THE JOURNAL OF EXPERIMENTAL ZOOLOGY 276:157-163 (1996) Environment Modifies the Testosterone Levels of a Female Bird and Its Eggs HUBERT SCHWABL Rockefeller University, Field Research Center for Ecology and Ethology, Millbrook, New York 12545 ABSTRACT Past studies have shown that the yolk of the canary (Serinus canaria) egg contains maternal testosterone, that its concentrations increase in the subsequently formed eggs of a clutch, and that testosterone influences development. The present study investigated 1)if the testosterone levels vary in the female during yolk formation; 2) how such putative variations may be related to the concentrations of maternal testosterone in the yolk; and 3) if environmental factors, such as day length, can modify the testosterone levels in the mother and her eggs. Maternal testosterone levels, measured in the females' feces, increased during yolk formation and egg laying, and decreased during incubation. This pattern was modified by day length. In a photoperiod of 12 h of light and 12 h of darkness (12L 12D), female testosterone levels decreased after the last egg of the clutch was laid while in a photoperiod of 16L 8D, they decreased after the first egg was laid. These different patterns were reflected in the testosterone concentrations in the egg yolk. Further, the eggs of subsequent clutches that were produced under a naturally changing photoperiod differed significantly in their testosterone concentrations. Finally, the doses of testosterone in the yolk of individual eggs were positively correlated with the concentrations of testosterone in the female during the yolk phase of each egg. I conclude that here we have a mechanism which communicates environmental conditions from the mother to the offspring, and that this mechanism serves to optimize reproduction andor modify offspring traits. @ 1996 Wiley-Liss, Inc. The yolk of vertebrate eggs contains maternal hormones and growth factors which influence development (e.g., de Pablo et al., '82; Brown et al., '88; Scavo et al., '89; Dickhoff et al., '90; Schreck et al., '91; Sechman and Bobek, '88; Adkins-Regan et al., '95). The yolk of the canary contains the ovarian sex steroid hormone testosterone in increasing doses in each subsequently laid egg of a clutch (Schwabl, '93). These differential doses of maternal testosterone in the eggs of a clutch influence the growth of the hatchlings (Schwabl, '96). Moreover, the aggressiveness of the fully grown offspring is positively correlated with the doses of maternal testosterone in the eggs (Schwabl, '93; unpublished results). Two hypotheses emerged from these observations: 1) female birds can enhance their reproductive fitness by hormonal influences on sibling competition and growth (Winckler, '93; Schwabl, '96); and 2) the maternal hormones of birds can permanently alter the phenotype of the offspring (Schwabl, '93, '96) in a way that is analogous to how hormones transferred in utero between littermates in mammals affect development (Clark and Galef, '95). The functions for optimal reproduction and adaptation to a stochastic environment imply that the contents of maternal 0 1996 WILEY-LISS, INC. hormones in the eggs vary with maternal and/or environmental conditions. The present study probed for such environmental influences on the testosterone levels in the mother and the egg yolk. The major sources of sex steroid hormones, including testosterone, in female birds are the theca and granulosa cell layers that surround the ovarian follicles. The production of testosterone by these cells changes as follicles are recruited into the follicular hierarchy and sequester yolk (e.g., Bahr et al., '83; Etches and Duke, '84;Porter et al., '89) and this is reflected in the circulating levels of ovarian hormones in the blood (e.g., Shahabi et al., '75; Johnson and van Tienhoven, 'SO). While the secretion of ovarian steroid hormones is well studied with regard to the regulation of vitellogenesis, ovulation, egg laying, and behavior (e.g., Gilbert, '71; Schwabl, '921, it has not been studied with regard t o the possibility of transfer of these hormones to the egg. The different doses of testosterone in the yolk of the eggs may reflect either the androgenic activity of all Received March 4, 1996; accepted May 3, 1996. Address reprint requests to Hubert Schwabl, Department of Zoology, Washington State University, Pullman, WA 99164-4236. 158 H. SCHWABL ovarian follicles or the androgenic activity of each individual follicle. To partly differentiate between these possibilities and to test the prediction derived from the above-presented hypotheses the present study investig,ated 1)how the testosterone levels of female canaries change during the yolk phases of the subsequently formed eggs of a clutch; 2) if such changes are reflected in the concentrations of testosterone in the yolk; and 3) if the testosterone levels in the female and her eggs are influenced by environmental conditions, in this case photoperiod. Answering these questions required daily measurements of the hormone concentrations in the same female during egg formation and laying. Therefore, I monitored female testosterone levels non-invasively by the analysis of the concentrations in their feces. Previous studies with Japanese quail Coturnixjaponzca (Bishop and Hall, '91) and domestic fowl Gallus domesticus (Cockrem and Rounce, '94) have showin that the testosterone concentrations in the feces correlate with the concentrations in the blood. MATERIALS AND METHODS Birds and breeding conditions Subjects were Water xhlager canaries from the stock at the Rockefeller University Field Research Center. Breeding under constant photoperiodic conditions The breeding pairs used in this study had hatched in spring 1993. They had experienced the photoperiod of Millbrook, NY (42"N), until the beginning of the experiment. In October 1994 these birds were transferred t o a day length of 8 h of light and 16 h of darkness (8L 16D). Females were individually housed in breeding cages (46 x 26 x 23 cm), while males were kept in groups of 6 in flight cages (62 x 38 x 52 cm). After 6 weeks pairs were formed by introducing a male t o each female. Twelve pairs were then exposed to a photoperiod of 12L 12D and 12 pairs to one of 16L 8D. Birds were provided with fresh food and water daily as described previously (Schwabl, '93). Breeding under natural photoperiodic conditions Here birds experienced the naturally changing day length of Millbrook, NY,and were paired in March, Housing conditions were as described earlier (Schwabl, '96). Collection and preparation of fecal samples After exposure to long days droppings were collected daily from each female 1-2 h after the lights were turned on. For collection of fecal material the male and the female of a pair were separated by the insertion of a wire partitioning into the cage for 10-30 min. One to three droppings were then scraped from clean paper at the bottom of the cage that had been placed there after separation of the pair partners. Droppings were dried at room temperature for 24 h and weighed. The remaining procedure of measuring the concentrations of "free," unconjugated testosterone in the feces followed the protocol of Bishop and Hall ('91). Validation of testosterone measurements in feces The validation of this protocol for measurements of free testosterone in the feces of female canaries included the following steps. Specificity Three fecal samples (diluted 1:60 with phosphate-buffered saline, w:v) that were serially diluted with phosphate-buffered saline yielded displacement curves that were parallel t o the serial dilution of the testosterone standard (Fig. 1). The antiserum (Wien Laboratories, Inc., Succa-Sunna, New Jersey) that was used in the radioimmunoassays cross-reacts to 55% with 5adihydrotestosterone and may also cross-react with other steroid hormone metabolites in the feces. To test if the testosterone immunoreactivity that is measured in a buffer solution of feces without extraction and purification of steroids reflects testosterone, 12 fecal samples were treated as follows: 20 pl of fecal buffer solution was subjected to hydrolysis following the protocol of Payne and Talahay ('86). After hydrolysis with a mixture of P-glucuronidase and sulfatase from Helix pomatia (Boehringer-Mannheim Biochemicals, Indianapolis, IN) for 18 h steroids were extracted with 3 x 3 ml of diethylether-petroleum ether (1:l) using Extrelut minicolumns. Extracts were dried under a stream of nitrogen and steroids were separated and partially purified on diatomaceous earth chromatographic columns a s described earlier for blood samples (Schwabl, '92, '93). The fraction containing testosterone was then subjected t o a testosterone radioimmunoassay. The concentrations of immunoreactive testosterone that were measured in the unextracted fecal buffer solutions were 159 TESTOSTERONE IN MOTHER AND EGG Sample volume (PI) 6.25 100- 80 - B/Bo (“A) 12.5 I I I I 25 50 I 1 I I 60 - 40 - 200 ’, 1.95 I 3.9 7.8 15.6 31.2 62.4 I 125 I 250 1 500 Testosterone (pgltube) Fig. 1. Displacement curves of serially diluted fecal samples from three female canaries and testosterone standard in the testosterone radioimmunoassay. is present in the feces in “free” form 100 pl of fecal buffer solutions of the same samples was extracted with 3 x 3 ml of diethylethedpetroleum ether (1:l).The dried organic extracts were purified on diatomaceous earth chromatographic columns and subjected to testosterone assay. The remaining buffer solution from which free steroids had been extracted was then hy5000 r=0.72 drolyzed, extracted as described above, the expco.01 tract purified, and assayed for testosterone. The comparison of the concentrations measured before and after hydrolysis showed that 95% of the testosterone is present in the buffer solution in conjugated form. “Free” and conjugated testosterone concentrations were positively correlated (r = 0.70, n = 13, P < 0.01). The concentrations of testosterone in the buffer solution were positively correlated with the concentrations of conjugated (r = 0.79, n = 13, P < 0.01; data not shown) and free testosterone (r = 0.56, n = 13, P c 0.05; data not shown). Thus the concentrations 0 2500 5000 7500 of testosterone that are measured in unextracted Total testosterone (pglml) feces provide a good estimate of the total test(hydrolysis-extraction-chromatography) osterone levels in the female. strongly correlated with the concentrations of testosterone measured after hydrolysis, organic extraction, and chromatographic separation from 5a-dihydrotestosterone and other metabolites (Fig. 2). To investigate the fraction of testosterone that - . A E CI) P Y Q, c E c v) 0 c v) I-“ .-L . e/l n Fig. 2. Correlation of the immunoreactive (ir) testosterone concentrations measured in a phosphate buffer extract of the feces with the testosterone concentrations measured after hydrolysis, organic extraction, and separation of steroids by chromatography. Precision All samples were measured in a single assay and the intra-assay variation was 7%. H. SCHWAJ3L 160 '1 lkstosterone concentrations in the egg yolk Constant photoperiods Eggs were collected and replaced with dummy eggs at the day they were laid. The yolk was separated from the albumin, weighed, and frozen at 20°C. The testosterone concentrations in the yolk were assayed in 10-20 mg of yolk. All samples were analyzed in a single assay. A) 12L 12D (n=5) Incuballon E . A m C I Natural photoperiod Yolk samples were obtained by biopsy from freshly laid eggs of first clutches (laid in March 1994) and third or fourth clutches (laid in May) as described earlier (Schwabl, '93). The yolk samples were weighed, suspended in 0.5 ml distilled water, and frozen at -20°C. The extraction and measurement of testosterone followed the previously published protocol (Schwabl, '93). Statistics 0 ?! 7 al C Eal c v) 0 c -6 5 -4 -2 0 2 4 6 8 2 4 6 8 8 ) 16L 8D (n=5) v) a t -m E c s Data were analyzed by analysis of variance (ANOVA)for repeated rneasurements with the levels of significance for posthoc pair comparisons set a t P = 0.05. RESULTS Maternal testosterone levels during yolk formation In both photoperiodic regimens female testosterone levels varied significantly during vitellogenesis, egg laying, and inleubation [12L 12D: F(7,28) = 2.82, P = 0.024; 16Li 8D: F(7,26) = 13.32, P < 0.00011. In 12L 12D female testosterone levels increased significantly 1day before the first egg was laid. These levels were about twice as high as 4 days before the onset of laying. Levels remained high until the third of four eggs was laid and incubation started. During incubation testosterone levels declined to levels that were lower than those prior t o the laying (Fig. 3A). This pattern was different in females which laid in a photoperiod of 16L 8D. Here there waij a single peak of testosterone on the day when tbe first egg was laid. Testosterone levels dropped sharply with the second laid egg and continued t o decrease during incubation which began with the second egg. Levels during incubation were significantly lower than during the prelaying phase (Fig. 3B). -6 -4 -2 0 Days from first laid egg Fig. 3. Changes of immunoreactive (ir) testosterone concentrations (means with 1 S.E.M.) in female canaries during egg formation, laying (day 0 = first egg laid), and incubation. A: Females kept in a photoperiod of 12L 12D. B: Females kept in a photoperiod of 16L 8D. Four growing follicles (FlF4) are sketched into each graph t o indicate the phases of yolk deposition and ovulation. Asterisks indicate significant differences between two adjacent values (P< 0.05, two-tailed paired t-tests). quently laid eggs of a clutch contained increasing doses of maternal testosterone (Fig. 4A). The dosages of testosterone and the pattern of increase from egg to egg appeared to be influenced by photoperiod although the differences were not significantly different, probably due to the small sample sizes. First eggs of clutches that were produced in 12L 12D seemed t o contain more testosterone than first eggs that were produced in 16L 8D. Moreover, in 12L 12D the yolk testosterone contents tended to continue to increase up to the fourth egg, while in 16L 8D there appeared to be Yolk testosterone concentrations no further increase from the third to the fourth As reported earlier for eggs that were laid in a egg. These overall patterns of the testosterone conphotoperiod of 14L 101) (Schwabl, '93) the subse- tents of subsequently laid eggs reflect the differ- 161 TESTOSTERONE IN MOTHER AND EGG 120 100 80 1 60 40 20 n 1 2 3 4 T i Correlation of the yolk testosterone concentrations with the testosterone levels in the female during vitellogenesis The relationship between the testosterone levels in the mother and in the eggs was further explored by correlating the concentrations in individual yolks with the levels of the female during vitellogenesis of each egg. It was assumed 1) that the major portion of the yolk of each egg was sequestered during the last 72 h before the egg was ovulated (Romanoff and Romanoff, '49; Warren and Conrad, '39) and 2) that ovulation occurred 24 h before the egg was laid (Warren and Scott, '34). Pilot work (Schwabl, unpublished results) investigating "yolk rings," which can be used t o estimate the duration of yolk deposition in an egg (Grau, '76; Meathrel, '91), supports the former assumption. The results show that the testosterone concentrations in the yolk are positively correlated with the mean testosterone levels in the female during yolk deposition (Fig. 5). DISCUSSION 1 2 3 4 Laying order Fig. 4. Testosterone concentrations (means with 1S.E.M.) in the yolk of eggs that were produced (A) in constant photoperiods of 12L 12D or 16L 8D and (B) during different times of the year in a naturally changing photoperiod of 42"N. Asterisks indicate significant differences in the testosterone contents among the eggs of a clutch (P< 0.05, two-tailed paired t-test). This study reports four major findings: 1)The testosterone levels of female canaries increase during vitellogenesis and egg laying and decrease during incubation. 2) This pattern is influenced by photoperiod. 3) The Concentrations of testosterone in the eggs vary also with photoperiod and change with the progress of the breeding season. 4) The testosterone concentrations in the egg yolk are . 16L 8D(n=4) 0 140- 0 12L12D(n=3) l*O j rs=0.47 pco.01 0 .. . 0 100ent duration of elevated female testosterone levels in the two photoperiodic conditions. a 80 In naturally changing photoperiod conditions 0. the eggs of first clutches, laid in March when phoa 0 60 0 toperiod was approximately 12L 12D, had significantly higher testosterone contents than the eggs of third or fourth clutches, laid in May when photoperiod was approximately 15L 9D [F(1,21) = i 8.36, P = 0.0231. While there was variation among 0 . 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 the eggs of the March clutches [F(3,16) = 2.79, P = 0.0741, there was no indication for variation in Maternal irTestosterone 1 :60 (ng/ml) the May clutches [F(3,12) = 1.023, P = 0.4171. Third and fourth eggs of the March clutches had 5. Correlation of the testosterone concentrations in significantly higher testosterone levels than first theFig. yolk of individual eggs with the levels of testosterone in egg. Such an increase was absent in the May the female during the putative yolk phase of each egg. rs: clutches (Fig. 4B). Spearman rank correlation coefficient. 182 H. SCHWABL positively correlated with the testosterone concentrations in the female during vitellogenesis. These results clearly show that environmental factors influence the testosterone levels of the egg-laying female and her eggs. As four to five follicles are recruited into the follicular hierarchy and they grow by the incorporation of yolk the testlosterone production of the steroidogenic cells of tlhe follicular walls changes (e.g., Bahr et al., '83; Etches and Duke, '84). These changes in steroidogenic activity of the ovary which are reflected in the blood levels of the hormones have been studied extensively with regard to the control of the follicular hierarchy, vitellogenesis, ovulation, egg laying, and behavior (Gilbert, '71). The results presented here show that the production of testosterone is modified by photoperiod. Apparently vi tellogenesis, ovulation, and laying can successfully proceed despite such differences in the testostei~onelevels, suggesting that testosterone is only loosely associated with the control of these processes. The differences in female plasma levels of testosterone during egg formation among avian species (e.g., Hector et al., '86a,b; Donham, '79; Fowler et al., '94; Hannon and Wingfield, '90; Johnson and Van Tienhoven, '801, and more importantly, within species when they produce multiple clutches in a season (e.g., Hegner and Wingfield, '86; Windield, '85; Schwabl et al., '80; Licht et al., '79; McPheieson et al., ,821, suggest the same. One conclusion tlhat emerges from these observations is that, instead of being intimately involved in the control of reproductive functions, testosterone may be associated with behavioral responses to variable environmental conditions. This could be similar to the social control of the testosterone levels in males (Wingfield et al., '90). The results presented here, together with the previously published work (Schwabl, '93, '961, suggest, however, yet another, new possibility: testosterone and other hormones may convey to the offspring information that is perceived by the mother. The results of the present study are consistent with two predictions from this hypothesis: 1) the maternal testosterone levels vary with environmental conditions; and 2) this varia1;ion is reflected in the testosterone contents of the eggs. This new fundion of maternal steroid hormones as communicators between the mother and the egg is not at odds with their well-known roles in ovarian physiology. In fact, the dissociation of steroid production by the two steroidogenic cell layers which compose the follicular walls during follicular growth may reflect these different functions (e.g., Nitta et al., '93; Etches and Duke, '84). The outer theca cells, which are in intimate contact with the mother's bloodstream, might be the principal source of steroids that affect the mother, while the inner granulosa cells, which are not vascularized and are closer to the yolk, might be the source of steroids, which via the yolk affect the offspring. Further experiments are necessary to test this idea. The adaptive significance of environmentally modified transmission of maternal hormones to the offspring may come &om two mutually not exclusive effects. First, seasonal differences of the concentrations of maternal hormones in the eggs, caused by changing day length, may influence developmental rates in offspring born at various times during the breeding season. Moreover, they may cause individual differences in life history strategies (e.g., Adriaensen, '86; Smith and Nilsson, '87; Clark and Galef, '95; Gross, '96). Whether other environmental factors, such as population density, degree of competition for resources, and male courtship behavior can also influence the transmission of maternal hormones into the egg is currently under study. Second, the variation among clutches in the doses and pattern of testosterone may allow females to optimize reproduction by favoring individual offspring in a brood (Schwabl, '96; Schwabl et al., manuscript submitted). In summary, I conclude that testosterone is produced by females to modify, in an epigenetic way, offspring characteristics in response to environmental conditions. Further studies, ideally with animals in their natural environment, are necessary to probe the proposed adaptive functions and evolutionary significance of such a new role of hormones (Bern, '90; Schreck et al., '91), in this case ovarian sex steroids, as communicators between generations. ACKNOWLEDGMENTS This work was supported by NIMH grant MH 49877 and a grant from the Harry Frank Guggenheim Foundation. I thank Fernando Nottebohm for his support during this study. LITERATURE CITED Adkins-Regan, E., M.A. Ottinger, and J. Park (1995) Maternal transfer of estradiol to egg yolks alters sexual differentiation of avian offspring. J. Exp. Zool., 271:466-470. Adriaensen, F. (1986) Differences in migratory behavior between early and late brood nestlings of the robin Erithacus rubecula. Ibis, 129:263-267. Bahr, J.M., S.-C. Wang, M.Y. Huang, and F.O. Calvo (1983) Steroid concentrations in isolated theca and granulosa layers of preovulatory follicles during the omlatory cycle of the domestic hen. Biol. Reprod., 29:326-334. TESTOSTERONE IN MOTHER AND EGG Bern, H.A. (1990) The “new” endocrinology: Its scope and its impact. Am. Zool., 30:877-885. Bishop, C.M., and M.R. Hall (1991) Non-invasive monitoring of avian reproduction by simplified faecal steroid analysis. J. Zool. Lond., 224:649-668. Brown, C.L., S.I. Doroshov, J.M. Nunez, C. Hadley, J. Vaneenennaam, R.S., Nishioka, and H.A. Bern (1988) Maternal triiodothyronine injections cause increases in swimbladder inflation and survival rates in larval striped bass, Morone saxatilis. J. Exp. Zool., 248:168-176. Clark, M.M., and B.G. Galef (1995) Prenatal influences on reproductive life history strategies. TREE, 10~151-153. Cockrem, J.F., and J.R. Rounce (1994) Faecal measurements of oestradiol and testosterone allow the non-invasive estimation o f plasma steroid concentrations in the domestic fowl. Br. Poult. Sci., 35:433-443. De Pablo, F., J . Roth, E. Hernandez, and R.M. Pruss (1982) Insulin is present in chicken eggs and early chick embryos. Endocrinology, 111:1109-1916. Dickhoff, W.W., C.L. Brown, C.V. Sullivan, and H.A. Bern (1990) Fish and amphibian models for developmental endocrinology. J. Exp. Zool. Suppl., 4:90-97. Donham, R.S. (1979) Annual cycle of plasma luteinizing hormone and sex hormones in male and female mallards (Anas platyrhynchos). Biol. Reprod., 2213273-1285. Etches, R.J., and C.E. Duke (1984) Progesterone, androstenedione and estradiol content o f theca and granulosa tissues of the four largest ovarian follicles during the ovulatory cycle of the hen (Gallus domesticus). J. Endocrinol., 103~71-76. Fowler, G.S., J.C. Wingfield, P.D. Boersma, and R.A. Sosa (1994) Reproductive endocrinology and weight change in relation to reproductive success in the Magellanic penguin (Speniscus magellanicus). Gen. Comp. Endocrinol., 94:305-315. Gilbert, A.B. (1971) The endocrine ovary in reproduction. In: Physiology and Biochemistry of the Domestic Fowl, Vol. 3. D.J. Bell and B.M. Freeman, eds. Academic Press, New York, pp. 1449-1468. Grau, C.R. (1976) Ring structure of avian egg yolk. Poult. Sci., 55:1418-1422. Gross, M.R. (1996) Alternative reproductive strategies and tactics: Diversity within sexes. TREE, 11:92-98. Hannon, S.J., and J.C. Wingfield (1990) Endocrine correlates of territoriality, breeding stage, and body molt in free-living Willow ptarmigan of both sexes. Can. J. Zool., 68:2130-2134. Hector, J.A.L., J.P. Croxall, and B.K. Follett (1986a) Reproductive endocrinology of the wandering albatross, Diomedea exulans, in relation to biennial breeding and deferred sexual maturity. Ibis, 128:9-22. Hector, J.A.L., B.K. Follett, and P.A. Prince (198613) Reproductive endocrinology of t h e black-browed albatross Diomedea melanophrys and the grey-headed albatross D. chrysostoma. J. Zool. Lond., 208:237-253. Hegner, R.E., and J.C. Wingfield (1986) Behavioral and endocrine correlates of multiple brooding in the semicolonial house sparrow Passer domesticus. 11. Females. Horm. Behav., 20:313-326. Johnson, A.L., and A. Van Tienhoven (1980) Plasma concentrations of six steroids and LH during the ovulatory cycle of the hen, Gallus domesticus. Biol. Reprod., 23r386-393. Licht, P., J. Wood, D.W. Owens, and F. Wood (1979) Serum gonadotropins and steroids associated with breeding activities in the green sea turtle Chelonia mydas. I. Captive animals. Gen. Comp. Endocrinol., 39:274-289. McPherson, R.J., L.R. Boots, R. MacGregor, 111, and K.R. 163 Marion (1982) Plasma steroid associated with seasonal reproductive changes i n a multiclutched freshwater turtle, Sternotherus odoratus. Gen. Comp. Endocrinol 48:440-451. Meathrel, C.E. (1991) Variation in eggs and the period of rapid yolk deposition of the silver gull Larus nouaeholZandiae during a protracted laying season. J. Zool. Lond., 223:501-508. Nitta, H.. J.I. Masin, and J.M. Bahr (1993) Localization of 3p-hydroxysteroid dehydrogenase in the chicken ovarian follicle shifts from the theca layer to granulosa layer with follicular maturation. Biol. Reprod., 48:llO-116. Payne, D.W., and P. Talalay (1986) A one-step enzymatic assay for the measurement of 17p-hydroxy- and 17-0x0-steroid profiles in biological samples. J. Steroid Biochem., 25:403%410. Porter, T.E., B.M. Hargis, J.L. Silsby, and M.E. El Halawani (1989) Differential steroid production between theca interna and theca externa cells: A three-cell model for follicular steroidogenesis in avian species. Endocrinology, 125:109-116. Romanoff,A.L., and A.J. Romanoff ( 1949) The Avian Egg. John Wiley & Sons, Inc., New York. Scavo, L., J. Alemany, J. Roth, and F. De Pablo (1989) Insulinlike growth factor I activity is stored in the yolk of the avian egg. Biochem. Biophys. Res. Commun., 162:1167-1173. Schreck, C.B., M.S. Fitzpatrick, G.W. Geist, and C.-G. Yeoh (1991) Steroids: Developmental continuum between mother and offspring. In: Proceedings of the Fourth International Symposium on the Reproductive Physiology of Fish. A.P. Scott, J.P. Sumpter, D.E. Kime, and M.S. Rolfe, eds. Fish Symp. 91. Sheffield, pp. 256-258. Schwabl, H. (1992) Winter and breeding territorial behavior and levels of reproductive hormones o f migratory European robins. Ornis Scand., 23:271-276. Schwabl, H. (1993) Yolk is a source of testosterone for developing birds. Proc. Natl. Acad. Sci. U.S.A., 90:11446-11450. Schwabl, H. (1996) Maternal testosterone in the avian egg enhances post-natal growth. Comp. Physiol. Biochem. 114 (in press). Schwabl, H., J.C. Wingfield, and D.S. Farner (1980) Seasonal variation in plasma levels of luteinizing hormone and steroid hormones in the European blackbird n r d u s merula. Vogelwarte, 30:283-294. Sechman, A., and S. Bobek (1988) Presence of iodothyronines in the yolk of the hen’s egg. Gen. Comp. Endocrinol., 69:99-105. Shahabi, N., J.M. Bahr, and A.V. Nalbandov (1975) Steroid levels in follicles and the plasma of hens during the ovulatory cycle. Endocrinology, 96:962-968. Smith, G.S., and J.A. Nilsson (1987) Intraspecific variation in migratory pattern of a partial migrant, the blue tit (Parus caeruleus): An evaluation of different hypotheses. Auk, 104:109-115. Warren, D.C., and R.M. Conrad (1939) Growth of the hen’s ovum. J. Agric. Res., 589375. Warren, D.C., and H.M. Scott (1934) Ovulation in the domestic hen. Science, 80:461. Winckler, D. (1993) Testosterone in egg yolks: An ornithologist’s perspective. Proc. Natl. Acad. Sci. U.S.A., 90:11439-11441. Wingfield, J.C. (1985) Influences of weather on reproductive function in female song sparrows, Melospiza melodia. J. Zool. Lond., 205.545-558. Wingfield, J.C., R.E. Hegner, A.M. Dufty, Jr., and G.F. Ball (1990) The “challenge hypothesis”: Theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. Am. Nat., 136:829-846.