## Abstract

Fisher's theory predicts equal sex ratios at the end of parental care if the costs and benefits associated with raising each sex of offspring are equal. In raptors, which display various degrees of reversed sexual size dimorphism (RSD; females the larger sex), sex ratios biased in favor of smaller males are only infrequently reported. This suggests that offspring of each sex may confer different fitness advantages to parents. We examined the relative returns associated with raising each sex of offspring of the brown falcon Falco berigora , a medium-sized falcon exhibiting RSD (males approximately 75% of female body mass) and subsequent sex ratios. Female nestlings hatched either first or second did not receive more food nor did they hatch from larger eggs or remain dependent on parents for longer periods than male offspring from these hatch orders. Together with previous studies this result indicates that even in markedly dimorphic species, the required investment to raise the larger sex is likely to be less than that predicted by body size differences alone. Moreover, among last-hatched nestlings, both sexes faced a reduced food allocation and suffered a slower growth rate and thus final body size, with a concurrent increased probability of mortality. For last-hatched females the reduction in food allocation was more marked, with complete mortality of all last-hatched female nestlings monitored in this study. Once independent, males of any size but only larger females are likely to be recruited into the breeding population. The sex-biased food allocation among last-hatched offspring favoring males thus reflects the relative returns to parents in raising a small member of each sex.

Sex allocation theory has been one of the most successful areas of evolutionary ecology, predicting the level of parental investment and associated sex ratio skews observed in many invertebrates with considerable accuracy ( Charnov, 1982 ). However, among mammals ( Clutton-Brock and Iason, 1986 ) and, in particular, birds ( Clutton-Brock, 1986 ; Ewen et al., 2004 ; Gowaty, 1993 ; Sheldon, 1998 ), results have been less compelling. While apparent manipulation independent of differential mortality has recently been reported ( Appleby et al., 1997 ; Heinsohn et al., 1997 ; Komdeur et al., 1997 ), biases are usually slight and often inconsistent between or even temporally within populations, reflecting the numerous and complex selective pressures likely to be influencing sex allocation on many scales in higher vertebrates ( Byholm et al., 2002 ; Charnov, 1982 ; Cockburn et al., 2002 ; Frank, 1990 ).

Sex allocation patterns among raptor populations have attracted considerable research attention ( Krackow, 1993 ; Newton, 1979 ; Olsen and Cockburn, 1991 ), as most raptorial birds exhibit reversed sexual size dimorphism (RSD; females the larger sex), suggesting larger daughters may be more costly as they would require more parental investment in the form of food requirements than sons. Fisher's hypothesis ( Fisher, 1930 ) consequently predicts biased population sex ratios at the end of parental care in favor of males, the cheaper sex, assuming that the benefits of raising either sex are equal ( Charnov, 1982 ; Frank, 1990 ). However, male-biased sex ratios have only rarely been observed in raptors ( Brommer et al., 2003 ; Rosenfield et al., 1996 ), with sex ratios at parity ( Negro and Hiraldo, 1992 ) or even biased towards females ( Appleby et al., 1997 ; Tella et al., 1996 ) equally common. One explanation for this discrepancy is that body size may be a poor indicator of potential parental costs, as increased investment in the form of increased food requirements of the larger sex has been observed in some raptors ( Anderson et al., 1993b ; Krijgsveld et al., 1998 ; Riedstra et al., 1998 , see also Laaksonen et al., 2004 ) but not others ( Boulet et al., 2001 ; Brommer et al., 2003 ; Collopy, 1986 ).

By contrast, with limited evidence for population-level biases, a number of authors have reported sex-biased hatch or laying order sequences ( Bednarz and Hayden, 1991 ; Bortolotti, 1986 ; Leroux and Bretagnolle, 1996 ; Olsen and Cockburn, 1991 ). The prevailing sex may change according to lay date within seasons, with the sex most likely to breed at 1 year of age suggested to be overproduced in early broods ( Dijkstra et al., 1990 ). Finally, some raptors produce more of the smaller sex (males) when parents are in poor condition or prey supplies are low ( Aparicio and Cordero, 2001 ; Wiebe and Bortolotti, 1992 ).

In an attempt to bring some focus to these apparently disparate research findings, we investigated sex allocation patterns in the brown falcon Falco berigora , a medium-sized falcon exhibiting moderate RSD (males: 486 g ± 5 SE, n = 69; females: 658 g ± 7 SE, n = 91; McDonald et al., 2005 ). The species has a socially monogamous breeding system with individual pairs maintaining exclusive all-purpose territories throughout the year ( McDonald, 2004 ; McDonald et al., 2003 ). Larger female brown falcons are more likely to gain breeding territories, but large males gain no such advantage ( McDonald et al., 2005 ). Young are not siblicidal.

Previous research has demonstrated that there are no differences between the sexes in the duration of parental care through the nestling and postfledging periods, ( McDonald, 2004 ). However, the sex ratio, investment in eggs, or investment in the amount of food provided to young of different sexes could still differ. In this study we examined the sex ratio of broods throughout the season, the investment in eggs containing male and female embryos, and the provisioning of male and female chicks and its consequences for growth, survival, and recruitment.

## METHODS

### Study area and general field methods

The study was conducted between August 2000 and June 2002, 35 km southwest of Melbourne, southeast Australia, at the Western Treatment Plant, Werribee (38°0′ S, 144°34′ E), the adjacent Avalon Airport (38°2′ S, 144°28′ E), and small areas of surrounding private land. Details of the approximately 150 km 2 study site ( Baker-Gabb, 1984 ) and study population have been provided elsewhere ( McDonald, 2003a , b , 2004 ; McDonald et al., 2003 , 2004 , 2005 ).

### Egg volume and food allocation according to sex and hatch order

All nests within the study area were found and their contents determined directly. Eggs were marked to facilitate individual recognition and measured to the nearest 0.01 cm with vernier calipers at the longest and widest point; volume was estimated using the formula 0.51 × length × (breadth) 2 in cm ( Hoyt, 1979 ). For several days prior to hatching, nests were visited daily and when possible nestlings matched to the eggs from which they hatched. During these visits hatch order was also ascertained either directly or using features described in McDonald (2003b) . Generally the first and second nestlings to hatch emerge from eggs within 24 h of each other, last-hatched nestlings follow 1–2 days later. Once emerged from eggs, nestlings were marked until they reached an appropriate age for banding with a small piece of Micropore tape (3M; Pymble, New South Wales, Australia), loosely folded around either the left or right humerus. Once hatched, a small blood sample (approximately 10–20 μl) was collected from each nestling's alar vein and a polymerase chain reaction (PCR)/ Hae III digest reaction used to assign sex to each nestling ( Griffiths et al., 1998 ). Chicks were fitted with color band combinations prior to fledging at approximately 28 days of age.

To determine the amount of prey required to raise nestlings of each sex, small cameras with infrared lights (Model 43150674; Radio Parts Group, Melbourne) were placed at nests for a 48-h observation period during each of the early- (eldest nestling 0–14), mid- (eldest nestling 15–27), and late-nestling phase (eldest nestling greater than 27 days old). They were powered by deep cycle batteries (Besco N70T) and connected to time-lapse video recorders run at 1/8 normal speed (Hitachi VT1200E); this minimized visits to the nest tree to 24-h intervals to change batteries and tapes. During each feeding event (prey brought to the nest cup and distributed among offspring by the female parent) the closest nestling relative to the prey being offered was noted at the beginning of each feed. For each feeding event the number of mouthfuls of prey provided to each nestling was also recorded, with the nestling receiving the first mouthful noted as being ‘fed first.’

### Recruitment of offspring into the breeding population

To determine possible benefits associated with raising one sex over the other, recruitment rates of banded offspring were monitored throughout the study period.

### Growth and survival to independence

To determine the biological significance of detected biases in resource allocation among broods monitored with surveillance cameras, wing length, head to bill length, and body mass were measured weekly as per the methods outlined in McDonald (2003b) . Survival to fledging and independence for each nestling were also monitored during weekly visits to the nest site or biweekly visits to the natal territory postfledging.

### Statistical analyses

Sex ratio data were assessed for deviances from parity using two-tailed binomial tests. To assess differences in egg volume between male and female eggs, after controlling for clutch and hatch order, we used linear mixed models that were evaluated with restricted maximum likelihood (REML) procedures. The probability of a given nestling fledging according to sex and hatch order was assessed using a generalized mixed model with a binomial error distribution ( Genstat Committee, 1993 ). Contingency tables were used to assess biases in nestlings recorded as being fed first or closest to prey items during feeding events according to sex and hatch order.

To model the size of nestlings and number of mouthfuls of food that chicks received during feeding events we used REML procedures. Random effects in the REML models were the year of the breeding attempt, the pair raising the chicks, the observation session (number of times cameras had been used at that nest, e.g., first, second, so on), the feeding event within the observation session, and the identity of the chick being fed. The number of mouthfuls supplied to each chick during a feeding event was ln( X + 1) transformed, in order to account for zeros and generate a normalized distribution of residuals. Fixed terms examined in the model included terms for brood age (early, mid, or late in the nestling phase), breeding season, Julian hatch date of the eldest nestling, and observation date for each brood, as well as nestling sex, hatch order, and whether or not the nestling survived to fledging.

Assessments of nestling growth rates were confined to the periods 7–20 days old and greater than 20 days old, over which growth is linear ( McDonald, 2003b ). We utilized a random model incorporating pair number, season, and the identity of the chick within the brood, with the main models used incorporating hatch order, brood size, and nestling age. The latter was calculated separately for each chick at every measurement using the methods outlined in McDonald (2003b) .

Initial models fitted included all explanatory terms and biologically meaningful two-way interactions, with terms then dropped in a step-wise fashion by examining the change in deviance between the full model and the submodel fitted, until only significant terms and/or interactions remained. For simplicity, interaction terms are presented only when significant. All calculations were carried out using Genstat 5 Release 4.21 ( Genstat Committee, 1993 ).

## RESULTS

### Sex ratios

Sex ratios did not differ between years across the study population ( Table 1 ), and both years of data collection were combined in further analyses. Sex ratios did not deviate significantly from parity at hatching, fledging, or independence among all chicks combined or within each hatch order ( Table 1 ), though it is notable that all female last-hatched offspring perished prior to fledging, suggesting that this effect might become significant were more data available ( Table 1 ).

Table 1

Sex ratio data for the study population as a whole and within each hatch order

Parameter

n

Sex ratio

p

Year of study
2000 hatching sex ratio 40 0.6 .27
2001 hatching sex ratio 24 0.54 .84
All offspring
Hatching sex ratio 64 0.58 .26
Fledgling sex ratio 49 0.61 .15
Sex ratio at independence 38 0.63 .13
First-hatched offspring
Hatching sex ratio 27 0.67 .12
Fledgling sex ratio 24 0.67 .15
Sex ratio at independence 17 0.71 .14
Second-hatched offspring
Hatching sex ratio 24 0.42 .54
Fledgling sex ratio 20 0.45 .82
Sex ratio at independence 17 0.47
Third-hatched offspring
Hatching sex ratio 13 0.69 .27
Fledgling sex ratio .06
Sex ratio at independence

4

1

.13

Parameter

n

Sex ratio

p

Year of study
2000 hatching sex ratio 40 0.6 .27
2001 hatching sex ratio 24 0.54 .84
All offspring
Hatching sex ratio 64 0.58 .26
Fledgling sex ratio 49 0.61 .15
Sex ratio at independence 38 0.63 .13
First-hatched offspring
Hatching sex ratio 27 0.67 .12
Fledgling sex ratio 24 0.67 .15
Sex ratio at independence 17 0.71 .14
Second-hatched offspring
Hatching sex ratio 24 0.42 .54
Fledgling sex ratio 20 0.45 .82
Sex ratio at independence 17 0.47
Third-hatched offspring
Hatching sex ratio 13 0.69 .27
Fledgling sex ratio .06
Sex ratio at independence

4

1

.13

Sex ratios presented as the proportion of offspring that are male, reported p values are for two-tailed binomial tests.

### Sex differences in egg volume

Female chicks did not hatch from eggs of a larger volume than their male counterparts ( F1,19 = 1.17, p = .29). Egg volume was unaffected by hatch order ( F2,19 = 0.69, p = .52) or the interaction between hatch order and sex ( F2,19 = 0.42, p = .67; mean volume = 42.4 cm 3 ± 0.8 SE, n = 26).

### Sex differences in food allocation

Delivery of a total of 1193 mouthfuls of food from 567 prey items was recorded during video surveillance of 27 broods (17 in 2000 and 10 in 2001). The number of mouthfuls supplied to nestlings within a brood during feeding events was influenced by three factors. First, nestlings in smaller broods were fed more per capita than those in larger broods ( Figure 1a ; Table 2 ). Second, nestlings older than 27 days tended to get less mouthfuls of food than younger classes of nestlings ( Figure 1b ; Table 2 ). Finally, a sex by hatch-order interaction existed ( Figure 1c ; Table 2 ). While male and female nestlings that hatched either first or second received similar amounts of food per feeding event ( Figure 1c ) and last-hatched males received similar proportions to earlier-hatched males, last-hatched females received far fewer mouthfuls than any other class of young ( Figure 1c ).

Figure 1

The number of mouthfuls of food supplied to brown falcon nestlings by parents according to (a) brood size, (b) nestling age, and (c) a sex by hatch-order interaction. Bars indicate REML model predictions ( Table 2 ), data points and error bars represent means ± 1SE, with numbers in parentheses representing sample sizes.

Figure 1

The number of mouthfuls of food supplied to brown falcon nestlings by parents according to (a) brood size, (b) nestling age, and (c) a sex by hatch-order interaction. Bars indicate REML model predictions ( Table 2 ), data points and error bars represent means ± 1SE, with numbers in parentheses representing sample sizes.

Table 2

Summary of analyses of (a) random and (b) fixed effects influencing the number of mouthfuls supplied to nestlings within brown falcon broods during 1193 feeding events assessed using restricted maximum likelihood modeling

(a) Random effects

Term of interest

Component

SE

p Value

Year 0.008 0.02 >.05
Pair/observation 0.09 0.04 <.05
Pair/observation/feeding event 0.3 0.05 <.001
(a) Random effects

Term of interest

Component

SE

p Value

Year 0.008 0.02 >.05
Pair/observation 0.09 0.04 <.05
Pair/observation/feeding event 0.3 0.05 <.001
(b) Fixed effects

Term of interest

df

Change in deviance

p Value

Brood size 1 29.0 <.001
Sex × hatch order 2 9.5 .009
Nestling age 2 7.3 .03
Sex 1 0.8 .37
Hatch order 2 16.8 <.001
Julian hatch date 1.1 .29
Nestling survived

1

0.04

.84

(b) Fixed effects

Term of interest

df

Change in deviance

p Value

Brood size 1 29.0 <.001
Sex × hatch order 2 9.5 .009
Nestling age 2 7.3 .03
Sex 1 0.8 .37
Hatch order 2 16.8 <.001
Julian hatch date 1.1 .29
Nestling survived

1

0.04

.84

Only unbound random terms are presented. Change in deviance statistics for each term initially included in the model are presented, terms included in the model are emboldened. Nestling age divided into three discrete categories: 0–14, 15–27 and greater than 27 days old.

To further investigate causes behind this disparity we also examined nestling position during feeding events relative to both sex and hatch order, excluding feeding events from single-chick broods ( n = 90). Nestling sex did not influence whether a nestling was recorded as being closest or fed first during feeding events both within and across all hatch orders ( Table 3 ). Biases according to nestling hatch order were apparent, however, with third- or last-hatched nestlings of both sexes less often than expected obtaining the position closest to prey or receiving the first mouthful during feeding events ( Figure 2 ; Table 3 ).

Figure 2

Observed (light bars) and expected (shaded bars) proportion of brown falcon nestlings from different hatch orders classed as either (a) closest to prey or (b) fed first during feeding events. Both graphs depict a significant departure from expected results ( p < .01 according to Bonferroni adjustment), numbers in parentheses indicate the number of feeding events observed.

Figure 2

Observed (light bars) and expected (shaded bars) proportion of brown falcon nestlings from different hatch orders classed as either (a) closest to prey or (b) fed first during feeding events. Both graphs depict a significant departure from expected results ( p < .01 according to Bonferroni adjustment), numbers in parentheses indicate the number of feeding events observed.

Table 3

Results of contingency table analyses assessing the distribution of nestlings that were closest to prey or supplied with the first mouthful of food during feeding events, according to nestling sex (within and across hatch orders) and hatch order

Parameter

χ 2

df

p

Closest nestling versus nestling sex
First hatched only 0.2 .64
Second hatched only 0.2 .70
Third hatched only 0.01 .91
All hatch orders 0.8 .38
Closest nestling versus hatch order
Both sexes 9.3 2 .009
Nestling fed first versus nestling sex
First hatched only 0.08 .78
Second hatched only 0.9 .35
Third hatched only 0.2 .65
All hatch orders 1.1 .30
Nestling fed first versus hatch order
Both sexes

28.1

2

<.0005

Parameter

χ 2

df

p

Closest nestling versus nestling sex
First hatched only 0.2 .64
Second hatched only 0.2 .70
Third hatched only 0.01 .91
All hatch orders 0.8 .38
Closest nestling versus hatch order
Both sexes 9.3 2 .009
Nestling fed first versus nestling sex
First hatched only 0.08 .78
Second hatched only 0.9 .35
Third hatched only 0.2 .65
All hatch orders 1.1 .30
Nestling fed first versus hatch order
Both sexes

28.1

2

<.0005

Significant terms according to Bonferroni adjustment ( p < .01) are given in bold.

### Consequences of differential food allocation for nestlings

Using a binomial generalized linear model, nestling mortality did not differ between sexes

$$(\mathrm{{\chi}}_{1}^{2}{=}1.9;{\,}p{=}.2).$$
Similarly yearly differences in mortality rates were not apparent
$$(\mathrm{{\chi}}_{1}^{2}{=}1.8;{\,}p{=}.18).$$
A significant hatch-order effect was found
$$(\mathrm{{\chi}}_{2}^{2}{=}10;{\,}p{=}.007;$$
Figure 3 ), with last-hatched nestlings of both sexes experiencing lower survival probabilities than their earlier-hatched counterparts. An interaction between sex and hatch order did not reach significance
$$(\mathrm{{\chi}}_{2}^{2}{=}2.6;{\,}p{=}.3;$$
Figure 3 ), however, while half the last-hatched males survived, none of their female counterparts did so ( Figure 3 ).

Figure 3

The proportion of brown falcon males (light bars) and females (dark bars) that survived to fledge in nests monitored with surveillance cameras according to nestling hatch order. Numbers above bars indicate sample sizes.

Figure 3

The proportion of brown falcon males (light bars) and females (dark bars) that survived to fledge in nests monitored with surveillance cameras according to nestling hatch order. Numbers above bars indicate sample sizes.

Nestling growth rates were assessed with REML modeling to determine the biological impact of differences in mouthfuls received during feeding events. Males tended to grow fastest (head to bill and body mass) in larger broods. However, female nestlings suffered a reduced head to bill growth rate in larger broods ( Table 4 ). Between 7 and 20 days, hatch-order effects were also marked with both male and female last-hatched nestlings having a lower body mass than earlier-hatched nestlings ( Figure 4a ). Only male nestlings could be assessed from 21 days of age through to fledging as no last-hatched females survived to this age. Among male nestlings greater than 20 days old, head to bill lengths were smaller in last-hatched individuals ( Figure 4b ; Table 4a ).

Figure 4

Residuals from a regression between body mass (a) or head to bill length (b) and nestling age according to hatch order. Nestlings were aged between 7 and 20 days (a) or were between 21 days of age and fledging (b). Bars represent means ± 1SE, with numbers in parentheses representing sample sizes. All three regressions significant at p < .05 (data not presented).

Figure 4

Residuals from a regression between body mass (a) or head to bill length (b) and nestling age according to hatch order. Nestlings were aged between 7 and 20 days (a) or were between 21 days of age and fledging (b). Bars represent means ± 1SE, with numbers in parentheses representing sample sizes. All three regressions significant at p < .05 (data not presented).

Table 4

Summary of analyses of (i) random and (ii) fixed effects influencing the growth of wing length, head to bill length and body mass in brown falcon nestlings aged 7–20 or greater than 20 days old using restricted maximum likelihood modeling

(i) Random effects

(a) Males

(b) Females

Term of interest

Component

SE

p Value

Component

SE

p Value

Nestlings aged 7–20 days old
(i) Wing length
Pair 40.4 17.3 <.05 — — —
Pair/hatch order 5.8 6.7 >.05 33.1 14.4 <.05
Year 9.9 16.9 >.05 25.1 39.6 >.05
(ii) Head to bill length
Pair — — — — — —
Pair/hatch order 0.8 0.9 >.05 0.9 1.1 >.05
Year — — — 0.2 0.6 >.05
(iii) Body mass
Pair 709.5 326.7 <.05 499 706 >.05
Pair/hatch order — — — 45 773 >.05
Year — — — 165 495 >.05
Nestlings older than 20 days
(iv) Wing length
Pair 37.3 16.0 <.05 — — —
Pair/hatch order — — — — — —
Year 4.5 8.7 >.05 — — —
(v) Head to bill length
Pair 0.1 0.4 >.05 — — —
Pair/hatch order — — — — — —
Year 0.2 0.5 >.05 — — —
(vi) Body mass
Pair 232.2 355.1 >.05 — — —
Pair/hatch order 389.5 398.9 >.05 — — —
Year — — — — — —
(i) Random effects

(a) Males

(b) Females

Term of interest

Component

SE

p Value

Component

SE

p Value

Nestlings aged 7–20 days old
(i) Wing length
Pair 40.4 17.3 <.05 — — —
Pair/hatch order 5.8 6.7 >.05 33.1 14.4 <.05
Year 9.9 16.9 >.05 25.1 39.6 >.05
(ii) Head to bill length
Pair — — — — — —
Pair/hatch order 0.8 0.9 >.05 0.9 1.1 >.05
Year — — — 0.2 0.6 >.05
(iii) Body mass
Pair 709.5 326.7 <.05 499 706 >.05
Pair/hatch order — — — 45 773 >.05
Year — — — 165 495 >.05
Nestlings older than 20 days
(iv) Wing length
Pair 37.3 16.0 <.05 — — —
Pair/hatch order — — — — — —
Year 4.5 8.7 >.05 — — —
(v) Head to bill length
Pair 0.1 0.4 >.05 — — —
Pair/hatch order — — — — — —
Year 0.2 0.5 >.05 — — —
(vi) Body mass
Pair 232.2 355.1 >.05 — — —
Pair/hatch order 389.5 398.9 >.05 — — —
Year — — — — — —
(ii) Fixed effects

(a) Males

(b) Females

Term of interest

Change in deviance

df

p Value

Change in deviance

df

p Value

Nestlings aged 7–20 days old
(i) Wing length (Wl, mm) ♂Wl = 71.2 + 6.3Age   ♀Wl = 76.9 + 7.1Age
Age 1247.8 1 <.001 1105.5 1 <.001
Hatch order 4.8 .09 0.67 .7
Brood size 3.13 .08 1.7 .2
(ii) Head to bill length (Hb, mm) ♂Hb = 50 + 1.3Age + 1 Brood size   ♀Hb = 52.02 + 1.5Age − 1.7 Brood size
Age 337.7 1 <.001 298.8 1 <.001
Hatch order 2.64 .3 0.6 .8
Brood size 5.63 1 .02 5.98 1 .02
(iii) Body mass (Mb, g) ♂Mb = 251 + 24Age + 21.4 Brood size + 14.8 (second hatched) − 21.2 (third hatched)   ♀Mb = 303.6 + 28.1Age − 19.5 (second hatched) − 94.82 (third hatched)
Age 439.3 1 <.001 203.5 1 <.001
Hatch order 8.36 2 .02 12.2 2 .002
Brood size 4.21 1 .04 2.75 .1
Nestlings older than 20 days
(iv) Wing length ♂Wl = 162.4 + 6.7Age
Age 1528.5 1 <.001 — — —
Hatch order 2.17 .3 — — —
Brood size 1.7 .2 — — —
(v) Head to bill length ♂Hb = 61.6 + 0.4Age − 1.3 (second hatched) − 1.9 (third hatched)
Age 41.9 1 <.001 — — —
Hatch order 8.8 2 .01 — — —
Brood size 0.4 .6 — — —
(vi) Body mass ♂Mb = 433.2 + 7.6Age
Age 39.1 1 <.001 — — —
Hatch order 4.02 .1 — — —
Brood size

1.3

1

.3

(ii) Fixed effects

(a) Males

(b) Females

Term of interest

Change in deviance

df

p Value

Change in deviance

df

p Value

Nestlings aged 7–20 days old
(i) Wing length (Wl, mm) ♂Wl = 71.2 + 6.3Age   ♀Wl = 76.9 + 7.1Age
Age 1247.8 1 <.001 1105.5 1 <.001
Hatch order 4.8 .09 0.67 .7
Brood size 3.13 .08 1.7 .2
(ii) Head to bill length (Hb, mm) ♂Hb = 50 + 1.3Age + 1 Brood size   ♀Hb = 52.02 + 1.5Age − 1.7 Brood size
Age 337.7 1 <.001 298.8 1 <.001
Hatch order 2.64 .3 0.6 .8
Brood size 5.63 1 .02 5.98 1 .02
(iii) Body mass (Mb, g) ♂Mb = 251 + 24Age + 21.4 Brood size + 14.8 (second hatched) − 21.2 (third hatched)   ♀Mb = 303.6 + 28.1Age − 19.5 (second hatched) − 94.82 (third hatched)
Age 439.3 1 <.001 203.5 1 <.001
Hatch order 8.36 2 .02 12.2 2 .002
Brood size 4.21 1 .04 2.75 .1
Nestlings older than 20 days
(iv) Wing length ♂Wl = 162.4 + 6.7Age
Age 1528.5 1 <.001 — — —
Hatch order 2.17 .3 — — —
Brood size 1.7 .2 — — —
(v) Head to bill length ♂Hb = 61.6 + 0.4Age − 1.3 (second hatched) − 1.9 (third hatched)
Age 41.9 1 <.001 — — —
Hatch order 8.8 2 .01 — — —
Brood size 0.4 .6 — — —
(vi) Body mass ♂Mb = 433.2 + 7.6Age
Age 39.1 1 <.001 — — —
Hatch order 4.02 .1 — — —
Brood size

1.3

1

.3

Change in deviance statistics for each term initially included in the model are presented, significant terms emboldened ( p < .05). 35 males (11 in 2000 and 24 in 2001) from 24 broods and 23 females (14 in 2000 and 9 in 2001) from 17 different broods were measured at 7–20 days old, with 29 males (20 in 2000 and 9 in 2001) from 21 broods measured at greater than 20 days old.

### Benefits associated with raising each sex of offspring: recruitment

Data on recruitment are based on a larger sample of broods, as it was not practical to attach cameras to all nests. A total of eight offspring, four birds of each sex, returned to the study site at approximately 23 months of age. All four males, three first-hatched and one third-hatched bird, obtained a territory within the study site immediately on returning. Three females, 1 first- and 2 second-hatched birds, were also successfully recruited on their return to the study site. The fourth bird, a third (last-) hatched female, remained in the floating population for a full breeding season before gaining a territory at just over 3 years of age. Both the last-hatched male and female were recaptured when fully grown at 2 years of age and remeasured. Both were still relatively small in comparison with a database of all birds measured (see McDonald et al., 2005 ), with the male ranked 46 th ( n = 69) and the female 79 th ( n = 90) largest in terms of head to bill length.

## DISCUSSION

Despite distinct RSD, male and female brown falcon offspring hatching either first or second were given equal numbers of mouthfuls of food during feeding events throughout the nestling period, without a reduction in the probability of female nestlings in these hatch positions fledging. Both sexes in these hatch orders were then recruited into the breeding population at the same age, indicating that benefits to parents of raising each sex were similar. However, evidence for a starvation penalty was found for third-hatched chicks, with birds in this hatch position receiving reduced amounts of food from mothers, with the reduction particularly marked for last-hatched female nestlings. The biological significance of our measure of food allocation was apparent, with the reduced growth rate and increased probability of mortality among all last-hatched individuals. For last-hatched females that experienced the lowest feeding levels of all sex and hatch orders, the reduced food resulted in the death of all individuals at nests monitored with cameras.

### Differential investment in raising male and female raptor offspring exhibiting RSD

Considering only first- and second-hatched offspring, no sex differences in egg volume or feeding rate (this study), incubation, nestling or postfledgling periods of parental care were observed ( McDonald, 2004 ), indicating that much of the potential differences in investment required to raise each sex, and thus presumably the costs involved to future parental reproductive effort, did not differ. Moreover, the sex ratio of a brood raised the previous year did not impose any costs on parental survival or breeding in subsequent seasons ( McDonald et al., 2004 ). It is possible that mothers manipulated a component of egg content, for example hormone content ( Schwabl et al., 1997 ), to achieve differential sex allocation. However, each sex hatched at a similar size, suggesting that any such difference did not translate into differences in growth rate ( McDonald, 2003b ). Similarly, while larger female offspring were not given more mouthfuls of food, it is conceivable that they may have been fed mouthfuls containing a different energy or nutritional content (e.g., Magrath et al., 2004 ). However, this is unlikely as videotape footage revealed no obvious evidence of different proportions of food, prey species, and/or tissues being fed preferentially to either sex ( McDonald, 2004 ). Virtually all meals comprised bird, mammal, or reptile meat ( McDonald et al., 2003 ).

Together, these data imply that females, despite being the larger sex in this species, did not impose increased costs of biological importance on their parents. A lack of sex differences in food requirements in other dimorphic raptors has been described previously, and may reflect differential growth, behavior, and/or metabolism ( Stamps, 1990 ; Torres and Drummond, 1999 ). Whatever the cause, it is now clear that sexual differences in food requirements in dimorphic birds are not as great as those predicted on the basis of body size differences alone ( Krijgsveld et al., 1998 ; Newgrain et al., 1993 ; Torres and Drummond, 1999 ), indicating that sex-specific costs should not necessarily be assumed a priori in dimorphic birds. As Fisher (1930) notes, it is the cost on future reproductive value or success of parents that is the key factor in determining sex allocation; biased sex ratios are predicted only if sex differences in food requirements are sufficient to detract from future reproductive success of parents. For the brown falcon, this was not apparently the case.

### Consequences of biased sex allocation

A persistent hatch-order effect on position during feeding events and subsequent growth and mortality rates as observed in this study is common in most asynchronously hatching altricial birds (reviewed by Krebs, 1999 ); however, this phenomenon cannot explain the biased sex allocation and resulting poor prospects of fledging among last-hatched nestlings. Moreover, as only female brown falcons fed offspring, differences in parental feeding strategies such as those described by Kilner (2002) are also irrelevant. It is possible that nests monitored by surveillance cameras with a last-hatched female were all under food stress initially and thus would have suffered brood reduction regardless of offspring sex. Experimental evidence is required to discount this hypothesis completely, however, it appears unlikely as the lay date of each nest from which a last-hatched female perished was earlier than average for the respective year. As is true for many other species, brown falcon nests laid earlier in the season have greater reproductive success ( McDonald et al., 2004 ) and, by extension, are therefore less likely to suffer food stress. More likely perhaps is the influence of sibling competition on feeding outcomes. Dimorphism can have mixed effects on nestlings, with larger chicks sometimes more prone to starvation in times of food stress ( Clutton-Brock et al., 1985 ; Griffiths, 1992 ; Røskaft and Slagsvold, 1985 ), alternatively in some species, particularly raptors, the larger sex is more able to monopolize prey items and outcompete the smaller during feeding events ( Anderson et al., 1993a ; Hipkiss et al., 2002 ). Male brown falcons did not procure more mouthfuls of food when hatched first or second nor did sex influence an individual's chance of being closest to the female or be fed the first mouthful during a feeding event for any hatch order. Given this, it seems unlikely that last-hatched male nestlings are somehow able to compete better with their older nest mates than females, last-hatched nestlings were outcompeted to food by virtue of their shorter size and thus reach ( McDonald, 2004 ), regardless of sex. Moreover, last-hatched females in this study had all perished before they were 19 days old. Yet it is not until the nestlings are around 15–17 days old that prey items are left in the nest, giving them the opportunity to compete directly ( McDonald, 2004 ), by which time most last-hatched females had already perished.

The reduced food supply to last-hatched chicks leads to slower growth than their earlier-hatched counterparts. Most critically, last-hatched falcons had smaller skeletal measurements (head to bill length) late in the nestling period, just prior to fledging. Growth is determinate in birds and, in skeletal terms at least, in brown falcons is all but completed at fledging ( McDonald, 2003b ). Disadvantages associated with poor nestling condition prevail in later life in other birds ( Thomas et al., 1999 ), including raptors ( Appleby et al., 1997 ; Arroyo, 2002 ), and two lines of evidence suggest that the small size of last-hatched brown falcon nestlings also persists into adulthood. Firstly, mortality of recently fledged raptors is characteristically high, with young birds generally possessing poor hunting skills and/or diets. In most species starvation is the main mortality factor in this period (see Newton, 1979 ; Olsen, 1995 for reviews), indicating that the likelihood of successfully obtaining extra resources for compensatory growth postfledging is low in many bird species ( Sedinger et al., 1995 ). Further, the last-hatched brown falcon offspring recaptured in this study at 2 years of age (fully grown) remained relatively small, potentially indicating that any compensatory growth was insufficient to recover hatch-order penalties postfledging. Moreover, even if compensatory growth was achieved in some individuals, recent research indicates that the negative effects of this can be varied and last throughout the individuals' lifetime inflicting severe fitness costs (see Metcalfe and Monaghan, 2001 for review).

Previous research has demonstrated that in the brown falcon, initial male recruitment and subsequent reproductive success is independent of body size. However, among female falcons, those with larger skeletal measures are more likely to prevail during intrasexual competition for territories and thus be recruited ( McDonald et al., 2005 ). Thus small last-hatched male offspring are still likely to successfully obtain a breeding territory, but females in this position have little prospect of successful recruitment. Recruitment for the sample of offspring assessed in this study followed these patterns, with males from first and last hatch orders obtaining territories in the season they returned to the study site, yet only first- and second-hatched females achieved this feat. The only last-hatched female that successfully returned to the study site remained in the floating population for more than 12 months before finally obtaining a territory in 2002, the worst breeding season recorded in the study and a year in which many territories were vacant, indicating relaxed competition.

This study demonstrates that last-hatched nestlings face a reduced food supply regardless of sex, however, the reduction was more severe for female offspring. It is possible that in extremely favorable seasons and/or with parents of high quality this last-hatched disadvantage could be overcome, although this was not the case during this study. This indicates last-hatched nestlings offer different benefits to parents in terms of their prospects of recruitment, with fitness returns from last-hatched sons likely to outweigh those of daughters in this hatch position. The allocation of a greater amount of resources to last-hatched males compared to females matches the broad predictions of the Trivers and Willard (1973) hypothesis because males in this position enjoy a greater chance of providing fitness returns to parents with continued investment. Thus for the brown falcon at least, sexual dimorphism influences the benefits of raising offspring in each sex, rather than the costs as would have been predicted a priori.

We thank Melbourne Water, Avalon Airport, Werribee CSR Readymix, Avalon Mountain View Quarry, and various private landowners in the area for allowing access to their land. Christine Donnelly and Ann Cowling from the Statistical Consulting Unit at the Australian National University (ANU) provided statistical advice. Sidney McDonald and Bob Phillips helped construct and maintain the nest surveillance cameras. The Australian Bird and Bat Banding Scheme provided the leg bands used in this study. P.M. was supported during the project by an ANU Graduate School Scholarship, with additional funds supplied by Stuart Leslie Bird Research Awards, a Cayley 2000 Memorial Scholarship, Birds Australia VicGroup Research Grants, and the Joyce W. Vickery Scientific Research Fund. Two anonymous reviewers greatly improved this manuscript. Study methods were approved by the ANU Animal Experimentation Ethics Committee (F.BTZ.02.99) and comply with current Australian laws.

## References

Anderson DJ, Budde C, Apanius V, Gomez JEM, Bird DM, Weathers WW,
1993
a. Prey size influences female competitive dominance in nestling American kestrels ( Falco sparverius ).
Ecology

74
:
367
–376.
Anderson DJ, Reeve J, Gomez JEM, Weathers WW, Hutson S, Cunningham HV, Bird DM,
1993
b. Sexual size dimorphism and food requirements of nestling birds.
Can J Zool

71
:
2541
–2545.
Aparicio JM, Cordero PJ,
2001
. The effects of the minimum threshold condition for breeding on offspring sex-ratio adjustment in the lesser kestrel.
Evolution

55
:
1188
–1197.
Appleby BM, Petty SJ, Blakey JK, Rainey P, MacDonald DW,
1997
. Does variation of sex ratio enhance reproductive success of offspring in tawny owls ( Strix aluco )?
Proc R Soc Lond B

264
:
1111
–1116.
Arroyo BE,
2002
. Fledgling sex ratio variation and future reproduction probability in Montagu's harrier, Circus pygargus .
Behav Ecol Sociobiol

52
:
109
–116.
Baker-Gabb DJ,
1984
. The feeding ecology and behavior of seven species of raptor overwintering in coastal Victoria.
Aust Wild Res

11
:
517
–532.
Bednarz JC, Hayden TJ,
1991
. Skewed brood sex ratio and sex-biased hatching sequence in Harris's hawks.
Am Nat

137
:
116
–132.
Bortolotti GR,
1986
. Influence of sibling competition on nestling sex ratios of sexually dimorphic birds.
Am Nat

127
:
495
–507.
Boulet M, Olsen PD, Cockburn A, Newgrain K,
2001
. Parental investment in male and female offspring by the peregrine falcon Falco peregrinus .
Emu

101
:
95
–103.
Brommer JE, Karell P, Pihlaja T, Painter JN, Primmer CR, Pietiäinen H,
2003
. Ural owl sex allocation and parental investment under poor food conditions.
Oecologia

137
:
140
–147.
Byholm P, Ranta E, Kaitala V, Linden H, Saurola P, Wikman M,
2002
. Resource availability and goshawk offspring sex ratio variation: a large-scale ecological phenomenon.
J Anim Ecol

71
:
994
–1001.
Charnov EC,
1982
. The theory of sex allocation. Princeton: Princeton University Press.
Clutton-Brock TH,
1986
. Sex ratio variation in birds.
Ibis

128
:
317
–329.
Clutton-Brock TH, Albon SD, Guinness FE,
1985
. Parental investment and sex differences in juvenile mortality in birds and mammals.
Nature

313
:
131
–133.
Clutton-Brock TH, Iason GR,
1986
. Sex ratio variation in mammals.
Q Rev Biol

61
:
339
–374.
Cockburn A, Legge S, Double MC,
2002
. Sex ratios in birds and mammals: can the hypotheses be disentangled? In: Sex ratios: concepts and research methods (Hardy ICW, ed). Cambridge: Cambridge University Press; 266–286.
Collopy MW,
1986
. Food consumption and growth energetics of nestling golden eagles.
Wilson Bull

98
:
445
–458.
Dijkstra C, Daan S, Buker JB,
1990
. Adaptive seasonal variation in the sex ratio of kestrel broods.
Funct Ecol

4
:
143
–147.
Ewen JG, Cassey P, Moller AP,
2004
. Facultative primary a lack of evidence sex ratio variation: a lack of evidence in birds?
Proc R Soc Lond B

271
:
1277
–1282.
Fisher RA,
1930
. The genetical theory of natural selection. Oxford: Oxford Clarendon.
Frank SA,
1990
. Sex allocation theory for birds and mammals.
Annu Rev Ecol Syst

21
:
13
–55.
Genstat Committee,
1993
. Genstat 5 release 3 reference manual. Oxford: Clarendon Press.
Gowaty PA,
1993
. Differential dispersal, local resource competition, and sex ratio variation in birds.
Am Nat

141
:
263
–280.
Griffiths R,
1992
. Sex-biased mortality in the lesser black-backed gull Larus fuscus during the nestling stage.
Ibis

134
:
237
–244.
Griffiths R, Double MC, Orr K, Dawson RJG,
1998
. A DNA test to sex most birds.
Mol Ecol

7
:
1071
–1075.
Heinsohn R, Legge S, Barry S,
1997
. Extreme bias in sex allocation in Eclectus parrots.
Proc R Soc Lond B

264
:
1325
–1329.
Hipkiss T, Hörnfeldt B, Eklund U, Berlin S,
2002
. Year-dependent sex-biased mortality in supplementary-fed Tengmalm's owl nestlings.
J Anim Ecol

71
:
693
–699.
Hoyt DF,
1979
. Practical methods of estimating volume and fresh weight of bird eggs.
Auk

96
:
73
–77.
Kilner RM,
2002
. Sex differences in canary ( Serinus canaria ) provisioning rules.
Behav Ecol Sociobiol

52
:
400
–407.
Komdeur J, Daan S, Tinbergen JM, Mateman C,
1997
. Extreme adaptive modification in sex ratio of the Seychelles warbler's eggs.
Nature

385
:
522
–525.
Krackow S,
1993
. Note on falconiforme sex ratios given in Olsen and Cockburn 1991: avian raptors exhibit no unique sex-ratio bias.
Behav Ecol Sociobiol

32
:
429
–430.
Krebs EA,
1999
. Last but not least: nestling growth and survival in asynchronously hatching crimson rosellas.
J Anim Ecol

68
:
266
–281.
Krijgsveld KL, Dijkstra C, Visser GH, Daan S,
1998
. Energy requirements for growth in relation to sexual size dimorphism in marsh harrier Circus aeruginosus nestlings.
Physiol Zool

71
:
693
–702.
Laaksonen T, Fargallo JA, Korpimäki E, Lyytinen S, Valkama J, Poyri V,
2004
. Year- and sex-dependent effects of experimental brood sex ratio manipulation on fledging condition of Eurasian kestrels.
J Anim Ecol

73
:
342
–352.
Leroux A, Bretagnolle V,
1996
. Sex ratio variations in broods of Montagu's harriers Circus pygargus .
J Avian Biol

27
:
63
–69.
Magrath MJL, van Lieshout E, Visser GH, Komdeur J,
2004
. Nutritional bias as a new mode of adjusting sex allocation.
Proc R Soc Lond B

271
:
S347
–S349.
McDonald PG,
2003
a. Variable plumage and bare part colouration in the brown falcon Falco berigora : the influence of age and sex.
Emu

103
:
21
–28.
McDonald PG,
2003
b. Nestling growth and development in the brown falcon Falco berigora : a new method of sex determination and an improved ageing formula.
Wild Res

30
:
411
–418.
McDonald PG,
2004
. The breeding ecology and behaviour of a colour-marked population of brown falcons Falco berigora .
Emu

104
:
1
–6.
McDonald PG, Olsen PD, Baker-Gabb DJ,
2003
. Territory fidelity, reproductive success and prey choice in the brown falcon Falco berigora : a flexible bet-hedger?
Aust J Zool

51
:
399
–414.
McDonald PG, Olsen PD, Cockburn A,
2004
. Weather dictates reproductive success and survival in the Australian brown falcon Falco berigora .
J Anim Ecol

73
:
683
–692.
McDonald PG, Olsen PD, Cockburn A,
2005
. Selection on body size in a raptor with pronounced reversed sexual size dimorphism: is bigger better?
Behav Ecol

16
:
48
–56.
Metcalfe NB, Monaghan P,
2001
. Compensation for a bad start: grow now, pay later?
TREE

16
:
254
–260.
Negro JJ, Hiraldo F,
1992
. Sex ratios in broods of the lesser kestrel Falco naumanni .
Ibis

134
:
190
–191.
Newgrain K, Olsen PD, Green B, Brothers N, Bartos B, Mooney N,
1993
. Food consumption rates of free-living raptor nestlings. In: Australian raptor studies (Olsen P, ed). Melbourne: R.A.O.U.; 74–84.
Newton I,
1979
. Population ecology of raptors. Berkhamsted: Poyser.
Olsen PD,
1995
. Australian birds of prey. Sydney: University of New South Wales Press.
Olsen PD, Cockburn A,
1991
. Female-biased sex allocation in peregrine falcons and other raptors.
Behav Ecol Sociobiol

28
:
417
–423.
Riedstra B, Dijkstra C, Daan S,
1998
. Daily energy expenditure of male and female marsh harrier nestlings.
Auk

115
:
635
–641.
Rosenfield RN, Bielefeldt J, Vos SM,
1996
. Skewed sex ratios in Cooper's hawk offspring.
Auk

113
:
957
–960.
Røskaft E, Slagsvold T,
1985
. Differential mortality of male and female offspring in experimentally manipulated broods of the rook.
J Anim Ecol

54
:
261
–266.
Schwabl H, Mock DW, Gieg JA,
1997
. A hormonal mechanism for parental favouritism.
Nature

386
:
231
.
Sedinger JS, Flint PL, Lindberg MS,
1995
. Environmental influence on life-history traits: growth, survival, and fecundity in black brant ( Branta bernicla ).
Ecology

76
:
2404
–2414.
Sheldon BC,
1998
. Recent studies of avian sex ratios.
Heredity

80
:
397
–402.
Stamps JA,
1990
. When should parents differentially provision sons and daughters?
Am Nat

135
:
671
–685.
Tella JL, Donazar JA, Negro JJ, Hiraldo F,
1996
. Seasonal and interannual variations in the sex-ratio of lesser kestrel Falco naumanni broods.
Ibis

138
:
342
–345.
Thomas F, Kayser Y, Hafner H,
1999
. Nestling size rank in the little egret ( Egretta garzetta ) influences subsequent breeding success of offspring.
Behav Ecol Sociobiol

45
:
466
–470.
Torres R, Drummond H,
1999
. Does large size make daughters of blue-footed booby more expensive than sons?
J Anim Ecol

68
:
1133
–1141.
Trivers RL, Willard DE,
1973
. Natural selection of parental ability to vary the sex ratio of offspring.
Science

179
:
90
–92.
Wiebe KL, Bortolotti GR,
1992
. Facultative sex ratio manipulation in American kestrels.
Behav Ecol Sociobiol

30
:
379
–386.

## Author notes

aSchool of Botany and Zoology, Australian National University, Canberra A.C.T. 0200, and bSchool of Biological Sciences, University of Wales, Bangor, Gwynedd LL57 2UW, UK