcification. VII, Reversible inactivation of calcification in vitro and related studies. J. Biol. Client., 197, 669-85. Sri.iioc an, V., Gotal, J. P., Adamovig, V., Sokic, B. & Df.lak, M. (1971). Effect of technical grade DF)T and p,p'-DDT on adrenocortical functions in chicks. Poult. Sci., 50, 1271-9. Stephen, B. J,, Gari.ich, J. D. & Guthrie, F. F. (1970). Effect of DDT on induction of microsomal enzymes and deposition of calcium in the domestic chicken. Bull, environ. Contain & Toxicol., 5, 569-76. Stickel, L. F. & Rhodes, L. I. (1970). The thin eggshell problem. Oregon State Univ. Environ. Health Sci., Scries No. 1, 31-35. Stoewsand, G. S.. Anderson, J. L., Guteman, W. H., Bache, C. A. & Lisk, D. J. (1971). Eggshell thinning in Japanese quail fed mercuric chloride. Science, N.Y., 173, 1030-1. StuKKII!, P. D. (1965). Avian physiology. London, Baillidrc, Tindall & Cassell. Suzuki, S. & Strominger, J. L. (1959). Enzymatic synthesis of sulfated mucopolysaccharides in hen oviduct. Biochim. hiophys. Acta, 31, 283 -5. Suzuki, S. & Strominger, J. L. (I960n). Enzymatic sulfation of mucopolysaccharides in hen oviduct. I. Transfer of sulfate from T-phosphoadenosine 5'-phosphosuIphate to mucopoly saccharides. J. biol. Chcm., 235, 257-66. Suzuki. S. & Strominger, J. L. (19606). Enzymatic sulfation of mucopolysaccharides in hen oviduct. 11. Mechanism of the reaction studied with oligosaccharides and monosaccharides as acceptors. J. biol. Client., 235, 267-73. Suzuki, S. & Strominger, J. L. (1960c). Enzymatic sulfation of mucopolysaccharides in hen oviduct. III. Mechanism of sulfation of chondroitin and chondroitin sulfate A. J. biol. Client., 235, 274-6. Switzer, B., Le:w'IN, V. & Wot.rr:, F. H. (1971). Shell thickness, DDF. levels in eggs, and repro ductive success in common terns (Sterna hirundo). in Alberta. Can. J. Zoo!., 49, 69-73. Tai.f.kar. N. S. (1971). DDT and reproduction in birds. Pestic., Bombay, 5, 11-20. Taylor, L. W. & Burmester, B. R. (1940). Effect of thyroidectomy on production, quality and composition of chicken eggs. Poult. Sci., 19, 326-31. Tayi.or, T. G. (1966). The endocrine control of calcium metabolism in the fowl. In Physiology of the domestic fowl, cd. by C. Horton-Smith and E. C. Amoroso, 199 -202. Edinburgh, Oliver & Boyd. Tayi.or, T, G. (1970). How an eggshell is made. Scient. Am., 222, 88-95. Taylor. T. G, & Hertelf.ndy. F. (1961). Changes in the blood calcium associated with egg shell calcification in the domestic fowl. 2. Changes in the diffusible calcium. Poult. Sci., 40, 115-23. Taylor, T. G. & Moore, J. H. (1953). Avian medullary bone. Nature, Loud., 172, 504-5. Taylor, T. G. & Moore, J. H. (1956). The effect of calcium depletion on the chemical composi tion of bone minerals in laying hens. Br. J. Nutr.. 10, 250-63. Taylor, T. G,, Morris, T. R. & Herteiendy, F. (1962). The effect of pituitary hormones on ovulation in calcium deficient pullets. Vet. Pee., 74, 123-5. Taylor, T. G. & Stringer, D. A. (1965). Eggshell formation and skeletal metabolism. In Avian physiology, ed. by P. Sturkie, 485 -514. London, Bailli6rc, T indall & Cassell. Tejning. J. (1967). Biological effects of methyl mercury dicyandiamide-treated grain in the domestic fowl Callus galltts L. Oikos, Suppl., 8, 1--116. DSW 031354 * I STLCOPCB4015316 148 A. S. COOKE Thornton, P. A. & Morfng, R. E, (1959). Further evidence on the value of ascorbic acid for main- tcnancc of egg shell thickness at increased environmental temperatures. Poult. Sci., 38, 594-9. Torda, C. & Wolff, H. G. (1949). Effect of convulsant and anticonvulsant agents on the activity of carbonic anhydrase. J. Pharmac. exp. Ther., 95, 444 -7. Tuck, L. M. (1960). The Murres. Ottawa. Canadian Wildlife Service. Tucker, R. K. & Haf.gele, M. A. (1970). Eggshell thinning as influenced by method of DDT exposure. Bull, environ. Contam. & Toxicol.. 5, 191-4. Tucker, R. K. & Haegelf., M. A. (1971). Comparative acute oral toxicity of pesticides to six species of birds. Toxic, appl. Pharmac.. 20, 57-65. Turner. C-. W., Irwin, M. R. & Reinekf, E. P. (1945). Effect of the thyroid hormone on egg production of white leghorn hens. Poult. Sci., 24, 171-80. Tvi.fr . C. (1950). The effect of sulphanilamide on the metabolism ofcalcium, carbonate, phosphorus, chloride and nitrogen in the laying hen. Br. J. Nutr., 4, 112-28. Tylf.r, C. (1954). A further study of the effect of sulphanilamide on the metabolism of calcium and phosphorus in the laying hen. J. agric. Sci., Camb., 45, 156-63. Tyler, C. (1956). Studies on egg shells. VII. Some aspects of structure as shown by plastic models. J. Sci. Fd Agric., 7, 483 93. Tyler, C. (1961). Studies on egg shells. XVI. Variations in shell thickness over different parts of the same shell. J. Sci. Fd Agric.. 12, 459-70. Tyler, C. (1965). A study of the egg shells of the Sphenisciformes. J. Zoo!., Land., 147, 1-19. Tylf.r, C. (1969). A study of the egg shells of the Gaviiformcs, Procellariiformes, Podicipiti- formes and Pelecaniformcs. J. Zoo!., Load., 158, 395-412. Tylf.r, C. & Geake, F. H. (1960). Studies on egg shells. XIII. Influence of individuality, breed, season and age on certain characteristics or egg shells. J. Sci. Fd Agric., 11, 535-47. Tyler, C. & Geake, F. H. (1961). Studies on egg shells. XV. Critical appraisal of various methods of assessing shell thickness. J. Sci. Fd Agric., 12, 281-9. Tylfr, C. & Simkiss, K. (1958). Studies on egg shells. VIII. Determination of membrane-, pore- and matrix-protein. A reassessment. J. Sci. Fd Agric., 9, 375-80. Tylf.r, C. & Simkiss, K. (1959). Studies on egg shells. XI. Some abnormalities. J. Sci. Fd Agric., 10, 362-6. Urist, M. R. & Deutsch, N. W. (1960). Effects of cortisone upon blood, adrenal cortex, gonads and the development of osteoporosis in birds. Endocrinology, 66, 805-18. Vanderstoep, J. & Richards. J. F. (1971). The changes in egg shell strength during incubation. Poult. Sci., 49, 276-85, Vermeer, K. & Reynolds, L. M. (1970). Organochtorine residues in aquatic birds in the Canadian prairie provinces. Can. Fid Nat., 84, 117-30. Vos, J. G,, van der Maas, H. L., Musch. A. & Ram, E. (1971). Toxicity of hexachlorobcnzene in Japanese quail with special reference to porphyria, liver damage, reproduction and tissue residues. Toxic, appl. Pharmac., 18, 944-57. Wassf.rman, R. H. (1968). Calcium transport by the intestine: a model and comment on vitamin D action. Calcif. Tissue Res., 2, 301-13. Wasserman, R. H. & Taylor, A. N. (1966). Vitamin D3-induced calcium-binding protein in chick intestinal mucosa. Science, N.Y., 152, 791-3. Watabe, N. & Wilrur, K. M. (1960). Influence of the organic matrix on crystal type in molluscs. Nature. Land.. 188, 334. Welch, R. M., Lf.vin, W. & Conney, A. H. (1969). Estrogenic action of DDT and its analogs. Toxic, appl. Pharmac., 14, 358-67. Whitehead, C. C.. Downing, A. G. & Pettigrew, R. J. (1972). The effects of lindane on laying hens. Br. Poult. Sci., 13, 293-9. Whitehouse. M. W. & Lash, J. W. (1961). Effect of cortisone and related compounds on the biogenesis of cartilage. Nature, Land., 189, 37-9. Wiemeyer, S. N. & Porter, R. D. (1970). DDE thins eggshells of captive American kestrels. Nature, Lond., 227, 737-8. Wilbur, K. M. & Simkiss, K. (1968). Calcified shells. In Comprehensive biochemistry, ed. by M. Florkin and E. Stotz, 26A, 229-95. Wilbur, K. M. & Watabe, N. (1967). Mechanisms of calcium carbonate deposition in coccolitho- phorids and molluscs. Proc. hit. Conf. Trop. Oceangr., 5, 133-54. Wilson, W. O. (1949). High environmental temperatures as affecting the reaction of laying hens to iodized casein. Poult. Sci., 28, 581-92. Wrenn, T. R., Weyant, J. R., FKits, G. F. & Bitman, J. (1971), Influence of dietary o.p'-DDT on reproduction and lactation of ewes. J. Anim. Sci., 33, 1288-92, Young, H. (1968). A consideration of insecticide effects on hypothetical avian populations. Ecology, 49, 991-4. I | DSW 031355 STLCOPCB4015317 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANTS 149 NOTE ADDED fN PROOF Since this paper was submitted for publication, there has been a heated debate on the thin egg shell problem in the journal Nature. Switzer et al. (1972) (in addition to Hazelline, 1972; see p. 90) have been critical of some of the laboratory and field evidence, while Wieineyer & Porter (1972), Risebrough (1972) and Blus et at. (1972a) have answered these criticisms. Switzer et al. (1972) maintained that Wicmeyer & Porter (1970) had not, after all, demonstrated that DDE caused egg shell thinning in the American kestrel. They also claimed that these kestrels did not contain residue levels typical of those found in the field. Both of these objections were answered satisfactorily by Wiemeyer & Porter (1972). Residues in the birds or the eggs in most of the controlled experi ments shown in Table 1 of the main text were probably comparable to residues found in some predatory species in Britain or North America (e.g. see Ludwig & Tomolf, 1966; Risebrough et al., 1968; Anderson et al., 1969; Fyfe et al., 1969; Keith et al., 1970; Presit, 1970; Ratcliffe, 1970; Vermeer & Reynolds, 1970; Cade et al., 1971). As pointed out on p. 100, even the high tissue levels induced in quail by Bilman et al. (1969) are sometimes encountered in the field. Switzer et al. (1972), like Hazeltine (1972), were critical of the misuse of the correlation coefficient by Blus et al. (1972) but this does not invalidate Blus' conclusions. Replying to Hazeltine (1972) and Switzer et al. (1972), Blus et al. (1972a) pointed oul that, for their 49 brown pelican eggs collected from Florida, there was, indeed, a highly significant negative relationship between DDE content and shell thickness. Previously, after examining the South Carolina and California data of Blus et al. (1972), but for some reason completely ignoring the Florida data, Hazeltine (1972) had stated that significant within-state relationships could not be found. Switzer et al. (1972) believed that comparing museum egg shells with shells collected recently from the field could be misleading, since oologists may have favoured eggs with thicker shells. This, they suggested, may be why pelican eggs collected recently in South Carolina and Florida have thinner shells than museum eggs. However, despite using a sampling technique that introduced bias towards the collection of thicker shells, Schreiber & Risebrough (1972) found that eggs in Florida colonies in 1969 and 1970 had shells 9% thinner on average than those taken prior to 1943 (data of Anderson & Hickey, 1970). Blus et al. (1972a) also pointed out that two investigations (Ratcliffe, 1967; Hickey & Anderson, 1968) have depended entirely upon the measurement of shells in museums or private collections. The part-incubated eggs upon which Hazeltine (1972) based much of his attack on Blus et al. (1972) were apparently not incubated after all. Blus et al. (1972a) quoted J. O. Keith as stating that, in each of the eggs Hazeltine believed to be DSW 031356 150 A. S. CCOKE incubated, the vitelline membrane around the yolk had burst when the fresh egg was deep frozen and the yolk had become contaminated with albumen. In a short amendment Hazeltine (1972a) was less explicit about his mistake: `Recently disclosed errors in the original laboratory reports invalidate the conclusions of the ten-fold lipid and DDE residue decline with incubation'. He then continued that the positive relationship between lipid DDE and shell thickness was still valid, which is true. It is, however, difficult to understand why he then claims that this `refutes the contention of cause and effect between egg residue and shell thickness'. His data, based on only nine eggs, may be inconsistent with the theory, but they certainly do not refute it. Earlier he had analysed Risebrough's data, based on 65 eggs from Anacapa, and had found a significant negative relationship between lipid DDE and shell thickness (Hazeltine, 1972). He did not comment on this finding, which contradicted his own views, and he also failed to point out that his eggs, which contained on average 1040 ppm DDE in the yolk lipid, had shells on average 34% thinner than pre-1943 Californian shells examined by Anderson & Hickey (1970). Risebrough (1972) and Blus et al. (1972a) presented additional information which should convince anyone making an objective assessment of the problem that a relationship exists between thinning in brown pelican egg shells and egg DDE content. Blus et al. (1971, 1972) did in fact use part-incubated eggs in their studies {see p. 90). However, since a correction was made for the loss in weight of egg contents during incubation and since DDE loss as the eggs were incubated was likely to be negligible (Abou-Donia & Menzel, 1968; Guthrie & Donaldson, 1970; Cooke, 1971), the estimated DDE content of the egg must have been close to the actual pre-incubation content. Switzer et al. (1972) have been accused by Wiemeyer & Porter (1972) of choosing `to ignore obvious evidence' and Risebrough (1972) has made similar criticisms of Hazeltine (1972). Indeed, Risebrough (1972) was sufficiently exasperated to comment, `One begins to suspect intrigue and conspiracy'. The effect of papers such as Hazeltine's (1972) on public or even informed opinion should not be underestimated. On 18 October 1972, five days after Hazeltine's article was pub lished, it was featured by The Times newspaper under the heading `Pesticides: exploding the DDE myth'. The article began, `The belief that DDT residues make birds' egg shells dangerously thin is a myth'. In fact none of these papers changes in any way the conclusions listed at the end of the main text (pp. 138 to 140). Gunn (1972) deduced from the figures of RatclilTe (1967) that the shell thinning phase was completed for the peregrine by 1947 and for the sparrowhawk by 1948. Since Gunn's researches indicated that the use of DDT was very restricted at this time, he concluded that DDT could not have been responsible for shell thinning. Why he did not examine Ratcliffe's later, more comprehensive, graphs (1970) instead of those given in the 1967 paper is not clear, the later paper being included in Gunn's reference list but not in his text. In fact shell thickness reached a `typical' low level for the peregrine in 1948 and for the sparrowhawk in 1949 or 1950, DSW 031357 STLCOPCB4015319 SHELL THINNING IN AVIAN EGGS BY ENVIRONMENTAL POLLUTANIS 151 although shells of the latter species were at their thinnest in 1954 and 1955. After the war, 1950 was the first year in which every sparrowhawk clutch collected con tained thin shells (Ratcliffe, 1970). Ratcliffe (1970) discussed the early uses of DDT at great length and concluded that, in the peregrine, DDT sprayed or dusted to control ectoparasites on homing pigeons, the peregrine's principal prey south of the Scottish Highlands, could have caused shell thinning as early as 1946. This particular use was not mentioned by Gunn (1972). The early contamination of sparrowhawks is thought to have resulted from horticultural uses of DDT (Ratcliffe, 1970). There was an intensive survey of the British peregrine population in 1971 (Ratcliffe, 1972). An increase has occurred since 1963, but the population is still well below the pre-war level. The main cause of breeding failure was clutch depletion, thought to be due to deliberate shell breakage by the sublethally poisoned parents rather than to accidental breakage of the thin shells (see p. 135). The Montrose Chemical Corporation, the only manufacturer of DDT in the United States, has made a significant contribution to DDT pollution in Californian coastal waters by discharging effluent via the Los Angeles County sewer into the sea (Burnett, 1971). This effluent is believed to be the main source of the residues accumulated by the Anacapa pelicans (Schreiber & Risebrough, 1972). Several more laboratory studies have been published recently. Foster et al. (1972) fed White Leghorn chickens for six weeks on a diet containing 01 ppm technical DDT + 01 ppm chlordane + 01 ppm lindane + 0-5 ppm ethion -f 0-5 ppm atrazine + 0-5 ppm linuron. In contrast to the observations of Sauter & Steele (1972; see pp. 96 and 97), shell thickness was found to be unaffected. Screech owls Otus asio fed on a diet containing 10 ppm DDE (dry weight) laid eggs with shells 13% thinner than those of control owls (McLane & Hall, 1972). Thus this seems to be one of the most sensitive species yet tested (see Table 1). Methyl mercury injected or given orally to ring doves (10 mg/kg) or fed to American kestrels (100 ppm in diet) had no effect on shell thickness (Peakall & Lincer, 1972). Previously mercury compounds given to gallinaceous species have been found to reduce shell thickness or increase the incidence of soft shelled eggs (Tejning, 1967; Fimreite, 1971; Stoewsand et a!., 1971). Whitehead et al. (1972a) have extended their yBHC studies (Whitehead et a!., 1972). A level of 200 mg/kg in the diet of chickens reduced egg production by 30%, but the shells were again of normal thickness and scanning electron microscopy revealed no structural abnormalities. The X-ray diffraction study of Gould (1972) demonstrated no obvious differences in the relative amounts of the calcium carbonate polymorphs present in brown pelican shells of different thicknesses. Although there were appreciable amounts of aragonite and vaterite in most of the pelican shells examined, the mineral phase in chicken shells is almost entirely calcite (Cain & Heyn, 1964), showing once again the gulf that exists between chickens and field species. DSW 031358 STLCOPCB4015320 152 A. S. COOKE ADDITIONAL REFERENCES Ahou-Donia, M. B. & Menzf.l, D. B. (1968). The metabolism in vivo of l,l,l-trichloro-2,2-bis(p- chlorophenyl) ethane (DDT), l,l-dichloro-2,2-bis(p-chlorophenyl) ethane (DDD) and l,I-dichloro-2,2-bis(p-chlorophcnyl) ethylene (DDE) in the chick by embryonic injection and dietary ingestion. Biochem. Pharmac., 17, 2143-61. Blos, L. J., Gish, C. D., Belisle, A. A. & Prouty, R. M. (1972a). Further analysis of the logarithmic . relationship of DDE residues to eggshell thinning. Nature, Load., 240, 164-6. Burnett, R. (1971). DDT residues: distribution of concentrations in Emerita analoga (Stimpson) along coastal California. Science, N.Y., 174, 606-8. Cain, C. J. & Heyn, A. N. J. (1964). X-ray diffraction studies of the crystalline structure of the avian egg shell. Biophys. J,, 4, 23-39. Foster, T. S., Morley, H. V., Purkayastha, R., Greenhalgh, R. & Hunt, J. R, (1972). Residues in eggs and tissues of hens fed a ration containing low levels of pesticides with and without charcoal. J. econ. Ent., 65, 982-8. Gould, R. W. (1972). Brown pelican eggshells: X-ray diffraction studies. Bull, environ. Contain. & Toxicol., 8, 84-8. Gunn, D. L. (1972). Dilemmas in conservation for applied biologists. Ann. appl. Biol., 72, 105 -27. Hazeltine, W. (1972a). Nature, Lond., 240, 166. McLane, M. A. R. & Hall, L. C. (1972). DDE thins screech owl eggshells. Bull, environ. Contain. A Toxicol., 8, 65-8. Peakall, D. B. & Lincer, J. L. (1972), Methyl mercury: its effect on eggshell thickness. Bull, environ. Contam. & Toxicol., 8, 89-90. Ratcliefe, D. A. (1972). The peregrine population of Great Britain in 1971. Bird Study, 19,117-56, Risebrough, R. W. (1972). Reply to Hazelline. Nature, Lond., 240, 164. Schreiber, R. W. & Risf.brough, R. W. (1972). Studies of the brown pelican. 1. Status of brown pelican populations in the United States. Wilson Bull., 84, 119-35. Switzer, B. C., Wolfe, F. H, & Lewjn, V. (1972). Eggshell thinning and DDE. Nature. Lond., 240, 162-3. Whitehead, C. C., Downie, J. N. & Phillips, J. A. (1972a). BHC not found to reduce the shell quality of hens' eggs. Nature, Lond., 239, 411-2. Wiemeyer, S. N. & Porter, R. D. (1972). Nature, Lond., 240, 163. DSW 031359 STLCOPCB4015321