Manx Shearwater Chicks; Seasonal, Parental, and Genetic Influences On
Total Page:16
File Type:pdf, Size:1020Kb
The Condor 88:324-327 0 The Cooper Ornithological Society 1986 MANX SHEARWATER CHICKS: SEASONAL, PARENTAL, AND GENETIC INFLUENCES ON THE CHICK’S AGE AND WEIGHT AT FLEDGING’ M. DE L. BROOKED ZoologyDepartment, Edward Grey Institute, South Parks Road, Oxford OX1 3PS, United Kingdom Abstract. Manx Shearwater(Pu@zuspujinus) pairs on Skokholm Island, Wales, typically breed together over several seasons,rearing one chick per year. There is a seasonal decline of chick fledgingweight. Even after correction for this decline, pairs tend to rear chicksthat are consistently heavier or lighter than predicted on the basis of the chick’s hatching date. These deviations may reflect pair quality. Weight and hatching date of a chick are not correlated with age of fledging. However, fledging agesof full siblingsshow significantlyless variation than thoseof nonsiblings.The characterappears highly heritable, with a heritability value around 0.84. This is the first study to addressthe extent to which the fledging age of wild birds is genetically determined. Key words: Manx Shearwater;Puffinus puffinus;fledging age; heritability. INTRODUCTION Jones 1974) and laying date (Van Noordwijk There is apparent variation in the extent to et al. 1980). On the whole, “characters with which the age of fledging in birds mirrors en- the lowest heritabilities are those most closely vironmental circumstances.In certain species, connected with reproductive fitness,while the such as Audubon’s Shearwater (Pujinus Iher- characters with the highest heritabilities are minieri)(Harris 1969)and the Common Swift those that might be . the least important as (&us apus) (Lack 1956) age of fledging is determinants of natural fitness” (Falconer highly variable. When feeding conditions are 1960). favorable, chicks develop and fledge more rap- There are few studiesof heritability of growth idly than when food is scarce.In other species, rate and fledging age in wild birds. In one ex- for example, Sooty Shearwaters (P. griseus), ception, Ricklefs and Peters(198 l), usingcross- fledging occurs at a similar age regardlessof fostering experiments, found a low heritability the fledgling’s weight (Richdale 1963). Weight of growth rate in European Starlings (Sturnus presumably reflects the food supply provided vulgaris).No studiesaddress genetic influences for the chick by the parents. on fledging age, perhaps because in multi- Perrins et al. (1973) have shown that, in the young, nidicolous broods fledging is a process Manx Shearwater (P. pu$kzus)which naturally involving complex parent-chick interactions rears a single chick, experimentally twinned (Davies 1976). This problem arisesless mark- chicks fledge at a lighter weight than controls. edly in burrow-nestingProcellariiformes, since Nevertheless, they fledge at the same age as the chick is deserted by the parents sometime controls and with a similar wing length. This before fledging. (The Manx Shearwater deser- resistanceof fledging age to environmental in- tion period averages 8.5 days [Harris 19661.) fluences suggeststhat it is a character at least Fledging is then a singular event which occurs partly under geneticcontrol. The present study in the absence of contemporaneous parental explores the suggestionfurther. influences. Studies of the heritability of avian mor- phology have mostly reported heritability val- METHODS ues ranging from 40 to 90% (see Boag 1983). From 1973 through 1976 on Skokholm Island, These high values probably representheritable Wales, I maintained a group of study burrows morphological variation, rather than the shar- in which all breeding Manx Shearwaterswere ing of a common environment by parents and banded. Manx Shearwater pairs are normally offspring(Smith and Dhondt 1980, but seealso faithful in successiveyears to each other and James 1983). Lower heritability values have return to the same breeding burrow. As a re- generally been reported for other phenotypic sult, faithful pairs raise in different years a se- characters, such as clutch size (Perrins and ries of chicks that are full siblings. The max- imum possiblenumber of chicks in each such “family” in this study is four, the number of I Received 2 1 August 1985. Final acceptance26 March 1986. Edited by Charles Blem. years the study continued. Change of mate and 2Present address: Zoology Department, Downing Street, burrow are both more likely after breeding fail- Cambridge CB2 3EJ, United Kingdom. ure (Brooke 1978). 13241 FLEDGING MANX SHEARWATERS 325 TABLE 1. Mean weight and age of Manx Shearwater TABLE 3a. Analysis of variance of fledging weight of chicks when fledging on Skokholm, 1973-1976. chicks reared by 56 unchanged Manx Shearwater pairs that reared chicksin two or more of the years, 1973-1976. Yea1 Mean weight (9) + SE (n) Mean age (days) -t SE (n) 1973 438.2 * 5.4 (57) 71.3 ? 0.4 (53) SOUPX df 1974 455.2 -t 5.7 (57) 71.1 f 0.4 (56) 1975 427.1 * 5.1 (41) 71.4 + 0.4 (39) Between pairs 55 119,434 2,172 2.00 1976 444.7 f 5.5 (54) 70.9 + 0.4 (53) Within pairs 92 99,845 1,085 (P < 0.005) Total 147 219,279 Note: There are significant differences between years in fledging weight (F,,,,,= 4.27, P < O.Ol), but not in fledging age (F,,,,, = 0.22, IX). Burrows were inspecteddaily in the hatching tend to hatch on the same date. Consequently, period to determine the hatching date. Chicks it would be expected that unchanged pairs were examined daily from 35 days of age until would raise chicks of similar fledging weight fledging at about 70 days. During the latter in different years, and this was observed (Table period chicks were weighed daily, at the same 3a). time each day. Peak weight, which occurs be- Knowing the regression of fledging weight tween Days 39 and 61 (Harris 1966), was on hatching date for a particular year (Table therefore known, as were fledging weight, age 2) it is possible to predict the fledging weight when last fed, and age at fledging. As many as of a chick hatching on a particular date and 55 chicks were reared in the study burrows then determine the magnitude and sign of the eachyear. The sexof the chickswas not known. deviation from the prediction shown by any one chick. If unchangedManx Shearwaterpairs RESULTS show consistent differences in their ability to Mean weight of chicksfledging from study bur- feed chicks, then the within-pair variance in rows varied significantly between years (Table this residual value should be lessthan the be- l), but age of fledging did not vary. Since the tween-pair variance. This was observed (Table chicks are reared by a random selection of 3b). Thus, after correcting for date, individual breeding adults, these results imply that there pairs tend to fledge chicks at a weight which are signifcant differencesbetween years in en- is consistentlygreater or lessthan expected on vironmental (feeding?) conditions. the basis of the chick’s hatching date. In each year except 1976, there was a sig- There are insufficient data available to de- nificant negative correlation (P < 0.01) be- termine whether this result is in any way due tween fledgingweight and hatching date (Table to the fact that successfulpairs typically breed 2). The decline in weight through the season in the sameburrow in successiveyears (Brooke could either be due to a deterioration in feeding 1978). The result is not simply a reflection of conditions later in the seasonor to a tendency body size, for example, large parents raising for pairs lessproficient at feeding chicks to lay large heavy chicks. Although body size in and hatch their eggslater. The fact that the rate shearwaters appears to be strongly heritable of decline of fledging weight differed between (Brooke 1977), there is no significant corre- years (Table 2) suggeststhat environment is at lation (r = 0.134, II = 197, n.s.) between fledg- least partly responsiblefor that decline. ing weight and body size, as measured by 60- It is already known that female Manx Shear- day tarsus length. waters, which remain paired to the same male Multiple regression analysis of the age of in successiveseasons, tend to lay on the same fledgingof chicks used four independent vari- date, both absolutely and relative to the pop- ables:the chick’s age when last fed, the chick’s ulation (Brooke 1978). Their chicks therefore age when at peak weight, the fledging weight, TABLE 2. Correlation coefficientsand slopes(g/day) of TABLE 3b. Analysis of variance for residuals, from linear regressionsof fledging weight on hatching date for the regressionof fledging weight on hatching date, of the Manx Shearwaterchicks fledging on Skokholm, 1973-l 976. fledging weights of chicks reared by 56 unchangedManx Shearwaterpairs that reared chicks in two or more of the Sample years, 1973-1976. YCZX I Slope SE size 1973 -0.415 -2.44 0.749 53 Sum of MeaIl 1974 -0.393 -1.57 0.501 56 SOUP3 df squares square F 1975 -0.469 -1.72 0.533 39 Between pairs 79,755 1,450 1.54 1976 -0.200 -0.96 0.659 53 Within pairs 86,420 939 (P < 0.05) Note: There are no significant dilkences between years in the regression Total 147 166,175 coefficient. 326 M. DE L. BROOKE TABLE 4. Multiple regressionof the age in days (Y) at TABLE 5. Analysis of variance of fledgingage of chicks which Manx Shearwaterchicks fledge on four independent reared by 106 unchangedManx Shearwater pairs in one variables: X, = age (days) when last fed, X, = age (days) or more of the years, 1973-1976. when at peak weight; X, = fledgingweight(g); X, = hatch- ing date (10 June = 0). The regression equation Between pairs 105 1,134.7 10.81 2.63 Y= 58.5 + 0.174X, + 0.102X, 0.00916X, + 0.0254X, Within pairs 89 365.4 4.11 (P < 0.005) Standard deviations t-rat,0 = Total 194 1,500.l of coefficients coetlx3D P 4 0.0498 3.49 CO.001 X, 0.0394 2.60 CO.01 XX 0.00502 -1.82 n.s.