- Email: [email protected]
Maternal testosterone in the avian egg enhances postnatal growth
Camp. Biochem. Physiol. Vol. 114A, No. 3, pp. 271-276, Copyright 0 1996 Elsevier Science Inc.
ISSN 0300-9629/96/$15.00 PII SO300-9629(96)00009-6
1996
ELSEVIER
Maternal Testosterone in the Avian Egg Enhances Postnatal Growth Hubert Schwabl THE ROCKEFELLERUNIVERSITY FIELD RESEARCH CENTER FOR ECOLOGY AND ETHOLOGY, TYRREL ROAD, MILLBROOK, NY 12545, USA
ABSTRACT. reported
The eggs of the canary
experiments
differences
investigated
(Serinw
whether
canaria)
testosterone
contain influences
variable nestling
of the growth of nest mates that are caused by asynchronous
into the yolk of unincubated simultaneously
from control
eggs enhanced
the growth after hatching
eggs. These differences
were established
The with
hatching.
compared
Injections
to nestlings
of testosterone that had hatched
Exogenous
chicks
increases in each subsequently nestlings
to chicks
that hatched
that hatched
laid egg in a clutch.
Consistent
with the results obtained
from eggs with higher concentrations
synchronously
testos-
in the growth of control birds. Testoster-
also begged more often for food. Previous studies have shown that the content
one-treated testosterone compared
testosterone.
growth and how this interacts
within 22 hr of hatching.
terone promoted growth in both sexes and there was no sexual difference
one injections
doses of maternal
of maternal
from eggs with lower testosterone
testosterone
concentrations.
of maternal by testostergrew faster However,
due to asynchronous incubation of clutches. This direct effect of maternal testosterone on growth in combination with a flexible onset of incubation allows to selectively enhance the growth and fitness of individual offspring of a brood. COMP more testosterone
did not compensate
for reduced growth that was caused by later hatching
BIOCHEM PHYSIOL 114A;3:271-276, 1996.
KEY WORDS. Begging, canary, development,
optimal
reproduction,
nestling,
Serinus canuria, sibling compe-
tition
INTRODUCTION The postnatal growth of altricial birds depends on the food they obtain from the parents. The nestlings in a brood compete with each other for parental food which in some large predatory species results in siblicide (8,23,24).
Small passer-
ine birds compete for food by begging (11,12,20,30,33) and differences in the ability to obtain food influence nestling growth ( 16) can lead to death by starvation (28) and affect juvenile survival (19). In both cases hatching asynchrony, which is caused by the onset of incubation
of the clutch
before the last egg is laid, influences the abilities of nestlings to compete with each other for food indirectly by establishing an age/size hierarchy among the nestlings (15,4,18). Recently it was proposed that maternal hormones that are transferred into the egg affect post-natal growth or the competitive abilities of the nestlings directly (31,40). The yolks of the eggs of a canary clutch contain variable doses of maternal testosterone. These doses are low in the first laid eggs and increase in each subsequently laid egg in a clutch (31, H. Schwab1 1993, unpublished results). The Address re@ints reqursts to currentaddress: H. Schwabl, Department of Zoology, Washington State University, Pullman, WA 99164. Recewed 20 July 1995; revised 11 December 1995; accepted 18 December 1995.
here reported experiments tested the hypothesis that maternal testosterone enhances nestling growth. Experiment 1 investigated whether nestlings that hatch from eggs with experimentally increased contents of testosterone grow faster compared to nestlings that hatch from control eggs. Experiment
2 investigated when and how the growth differ-
ences are established.
Experiment
3 evaluated whether the
higher doses of maternal testosterone in the later laid eggs in a clutch can compensate for the adverse effects of hatching asynchrony
on the growth of younger nestlings.
To investigate the effects of high or low doses of testosterone in the egg on nestling growth I injected first and second eggs of the clutch sequence with a dose of testosterone
(100
ng) which is equivalent to the high doses measured in third, fourth or fifth eggs (31). The growth of the nestlings that hatched from these eggs was compared with the growth of nestmates hatching from control-injected first and second eggs that contained only the low maternal doses of testosterone (3 1, H. Schwabl, unpublished results). Confounding effects of egg mass (39) or asynchronous hatching (18) on growth were controlled for by matching the eggs for their mass and by experimentally inducing synchronous hatching. To estimate the relative influences of differential contents of maternal testosterone in the eggs of a clutch and hatching asynchrony on growth rates, I measured the con-
H. Schwabl
272
centrations of maternal testosterone in the eggs and allowed asynchronous incubation and hatching to occur naturally.
MATERIALS
AND METHODS
Subjects were Waterslager canaries bred at the Rockefeller University Field Research Center in Millbrook, New York. Testosterone EXPERIMENT
Injections were stimulated to to long days (14L:lOD) in November
1993 after 4 weeks of exposure to short days (8L:16D).
Eggs
were collected daily within 1 hr after the lights were turned on and replaced with dummy eggs. Eggs were weighed and the first and second eggs in the clutch sequence were selected for the experiment. They contained low doses of maternal testosterone
Data on begging performance
(3 1, H. Schwabl,
unpublished
results).
are compiled
from experi-
ments 1 and 2. To quantify begging the nest was removed from the cage and the number and duration of begging bouts of a newly hatched chick (within an hr of hatching) was recorded. To elicit begging the nest was tapped 5 times over the course of 1 min. A begging bout was defined as the response of the hatchling by raising its head and gaping. When
1. Ten pairs of canaries
breed by exposure
Begging Behavior
a hatchling
begged already when the nest was re-
moved this response was included.
Maternal
Testosterone
EXPERIMENT
Versus Asynchronous
Hatching
3. During the spring breeding
the concentrations
of maternal testosterone
season 1993 in each egg of
9 first clutches were measured. Eggs were taken from the nest in the morning they were laid. They were weighed and
Experimental eggs were injected into the yolk with 100 ng of testosterone in 5 ,~l of sterile sesame oil. This dose corre-
a small yolk sample was removed from each egg for measurement of the testosterone concentration by radioimmunoas-
sponds to the high maternal doses that have been measured in the later laid eggs in a clutch. Control eggs were injected with 5 ~1 of sesame oil. Injections were made through the
say following the methods of Schwab1 (31). Each egg was returned within 1 hr to the nest. At hatching the identity of each hatchling was recorded and each hatchling was indi-
small pole of the egg with a sterile needle (26 g) using illu-
vidually marked. Broods of various sizes (2-4)
mination
ious degrees of asynchronous
from underneath
to ascertain
that the tip of the
and with var-
and synchronous
hatching
needle had penetrated the yolk membranes. The hole in the shell was then patched with flexible wound dressing.
were obtained. Each nestling was weighed when it was 10 days old. During the nestling phase food was available ad
Each egg was individually marked and clutches of 4 eggs were formed consisting of 1 control and 1 experimental egg
libitum and replaced 3 times daily.
of each laying position (1 and 2). Growth differences due to variation in egg size (32,39) were controlled by matching
SEX. The sex of each subject was identified by visual in-
spection of the gonads at an age of 3 to 4 months.
the eggs of each clutch as closely as possible for their mass. Growth differences due to asynchronous hatching (4,18) were eliminated by returning the eggs of a clutch simultaneously into the nest of a female that had stopped laying and was firmly incubating dummy eggs. At the time of expected hatching the nests were inspected
at 2 hr intervals
and the egg from which
each
hatchling hatched identified. Hatchlings were weighed and the lengths of their tarsi were measured. After clipping claws for subsequent identification, hatchlings were returned to their nests. From then on measurements were taken twice daily once in the morning within 1 hr after the lights were switched on and once 4-6 hr before the end of the light phase. Growth curves were plotted for each indi-
STATISTICAL
ANALYSIS.
Data were analyzed by ANOVA
for repeated measures, followed by Student t-tests. Data on begging were analyzed by Mann-Whitney U-tests.
RESULTS Effects of Testosterone-injections on Nestling @owth (Experiment
1)
Treatment of the egg did not influence (testosterone: 62.1%; controls: 60.9%). terone
injections
accelerated
the hatching rate However, testos-
the post-natal
growth
and
EXPERIMENT z. The protocol for this experiment was similar to that of Experiment 1, except that it was performed with 20 breeding pairs during the breeding season in spring 1994 under the changing natural photoperiod of Millbrook,
development compared to synchronously hatching controls (Fig. 1). Although the slopes of fitted logistic growth curves (29) were not significantly different between treatment groups the hatchlings from testosterone-injected eggs reached 50% of their final tarsus length more quickly (3.4 days after hatching compared to 4.3 days in controls: t = 2.44, df = 14, p = 0.029). They also had developed detectable eye slits earlier than hatchlings from control eggs (t = 2.93, df = 11, p = 0.014). This lead in development of approximately 1 day persisted until 80% of the final tarsus length was reached (t = 2.16, df = 14, p = 0.049). Mass and size at the end of the growth phase were not different
NY.
between trearment
vidual (29) and mass and tarsus length interpolated for each 24 hr of age. A constant brood size had been intended, but due to infertility and failure to hatch brood sizes actually ranged from 2 to 4. Fresh food was provided ad libitum 3 times daily throughout the nestling phase.
groups.
273
Maternal Testosterone and Offspring Growth
E
test than control-treated
Although
the duration
2
22
of a single begging bout was not significantly
different be-
y
20
2 s ‘ij
18
tween treatment groups, this resulted in significantly more overall begging of testosterone-treated hatchlings (Fig. 3). Because the testosterone-treated hatchlings also gained
16 14
2 z
12
f
10
=.
8
s z!
6 4
ii
;
6
more weight and grew faster, their begging must have resulted in greater food acquisition (26,35), though this was not directly measured. 0
2
4
ii
6
8
testosterone (8) 10 12 14 16 18
Effects of Testosterone-injections Immediately
This experiment differences
on Embryonic Qrowth (Experiment 2)
after Hatching
addressed the question when and how the
in nestling growth are established.
Maternal Testosterone Contents Versus Hatching Asynchrony (Experiment To evaluate the relative contributions
Age WW
FIG. 1. Growth (means with standard errors) of nestlings from testosterone-injected or control eggs. Arrows on the curve for tarsus length indicate when the testosteronetreated and control nestlings had developed eye slits.
and @owth
hatchlings.
Testosterone
= 2, p < 0.01) and treatment (F = 3.33, df = 2, p < 0.05 and F = 3.67, df = 2, p < 0.05). There were no significant differences in mass or size at hatching. However, at an age of 22 hr the nestlings from the testosterone-injected eggs
of variable maternal
testosterone contents in the eggs of a clutch and asynchronous hatching to differences in nestling growth I measured the concentrations of maternal testosterone in each egg of 9 clutches and compared the mass of pairs of nestlings that had hatched in various intervals from each other. Nine pairs of siblings hatched within 8 hr from each other on the same day. At 10 days of age, the chick from the egg with the higher testosterone content (that is the later-laid egg) weighed more than the chick from the egg with the lower content in 6 of these 9 dyads. This mass difference decreased
did not enhance embryonic development, since testosterone-treated and control eggs hatched after the same duration of incubation (320 ? 3 hr for testosterone-injected eggs and 316 5 4 hr for control eggs, t = 0.88, p = 0.39). Post-hatching mass and tarsus length varied significantly with age (F = 47.94, df = 2 p < 0.01 and F = 49.59, df
3)
2.4
-
IA
Controls (ll=13)
2.2 1
8 -
2.0-
t :
1.6-
6 m”
1.6-
weighed more and had longer tarsi than those from the control eggs (Fig. 2A, B). There was no sex difference in the
lemales @A) testosterone:
A 0
early growth of controls and testosterone accelerated the growth of both sexes (Fig. 2A, B). At the age of 22 hr fe-
-3
males (rod) females (n=6)
D
3
6
9
4 12 15 10 21 24 27
0
3
6
9
12 15 18 21 24 27
males from testosterone-treated eggs were heavier than females from control eggs (t = 2.44, p < 0.02, l-tailed) and males from testosterone-treated eggs had longer tarsi than males from control eggs (t = 1.97, p < 0.05, l-tailed). Thus testosterone had identical effects on the sexes, which resulted in faster growth within the first day after hatching possibly through better food acquisition. To evaluate this possibility I investigated hatched chicks.
Effects of Exogenous 011 Begging Behavior
the begging behavior of the newly
Testosterone at Hatching
Testosterone-treated hatchlings were more likely to beg immediately after hatching than control-treated hatchlings. All 8 testosterone-treated hatchlings begged, while only 3 out of 6 control-treated hatchlings begged. Moreover, testosterone-treated hatchlings tended to beg more often per
J t?
6.8 -
d
6.6 -
6
6.4 -
z
P
3
6.2 6.0 -3
Age (hours) FIG. 2. Mass (A) and tarsus length (B) of hatchlings from testosterone- or control-injected eggs during the first 24 hr after hatching. Means and standard errors for measurements and age at measurement are depicted. Symbols plotted at 24 hr show the mass and tarsus length at the age of 22 hr for each sex and treatment.
H. Schwab1
274
. -6 -0 Testosterone
(n=B)
FIG. 3. Begging behaviour of testosterone-treated and cone trol chicks within 1 hr after hatching. Means and standard errors of the means are shown for the duration of beginning bouts and the overall duratioqof begging. The number of begging bouts per test are depicted as means with ranges. Significance levels for differences between treatment groups are indicated above bars for controls.
with seniority
of the chick
that
hatched from the egg with less testosterone (Fig. 4). In contrast, in only 1 out of 8 cases in which hatching occurred 20 or more hr apart the chick testosterone content weighed tosterone in the egg resulted pared to a chick that hatched ing 6-33
hr earlier
from the egg with the high more. On average, more tesin a 22% greater mass comwithin 4 hours, while hatch-
from an egg with less testosterone
resulted in 15% greater mass (Fig. 4, insert).
DISCUSSION
The results of the 3 experiments support the hypothesis that maternal testosterone in the eggs enhances post-natal growth and development. This adds differential doses of maternal testosterone in the eggs to asynchronous incubation (18) and differences in egg size (32,39) as maternal factor which can cause differential
growth among the nestlings of
a brood. I will first discuss the mechanisms by which testosterone may act and then the function of this hormonal influence on nestling growth.
4
8
12
16
difference
20
24
28
32
36
(hours)
FIG. 4. Mass difference between pairs of chicks at 10 days of age that had hatched in various intervals from each other from eggs with a high or low concentration of maternal testosterone. Abscissa: Difference in age between the chick that hatched from the egg with the low concentration of maternal testosterone and the chick that hatched from the egg with the high testosterone (positive values: the chick from the egg with the high testosterone concentration is younger). Ordinate: Mass of the chick from the egg with more maternal testosterone minus mass of the chick from the egg with less testosterone (positive values: the chick from the egg with more testoserone weighs more; negative values: it weighs less). The solid line is the linear regression of this mass difference versus the age difference for pairs of chicks that hatched on the same day within 8 hr (filled circles). Open circles represent data for pairs of chicks which hatched on different days. Insert: Left bar: Relative mass difference in percent (mean with standard error) that resulted from a higher testosterone content in the egg compared to a synchronously hatching chick (within 4 hr). Right bar: relative mass difference that resulted from hatching earlier (6-33 hr) compared to a chick from an egg with more maternal testosterone.
begging because yolk reserves are depleted at hatching. Support for this possibility comes from two lines of evidence: (a) Testosterone enhances the metabolism in adult birds and (b) metabolic deficits (hunger) intensify begging, at least in older nestlings (33,26). The perception of cues that elicit begging: Before we can answer the question if testosterone enhances the development of perceptual processes, we need more information about the nature of the cues that elicit begging and interactions of chicks and parents at hatching. It is possible that testosterone enhances begging by an influence
Mechanisms The mechanism
,‘f’~‘~‘~‘,‘~~~
0
Age
Egg treatment
rapidly and significantly
.I
-4
Control(n-6)
by which maternal testosterone
influences
nesting growth differs from those of asynchronous hatching or egg size variation which both act indirectly by causing differences in the age or size of nest mates (18,39, H. Schwabl, unpublished results). Testosterone has a different influence on the hatchling which results in its faster growth, probably by greater food acquisition through intensified food begging behavior. Begging involves several processes that testosterone might modify. 1. Metabolic needs: It is possible that testosterone increases the embryo’s metabolism, causing earlier and intensified
on attention or persistency, which have been shown to be improved in the newly hatched chick (G&s domestictls) by exposure to testosterone (1,2,5). The actual motor pattern of raising the head, extending the neck and gaping may be enhanced by effects on the development of muscles or the differentiation of motor control systems. The observations that exogenous androgens are accumulated in the avian embryonic spinal cord and spinal ganglia suggesting the presence of receptors (27) and that testosterone enhances the proliferation of motor neurons (14) lend some support to such an expla-
275
Maternal Testosterone and Offspring Growth
nation.
Presently,
it is not possible to distinguish
tween these possibilities.
However,
be-
it can be excluded
which the growth and fitness of the nestlings from the first laid eggs can be enhanced.
But the alternative
of synchro-
that the faster growth of testosterone-treated hatchlings resulted simply from the elimination of a sex difference (37) because (a) there was no sex difference in the
nous incubation cannot selectively enhance the fitness of the nestlings from later eggs. I suggest that the increasing
growth of controls,
nation with synchronous incubation provide a mechanism to optimize reproduction by selectively enhancing the
hanced
and (b) testosterone
injections
en-
the growth of both sexes.
in the subsequently laid eggs in combi-
growth and fitness of the offspring from the later laid eggs in a clutch.
Function The results demonstrate
that the differential doses of mater-
nal testosterone in the eggs of a clutch result in nestling growth differences, at least if hatching is nearly synchronous. They do not suggest, however, that the higher doses of maternal testosterone in the later laid eggs compensate for growth differences that result from asynchronous hatching, although it cannot be ruled out that testosterone somewhat mitigates the disadvantages of hatching later. At present it is parsimonious to assume (a) that the variable doses of maternal testosterone
in the egg cause growth differences
among synchronously hatching nestlings that can be as great as those resulting from asynchronous hatching; and (b) that asynchronous hatching overrides this hormonal influence. Since asynchronous hatching depends to a large extent on the onset of incubation of a clutch (4,18) the incubation strategy will decide whether hatching asynchrony or differences of the doses of maternal testosterone prevail to cause differential
doses of testosterone
growth among the nestlings of a brood. In the
canaries of this study the onset of incubation changes in subsequent clutches (H. Schwabl, unpublished results) and such changes with the progression of the breeding season reported for several have been other species (3,10,13,21,22,38). In the last part I discuss why it may be adaptive to influence nestling growth by two different mechanisms: on the one hand, by asynchronous incubation and, on the other hand, by higher contents of maternal testosterone in each subsequently laid egg in a clutch and synchronous incubation. Hypotheses of optimal clutch and brood size emphasize that selection
maximizes the number and size of offspring
that are produced (17,18,24,36). This implies that all offspring of a brood are of equal value to the parent, regardless from which egg they hatch which might, however, not be the case (25). For example, in European kestrels (F&o tinninculus) the value of an offspring that is produced at a given time of year depends on its sex (6). At the same time the primary sex ratio varies systematically with the laying order of the eggs in this species (7) as well as in other species including the canary (H. Schwabl, unpublished results). If the value of an offspring depends on its sex and the sex ratio varies predictably with egg sequence, it will be adaptive to evolve mechanisms by which the fitness of the offspring of specific eggs in the clutch can be selectively influenced. Asynchronous incubation is a well known mechanism by
I thankDam Jackson, Sharon Sepe, and Helen Ecklund for their help with breeding and Sibylle Kbnig, Douglas Mock, and Fernando Notrebohm for critical comments on earlier versions of this manuscript and 3 anonymous reviewers for valuable comments. The work was supported by the Mary Flagler Cary Charitable Trust, Public Health Service Grant MH 49877, and a g-rantfrom the Harry Frank Guggenheim Foundation.
References 1. Andrew, R.J. Effects of testosterone on the behaviour of the domestic chick: I. Effects present in males but not in females. Anim. Behav. 23:139-155;1975. 2. Andrew, R.J. Effects of testosterone on the behaviour of the domestic chick. II. Effects present in both sexes. Anim. Behav. 23:156-168;1975. 3. Beukeboom, L.; Dijkstra, C.; Daan, S.; Meijer, T. Seasonality of clutch size determination in the kestrel F&o tinnunculus. Om. Stand. 19:41-48;1988. 4. Clark, A.B.; Wilson, D.S. Avian breeding adaptations: hatching asynchrony, brood reduction and nest failure. Quart. Rev. Biol. 56:257-277;1981. 5. Clifton, P.G.; Andrew, R.J.; Brighton, L. Gonadal steroids and attentional mechanism in young domestic chicks. Physiol. Behav. 43:441-446;1988. 6. Daan, S.; Dijkstra, C.; Drent, R.H.; Meijer, T. Food supply and the annual timing of avian reproduction. In: Qiiellet, H. (ed). Acta XIX Congressus Intemationalis Ornithologici. Ottawa: University of Ottawa Press; 1989:392-407. 7. Dijkstra, C.; Daan, S.; Buker, J.B. Adaptive seasonal variation in the sex ratio of kestrel broods. Funct. Ecol. 4:143-147; 1990. 8. Drummond, H. Parent-offspring conflict and brood reduction in the Pelicaniformes. Colonial Waterbirds lO:l-15;1987. 9. Feuerbacher, 1.; Prinzinger, R. The effect of the male sex-hormone testosterone on body temperature and energy metabolism in male Japanese Quail (Coturnix co&x japonica). Comp. Biochem. Physiol. 70A:247-250;1981. 10. Gibb, J. The breeding biology of the great and blue titmice. Ibis 92:507-539;1950. 11. Gottlander, K. Parental feeding behaviour and sibling competition in the pied flycatcher Ficedula hypoleuca. Orn. Stand. 18:269-276;1987. 12. Greig-Smith, P.J. Weight differences, brood reduction, and sibling competition among nestling stonechats. Saxicola torquuta. J. Zool. (London) 205:453-465;1985. 13. Hafthom, S. Incubation during the egg-laying period in relation to clutch size and other aspects of reproduction tn the great tit Parus major. Om. Stand. 12:169-185;1981. CD. Androgen increases the 14. Hauser, K.F.; Torand-Allerand, number of cells in fetal mouse spinal cord cultures: implications for motomeuron survival. Brain Res. 485:157-164;1989. 15. Lack, D. The natural regulation of animal numbers. Oxford: Clarendon Press; 1964.
276
16. Liihrl, H. Das Nesthakchen als biologisches Problem. J. Omithol. 109:383-395;1968. 17. Magrath, R.D. Hatching asynchrony and reproductive success in the blackbird. Nature 339:536-538;1989. 18. Magrath, R.D. Hatching asynchrony in altricial birds. Biol. Rev. 65:587-622;1990. 19. Magrath, R.D. Nestling weight and juvenile survival in the blackbird, Turdus merula. J. Anim. Ecol. 60:335-351;1991. 20. McRae, S.B.; Weatherhead, P.J.; Montgomerie, R. American robin nestlings compete by jockeying for position. Behav. Ecol. Sociobiol. 33:101-106;1993. 21. Mead, P.S.; Morton, M.L. Hatching asynchrony in the whitecrowned sparrow (Zononichia leucophrys oriunthu): a selected or inherited trait? Auk 102:781-792;1985. 22. Meijer, T. Incubation development and clutch size in the starling. Orn. Stand. 21:163-168;1990. 23. Mock, D.W. Infanticide, siblicide, and avian nestling mortality. In: Hausfater, G.; Hrdy, S.B., (eds). Infanticide: comparative and evolutionary perspectives. New York: Aldine; 1984: 3-30. 24. Mock, D.W. Siblicide, parent-offspring conflict, and unequal parental investment by egrets and herons. Behav. Ecol. Sociobiol. 20:247-256~1987. 25. Mock, D.W.; Forbes, L.S. Parent-offspring conflict: a case of arrested development. TREE 7:409-413;1992. 26. Price, K.; Ydenberg, R. Begging and provisioning in broods of asynchronously-hatched yellow-headed blackbird nestlings. Behav. Ecol. Sociobiol. 37:201-208;1995. 27. Reid, F.A.; Gasc, J.-M; Stumpf, W.E.; Sar, M. Androgen target cells in spinal cord, spinal ganglia, and glycogen body of chick embryos. Exp. Brain Res. 44:243-248;1981. 28. Richter, W. Hatching asynchrony: the nest failure hypothesis and brood reduction. Am. Nat. 120:828-832;1982.
H. Schwab1
29. Ricklefs, R.E. A graphical method of fitting equations to growth curves. Ecology 48:978-983;1967. 30. Ryden, 0.; Bengtson, H. Differential begging and locomotory behavior by early and late hatched nestlings affecting distribution of food in asynchronously hatched broods of altricial birds. Z. Tierpsychol. 53:209-224;1980. 3 1. Schwabl, H. Yolk is a source of maternal testosterone for developing birds. Proc. Natl. Acad. Sci. USA 90:11449-l 1450; 1993. 32. Slagsvold, T.; Sandvik, T.; Rofstad, G., Lorentsen, 0.; Husby, M. On the adaptive value of intraclutch egg-size variation in birds. Auk 101:685-697;1984. 33. Smith, H.G.; Montgomerie, R. Nestling American robins compete with siblings by begging. Behav. Ecol. Sociobiol. 29: 307-312;1991. 34. Smith, CC.; Fretwell, S.D. The optimal balance between size and number of offspring. Am. Nat. 108:499-506;1974. 35. Stamps, J.; Clark, A.; Arrowood P.; Kus B. Begging behaviour in budgerigars. Ethology 81:177-192;1988. 36. Stearns, SC. The evolution of life histories. Oxford: Oxford Universiry Press, 1992. 37. Teather, K.L. An experimental study of competition for food between male and female nestlings of the redwinged blackbird. Behav. Ecol. Sociobiol. 31:81-87;1992. 38. van Balen, J.H. A comparative study of the breeding ecology of the Great tit Parus major in different habitats. Ardea 61: l-93;1973. 39. Williams, T.D. Intra-specihc variation in egg size and egg composition in birds: effects on offspring fitness. Biol. Rev. 68:35-59;1994. 40. Winkler, D.W. Testosterone in egg yolks: an ornithologist’s perspective. Proc. Natl. Acad. Sci. USA 90:11439-11441; 1993.
ISSN 0300-9629/96/$15.00 PII SO300-9629(96)00009-6
1996
ELSEVIER
Maternal Testosterone in the Avian Egg Enhances Postnatal Growth Hubert Schwabl THE ROCKEFELLERUNIVERSITY FIELD RESEARCH CENTER FOR ECOLOGY AND ETHOLOGY, TYRREL ROAD, MILLBROOK, NY 12545, USA
ABSTRACT. reported
The eggs of the canary
experiments
differences
investigated
(Serinw
whether
canaria)
testosterone
contain influences
variable nestling
of the growth of nest mates that are caused by asynchronous
into the yolk of unincubated simultaneously
from control
eggs enhanced
the growth after hatching
eggs. These differences
were established
The with
hatching.
compared
Injections
to nestlings
of testosterone that had hatched
Exogenous
chicks
increases in each subsequently nestlings
to chicks
that hatched
that hatched
laid egg in a clutch.
Consistent
with the results obtained
from eggs with higher concentrations
synchronously
testos-
in the growth of control birds. Testoster-
also begged more often for food. Previous studies have shown that the content
one-treated testosterone compared
testosterone.
growth and how this interacts
within 22 hr of hatching.
terone promoted growth in both sexes and there was no sexual difference
one injections
doses of maternal
of maternal
from eggs with lower testosterone
testosterone
concentrations.
of maternal by testostergrew faster However,
due to asynchronous incubation of clutches. This direct effect of maternal testosterone on growth in combination with a flexible onset of incubation allows to selectively enhance the growth and fitness of individual offspring of a brood. COMP more testosterone
did not compensate
for reduced growth that was caused by later hatching
BIOCHEM PHYSIOL 114A;3:271-276, 1996.
KEY WORDS. Begging, canary, development,
optimal
reproduction,
nestling,
Serinus canuria, sibling compe-
tition
INTRODUCTION The postnatal growth of altricial birds depends on the food they obtain from the parents. The nestlings in a brood compete with each other for parental food which in some large predatory species results in siblicide (8,23,24).
Small passer-
ine birds compete for food by begging (11,12,20,30,33) and differences in the ability to obtain food influence nestling growth ( 16) can lead to death by starvation (28) and affect juvenile survival (19). In both cases hatching asynchrony, which is caused by the onset of incubation
of the clutch
before the last egg is laid, influences the abilities of nestlings to compete with each other for food indirectly by establishing an age/size hierarchy among the nestlings (15,4,18). Recently it was proposed that maternal hormones that are transferred into the egg affect post-natal growth or the competitive abilities of the nestlings directly (31,40). The yolks of the eggs of a canary clutch contain variable doses of maternal testosterone. These doses are low in the first laid eggs and increase in each subsequently laid egg in a clutch (31, H. Schwab1 1993, unpublished results). The Address re@ints reqursts to currentaddress: H. Schwabl, Department of Zoology, Washington State University, Pullman, WA 99164. Recewed 20 July 1995; revised 11 December 1995; accepted 18 December 1995.
here reported experiments tested the hypothesis that maternal testosterone enhances nestling growth. Experiment 1 investigated whether nestlings that hatch from eggs with experimentally increased contents of testosterone grow faster compared to nestlings that hatch from control eggs. Experiment
2 investigated when and how the growth differ-
ences are established.
Experiment
3 evaluated whether the
higher doses of maternal testosterone in the later laid eggs in a clutch can compensate for the adverse effects of hatching asynchrony
on the growth of younger nestlings.
To investigate the effects of high or low doses of testosterone in the egg on nestling growth I injected first and second eggs of the clutch sequence with a dose of testosterone
(100
ng) which is equivalent to the high doses measured in third, fourth or fifth eggs (31). The growth of the nestlings that hatched from these eggs was compared with the growth of nestmates hatching from control-injected first and second eggs that contained only the low maternal doses of testosterone (3 1, H. Schwabl, unpublished results). Confounding effects of egg mass (39) or asynchronous hatching (18) on growth were controlled for by matching the eggs for their mass and by experimentally inducing synchronous hatching. To estimate the relative influences of differential contents of maternal testosterone in the eggs of a clutch and hatching asynchrony on growth rates, I measured the con-
H. Schwabl
272
centrations of maternal testosterone in the eggs and allowed asynchronous incubation and hatching to occur naturally.
MATERIALS
AND METHODS
Subjects were Waterslager canaries bred at the Rockefeller University Field Research Center in Millbrook, New York. Testosterone EXPERIMENT
Injections were stimulated to to long days (14L:lOD) in November
1993 after 4 weeks of exposure to short days (8L:16D).
Eggs
were collected daily within 1 hr after the lights were turned on and replaced with dummy eggs. Eggs were weighed and the first and second eggs in the clutch sequence were selected for the experiment. They contained low doses of maternal testosterone
Data on begging performance
(3 1, H. Schwabl,
unpublished
results).
are compiled
from experi-
ments 1 and 2. To quantify begging the nest was removed from the cage and the number and duration of begging bouts of a newly hatched chick (within an hr of hatching) was recorded. To elicit begging the nest was tapped 5 times over the course of 1 min. A begging bout was defined as the response of the hatchling by raising its head and gaping. When
1. Ten pairs of canaries
breed by exposure
Begging Behavior
a hatchling
begged already when the nest was re-
moved this response was included.
Maternal
Testosterone
EXPERIMENT
Versus Asynchronous
Hatching
3. During the spring breeding
the concentrations
of maternal testosterone
season 1993 in each egg of
9 first clutches were measured. Eggs were taken from the nest in the morning they were laid. They were weighed and
Experimental eggs were injected into the yolk with 100 ng of testosterone in 5 ,~l of sterile sesame oil. This dose corre-
a small yolk sample was removed from each egg for measurement of the testosterone concentration by radioimmunoas-
sponds to the high maternal doses that have been measured in the later laid eggs in a clutch. Control eggs were injected with 5 ~1 of sesame oil. Injections were made through the
say following the methods of Schwab1 (31). Each egg was returned within 1 hr to the nest. At hatching the identity of each hatchling was recorded and each hatchling was indi-
small pole of the egg with a sterile needle (26 g) using illu-
vidually marked. Broods of various sizes (2-4)
mination
ious degrees of asynchronous
from underneath
to ascertain
that the tip of the
and with var-
and synchronous
hatching
needle had penetrated the yolk membranes. The hole in the shell was then patched with flexible wound dressing.
were obtained. Each nestling was weighed when it was 10 days old. During the nestling phase food was available ad
Each egg was individually marked and clutches of 4 eggs were formed consisting of 1 control and 1 experimental egg
libitum and replaced 3 times daily.
of each laying position (1 and 2). Growth differences due to variation in egg size (32,39) were controlled by matching
SEX. The sex of each subject was identified by visual in-
spection of the gonads at an age of 3 to 4 months.
the eggs of each clutch as closely as possible for their mass. Growth differences due to asynchronous hatching (4,18) were eliminated by returning the eggs of a clutch simultaneously into the nest of a female that had stopped laying and was firmly incubating dummy eggs. At the time of expected hatching the nests were inspected
at 2 hr intervals
and the egg from which
each
hatchling hatched identified. Hatchlings were weighed and the lengths of their tarsi were measured. After clipping claws for subsequent identification, hatchlings were returned to their nests. From then on measurements were taken twice daily once in the morning within 1 hr after the lights were switched on and once 4-6 hr before the end of the light phase. Growth curves were plotted for each indi-
STATISTICAL
ANALYSIS.
Data were analyzed by ANOVA
for repeated measures, followed by Student t-tests. Data on begging were analyzed by Mann-Whitney U-tests.
RESULTS Effects of Testosterone-injections on Nestling @owth (Experiment
1)
Treatment of the egg did not influence (testosterone: 62.1%; controls: 60.9%). terone
injections
accelerated
the hatching rate However, testos-
the post-natal
growth
and
EXPERIMENT z. The protocol for this experiment was similar to that of Experiment 1, except that it was performed with 20 breeding pairs during the breeding season in spring 1994 under the changing natural photoperiod of Millbrook,
development compared to synchronously hatching controls (Fig. 1). Although the slopes of fitted logistic growth curves (29) were not significantly different between treatment groups the hatchlings from testosterone-injected eggs reached 50% of their final tarsus length more quickly (3.4 days after hatching compared to 4.3 days in controls: t = 2.44, df = 14, p = 0.029). They also had developed detectable eye slits earlier than hatchlings from control eggs (t = 2.93, df = 11, p = 0.014). This lead in development of approximately 1 day persisted until 80% of the final tarsus length was reached (t = 2.16, df = 14, p = 0.049). Mass and size at the end of the growth phase were not different
NY.
between trearment
vidual (29) and mass and tarsus length interpolated for each 24 hr of age. A constant brood size had been intended, but due to infertility and failure to hatch brood sizes actually ranged from 2 to 4. Fresh food was provided ad libitum 3 times daily throughout the nestling phase.
groups.
273
Maternal Testosterone and Offspring Growth
E
test than control-treated
Although
the duration
2
22
of a single begging bout was not significantly
different be-
y
20
2 s ‘ij
18
tween treatment groups, this resulted in significantly more overall begging of testosterone-treated hatchlings (Fig. 3). Because the testosterone-treated hatchlings also gained
16 14
2 z
12
f
10
=.
8
s z!
6 4
ii
;
6
more weight and grew faster, their begging must have resulted in greater food acquisition (26,35), though this was not directly measured. 0
2
4
ii
6
8
testosterone (8) 10 12 14 16 18
Effects of Testosterone-injections Immediately
This experiment differences
on Embryonic Qrowth (Experiment 2)
after Hatching
addressed the question when and how the
in nestling growth are established.
Maternal Testosterone Contents Versus Hatching Asynchrony (Experiment To evaluate the relative contributions
Age WW
FIG. 1. Growth (means with standard errors) of nestlings from testosterone-injected or control eggs. Arrows on the curve for tarsus length indicate when the testosteronetreated and control nestlings had developed eye slits.
and @owth
hatchlings.
Testosterone
= 2, p < 0.01) and treatment (F = 3.33, df = 2, p < 0.05 and F = 3.67, df = 2, p < 0.05). There were no significant differences in mass or size at hatching. However, at an age of 22 hr the nestlings from the testosterone-injected eggs
of variable maternal
testosterone contents in the eggs of a clutch and asynchronous hatching to differences in nestling growth I measured the concentrations of maternal testosterone in each egg of 9 clutches and compared the mass of pairs of nestlings that had hatched in various intervals from each other. Nine pairs of siblings hatched within 8 hr from each other on the same day. At 10 days of age, the chick from the egg with the higher testosterone content (that is the later-laid egg) weighed more than the chick from the egg with the lower content in 6 of these 9 dyads. This mass difference decreased
did not enhance embryonic development, since testosterone-treated and control eggs hatched after the same duration of incubation (320 ? 3 hr for testosterone-injected eggs and 316 5 4 hr for control eggs, t = 0.88, p = 0.39). Post-hatching mass and tarsus length varied significantly with age (F = 47.94, df = 2 p < 0.01 and F = 49.59, df
3)
2.4
-
IA
Controls (ll=13)
2.2 1
8 -
2.0-
t :
1.6-
6 m”
1.6-
weighed more and had longer tarsi than those from the control eggs (Fig. 2A, B). There was no sex difference in the
lemales @A) testosterone:
A 0
early growth of controls and testosterone accelerated the growth of both sexes (Fig. 2A, B). At the age of 22 hr fe-
-3
males (rod) females (n=6)
D
3
6
9
4 12 15 10 21 24 27
0
3
6
9
12 15 18 21 24 27
males from testosterone-treated eggs were heavier than females from control eggs (t = 2.44, p < 0.02, l-tailed) and males from testosterone-treated eggs had longer tarsi than males from control eggs (t = 1.97, p < 0.05, l-tailed). Thus testosterone had identical effects on the sexes, which resulted in faster growth within the first day after hatching possibly through better food acquisition. To evaluate this possibility I investigated hatched chicks.
Effects of Exogenous 011 Begging Behavior
the begging behavior of the newly
Testosterone at Hatching
Testosterone-treated hatchlings were more likely to beg immediately after hatching than control-treated hatchlings. All 8 testosterone-treated hatchlings begged, while only 3 out of 6 control-treated hatchlings begged. Moreover, testosterone-treated hatchlings tended to beg more often per
J t?
6.8 -
d
6.6 -
6
6.4 -
z
P
3
6.2 6.0 -3
Age (hours) FIG. 2. Mass (A) and tarsus length (B) of hatchlings from testosterone- or control-injected eggs during the first 24 hr after hatching. Means and standard errors for measurements and age at measurement are depicted. Symbols plotted at 24 hr show the mass and tarsus length at the age of 22 hr for each sex and treatment.
H. Schwab1
274
. -6 -0 Testosterone
(n=B)
FIG. 3. Begging behaviour of testosterone-treated and cone trol chicks within 1 hr after hatching. Means and standard errors of the means are shown for the duration of beginning bouts and the overall duratioqof begging. The number of begging bouts per test are depicted as means with ranges. Significance levels for differences between treatment groups are indicated above bars for controls.
with seniority
of the chick
that
hatched from the egg with less testosterone (Fig. 4). In contrast, in only 1 out of 8 cases in which hatching occurred 20 or more hr apart the chick testosterone content weighed tosterone in the egg resulted pared to a chick that hatched ing 6-33
hr earlier
from the egg with the high more. On average, more tesin a 22% greater mass comwithin 4 hours, while hatch-
from an egg with less testosterone
resulted in 15% greater mass (Fig. 4, insert).
DISCUSSION
The results of the 3 experiments support the hypothesis that maternal testosterone in the eggs enhances post-natal growth and development. This adds differential doses of maternal testosterone in the eggs to asynchronous incubation (18) and differences in egg size (32,39) as maternal factor which can cause differential
growth among the nestlings of
a brood. I will first discuss the mechanisms by which testosterone may act and then the function of this hormonal influence on nestling growth.
4
8
12
16
difference
20
24
28
32
36
(hours)
FIG. 4. Mass difference between pairs of chicks at 10 days of age that had hatched in various intervals from each other from eggs with a high or low concentration of maternal testosterone. Abscissa: Difference in age between the chick that hatched from the egg with the low concentration of maternal testosterone and the chick that hatched from the egg with the high testosterone (positive values: the chick from the egg with the high testosterone concentration is younger). Ordinate: Mass of the chick from the egg with more maternal testosterone minus mass of the chick from the egg with less testosterone (positive values: the chick from the egg with more testoserone weighs more; negative values: it weighs less). The solid line is the linear regression of this mass difference versus the age difference for pairs of chicks that hatched on the same day within 8 hr (filled circles). Open circles represent data for pairs of chicks which hatched on different days. Insert: Left bar: Relative mass difference in percent (mean with standard error) that resulted from a higher testosterone content in the egg compared to a synchronously hatching chick (within 4 hr). Right bar: relative mass difference that resulted from hatching earlier (6-33 hr) compared to a chick from an egg with more maternal testosterone.
begging because yolk reserves are depleted at hatching. Support for this possibility comes from two lines of evidence: (a) Testosterone enhances the metabolism in adult birds and (b) metabolic deficits (hunger) intensify begging, at least in older nestlings (33,26). The perception of cues that elicit begging: Before we can answer the question if testosterone enhances the development of perceptual processes, we need more information about the nature of the cues that elicit begging and interactions of chicks and parents at hatching. It is possible that testosterone enhances begging by an influence
Mechanisms The mechanism
,‘f’~‘~‘~‘,‘~~~
0
Age
Egg treatment
rapidly and significantly
.I
-4
Control(n-6)
by which maternal testosterone
influences
nesting growth differs from those of asynchronous hatching or egg size variation which both act indirectly by causing differences in the age or size of nest mates (18,39, H. Schwabl, unpublished results). Testosterone has a different influence on the hatchling which results in its faster growth, probably by greater food acquisition through intensified food begging behavior. Begging involves several processes that testosterone might modify. 1. Metabolic needs: It is possible that testosterone increases the embryo’s metabolism, causing earlier and intensified
on attention or persistency, which have been shown to be improved in the newly hatched chick (G&s domestictls) by exposure to testosterone (1,2,5). The actual motor pattern of raising the head, extending the neck and gaping may be enhanced by effects on the development of muscles or the differentiation of motor control systems. The observations that exogenous androgens are accumulated in the avian embryonic spinal cord and spinal ganglia suggesting the presence of receptors (27) and that testosterone enhances the proliferation of motor neurons (14) lend some support to such an expla-
275
Maternal Testosterone and Offspring Growth
nation.
Presently,
it is not possible to distinguish
tween these possibilities.
However,
be-
it can be excluded
which the growth and fitness of the nestlings from the first laid eggs can be enhanced.
But the alternative
of synchro-
that the faster growth of testosterone-treated hatchlings resulted simply from the elimination of a sex difference (37) because (a) there was no sex difference in the
nous incubation cannot selectively enhance the fitness of the nestlings from later eggs. I suggest that the increasing
growth of controls,
nation with synchronous incubation provide a mechanism to optimize reproduction by selectively enhancing the
hanced
and (b) testosterone
injections
en-
the growth of both sexes.
in the subsequently laid eggs in combi-
growth and fitness of the offspring from the later laid eggs in a clutch.
Function The results demonstrate
that the differential doses of mater-
nal testosterone in the eggs of a clutch result in nestling growth differences, at least if hatching is nearly synchronous. They do not suggest, however, that the higher doses of maternal testosterone in the later laid eggs compensate for growth differences that result from asynchronous hatching, although it cannot be ruled out that testosterone somewhat mitigates the disadvantages of hatching later. At present it is parsimonious to assume (a) that the variable doses of maternal testosterone
in the egg cause growth differences
among synchronously hatching nestlings that can be as great as those resulting from asynchronous hatching; and (b) that asynchronous hatching overrides this hormonal influence. Since asynchronous hatching depends to a large extent on the onset of incubation of a clutch (4,18) the incubation strategy will decide whether hatching asynchrony or differences of the doses of maternal testosterone prevail to cause differential
doses of testosterone
growth among the nestlings of a brood. In the
canaries of this study the onset of incubation changes in subsequent clutches (H. Schwabl, unpublished results) and such changes with the progression of the breeding season reported for several have been other species (3,10,13,21,22,38). In the last part I discuss why it may be adaptive to influence nestling growth by two different mechanisms: on the one hand, by asynchronous incubation and, on the other hand, by higher contents of maternal testosterone in each subsequently laid egg in a clutch and synchronous incubation. Hypotheses of optimal clutch and brood size emphasize that selection
maximizes the number and size of offspring
that are produced (17,18,24,36). This implies that all offspring of a brood are of equal value to the parent, regardless from which egg they hatch which might, however, not be the case (25). For example, in European kestrels (F&o tinninculus) the value of an offspring that is produced at a given time of year depends on its sex (6). At the same time the primary sex ratio varies systematically with the laying order of the eggs in this species (7) as well as in other species including the canary (H. Schwabl, unpublished results). If the value of an offspring depends on its sex and the sex ratio varies predictably with egg sequence, it will be adaptive to evolve mechanisms by which the fitness of the offspring of specific eggs in the clutch can be selectively influenced. Asynchronous incubation is a well known mechanism by
I thankDam Jackson, Sharon Sepe, and Helen Ecklund for their help with breeding and Sibylle Kbnig, Douglas Mock, and Fernando Notrebohm for critical comments on earlier versions of this manuscript and 3 anonymous reviewers for valuable comments. The work was supported by the Mary Flagler Cary Charitable Trust, Public Health Service Grant MH 49877, and a g-rantfrom the Harry Frank Guggenheim Foundation.
References 1. Andrew, R.J. Effects of testosterone on the behaviour of the domestic chick: I. Effects present in males but not in females. Anim. Behav. 23:139-155;1975. 2. Andrew, R.J. Effects of testosterone on the behaviour of the domestic chick. II. Effects present in both sexes. Anim. Behav. 23:156-168;1975. 3. Beukeboom, L.; Dijkstra, C.; Daan, S.; Meijer, T. Seasonality of clutch size determination in the kestrel F&o tinnunculus. Om. Stand. 19:41-48;1988. 4. Clark, A.B.; Wilson, D.S. Avian breeding adaptations: hatching asynchrony, brood reduction and nest failure. Quart. Rev. Biol. 56:257-277;1981. 5. Clifton, P.G.; Andrew, R.J.; Brighton, L. Gonadal steroids and attentional mechanism in young domestic chicks. Physiol. Behav. 43:441-446;1988. 6. Daan, S.; Dijkstra, C.; Drent, R.H.; Meijer, T. Food supply and the annual timing of avian reproduction. In: Qiiellet, H. (ed). Acta XIX Congressus Intemationalis Ornithologici. Ottawa: University of Ottawa Press; 1989:392-407. 7. Dijkstra, C.; Daan, S.; Buker, J.B. Adaptive seasonal variation in the sex ratio of kestrel broods. Funct. Ecol. 4:143-147; 1990. 8. Drummond, H. Parent-offspring conflict and brood reduction in the Pelicaniformes. Colonial Waterbirds lO:l-15;1987. 9. Feuerbacher, 1.; Prinzinger, R. The effect of the male sex-hormone testosterone on body temperature and energy metabolism in male Japanese Quail (Coturnix co&x japonica). Comp. Biochem. Physiol. 70A:247-250;1981. 10. Gibb, J. The breeding biology of the great and blue titmice. Ibis 92:507-539;1950. 11. Gottlander, K. Parental feeding behaviour and sibling competition in the pied flycatcher Ficedula hypoleuca. Orn. Stand. 18:269-276;1987. 12. Greig-Smith, P.J. Weight differences, brood reduction, and sibling competition among nestling stonechats. Saxicola torquuta. J. Zool. (London) 205:453-465;1985. 13. Hafthom, S. Incubation during the egg-laying period in relation to clutch size and other aspects of reproduction tn the great tit Parus major. Om. Stand. 12:169-185;1981. CD. Androgen increases the 14. Hauser, K.F.; Torand-Allerand, number of cells in fetal mouse spinal cord cultures: implications for motomeuron survival. Brain Res. 485:157-164;1989. 15. Lack, D. The natural regulation of animal numbers. Oxford: Clarendon Press; 1964.
276
16. Liihrl, H. Das Nesthakchen als biologisches Problem. J. Omithol. 109:383-395;1968. 17. Magrath, R.D. Hatching asynchrony and reproductive success in the blackbird. Nature 339:536-538;1989. 18. Magrath, R.D. Hatching asynchrony in altricial birds. Biol. Rev. 65:587-622;1990. 19. Magrath, R.D. Nestling weight and juvenile survival in the blackbird, Turdus merula. J. Anim. Ecol. 60:335-351;1991. 20. McRae, S.B.; Weatherhead, P.J.; Montgomerie, R. American robin nestlings compete by jockeying for position. Behav. Ecol. Sociobiol. 33:101-106;1993. 21. Mead, P.S.; Morton, M.L. Hatching asynchrony in the whitecrowned sparrow (Zononichia leucophrys oriunthu): a selected or inherited trait? Auk 102:781-792;1985. 22. Meijer, T. Incubation development and clutch size in the starling. Orn. Stand. 21:163-168;1990. 23. Mock, D.W. Infanticide, siblicide, and avian nestling mortality. In: Hausfater, G.; Hrdy, S.B., (eds). Infanticide: comparative and evolutionary perspectives. New York: Aldine; 1984: 3-30. 24. Mock, D.W. Siblicide, parent-offspring conflict, and unequal parental investment by egrets and herons. Behav. Ecol. Sociobiol. 20:247-256~1987. 25. Mock, D.W.; Forbes, L.S. Parent-offspring conflict: a case of arrested development. TREE 7:409-413;1992. 26. Price, K.; Ydenberg, R. Begging and provisioning in broods of asynchronously-hatched yellow-headed blackbird nestlings. Behav. Ecol. Sociobiol. 37:201-208;1995. 27. Reid, F.A.; Gasc, J.-M; Stumpf, W.E.; Sar, M. Androgen target cells in spinal cord, spinal ganglia, and glycogen body of chick embryos. Exp. Brain Res. 44:243-248;1981. 28. Richter, W. Hatching asynchrony: the nest failure hypothesis and brood reduction. Am. Nat. 120:828-832;1982.
H. Schwab1
29. Ricklefs, R.E. A graphical method of fitting equations to growth curves. Ecology 48:978-983;1967. 30. Ryden, 0.; Bengtson, H. Differential begging and locomotory behavior by early and late hatched nestlings affecting distribution of food in asynchronously hatched broods of altricial birds. Z. Tierpsychol. 53:209-224;1980. 3 1. Schwabl, H. Yolk is a source of maternal testosterone for developing birds. Proc. Natl. Acad. Sci. USA 90:11449-l 1450; 1993. 32. Slagsvold, T.; Sandvik, T.; Rofstad, G., Lorentsen, 0.; Husby, M. On the adaptive value of intraclutch egg-size variation in birds. Auk 101:685-697;1984. 33. Smith, H.G.; Montgomerie, R. Nestling American robins compete with siblings by begging. Behav. Ecol. Sociobiol. 29: 307-312;1991. 34. Smith, CC.; Fretwell, S.D. The optimal balance between size and number of offspring. Am. Nat. 108:499-506;1974. 35. Stamps, J.; Clark, A.; Arrowood P.; Kus B. Begging behaviour in budgerigars. Ethology 81:177-192;1988. 36. Stearns, SC. The evolution of life histories. Oxford: Oxford Universiry Press, 1992. 37. Teather, K.L. An experimental study of competition for food between male and female nestlings of the redwinged blackbird. Behav. Ecol. Sociobiol. 31:81-87;1992. 38. van Balen, J.H. A comparative study of the breeding ecology of the Great tit Parus major in different habitats. Ardea 61: l-93;1973. 39. Williams, T.D. Intra-specihc variation in egg size and egg composition in birds: effects on offspring fitness. Biol. Rev. 68:35-59;1994. 40. Winkler, D.W. Testosterone in egg yolks: an ornithologist’s perspective. Proc. Natl. Acad. Sci. USA 90:11439-11441; 1993.