sign in sign up
<<
Home , Brown snake eagle, Circaetus, Crozet shag, Greater spotted eagle, Hornbill, Tawny eagle

CONDOR Tuesday May 25 2004 01:37 PM cond 106_309 Mp_540 Allen Press • DTPro System GALLEY File # 09TQ

The Condor 106:540±548 ᭧ The Cooper Ornithological Society 2004

ONTOGENY OF LARGE : MIGRANTS DO IT FASTER

SHAI MEIRI1 AND YORAM YOM-TOV Department of Zoology, Tel Aviv University, Tel Aviv 69978

Abstract. We compared incubation and ¯edging times between large (Ͼ2 kg) migrating and sedentary birds. We found that while length of incubation period does not differ between migrants and nonmigrants, ¯edging period is signi®cantly shorter in the former. This pattern is apparent in the class as a whole, as well as within orders, families, and genera. Additional, albeit weak, evidence suggests that clutch sizes of migrants are smaller than those of closely related resident birds. We hypothesize that the need to migrate constrains the length of developmental period of large migrating birds, especially in that undertake long- distance migration. Key words: body size, constraints, development, ¯edging, incubation, migration.

Ontogenia de Aves Grandes: Los Migrantes lo Hacen maÂs RaÂpido Resumen. Comparamos los tiempos de incubacioÂn y emplumamiento entre aves grandes (Ͼ2 kg) migratorias y sedentarias. Encontramos que mientras la longitud del perõÂodo de incubacioÂn no di®ere entre migrantes y no migrantes, el perõÂodo de emplumamiento es signi®cativamente maÂs corto en las migrantes. Este patroÂn resulta evidente a nivel de toda la clase, asõÂ como al interior de oÂrdenes, familias y geÂneros. Existe evidencia adicional, aunque deÂbil, que sugiere que las nidadas de las migrantes son maÂs pequenÄas que las de aves residentes estrechamente relacionadas. Proponemos la hipoÂtesis de que la necesidad de migrar limita la longitud del perõÂodo de desarrollo de las aves migrantes grandes, especial- mente en las especies que realizan migraciones a traveÂs de grandes distancias.

INTRODUCTION longest of any . Its far larger congener, the A bird's breeding cycle consists of the periods 34-kg Emperor Penguin (A. forsteri; Dunning of courtship, nest building, egg laying, incuba- 1993), feeds in Antarctic waters but breeds in- tion, and ¯edging. Many biological characters, land. This species starts its breeding cycle in the including the length of incubation and ¯edging Antarctic winter, and its chicks attain adult periods, scale positively with body mass (Rahn 214 days after incubation begins (del et al. 1975, Calder 1984, Schmidt-Nielsen Hoyo et al. 1992, Reilly 1994). Could it be that 1984). It is therefore obvious that the larger a the migratory behavior of the Emperor Penguin bird is, the longer it needs to develop from a forces it to shorten its development time, relative zygote to a ¯ying juvenile. Juveniles of some to its smaller, sedentary congener? large birds do indeed require a long period be- Whereas chicks of migratory birds need to fore taking ®rst ¯ight: the combined incubation complete their growth and gain ¯ight capability and ¯edging periods for the Andean Condor by fall, long developmental periods pose little (Vultur gryphus) and the Californian Condor problem for sedentary birds. Large migratory (Gymnogyps californianus) are 239 and 238 birds with a long breeding cycle would thus be days, respectively; and developmental periods in at a disadvantage in high latitudes, where the excess of 300 days have been recorded for the breeding season is short. Members of large mi- larger species of albatrosses (del Hoyo et al. gratory species might therefore have accelerated 1992). The ¯ightless King Penguin (Aptenodytes their development in comparison with similar- patagonicus; 13.2 kg, Dunning 1993), which sized residents. A short breeding season at high breeds close to its feeding grounds, has a de- latitude, combined with high wing load and a velopment period of just over a year (385 days; consequent reduction in speed of ¯apping ¯ight del Hoyo et al. 1992, Reilly 1994), perhaps the might explain why large birds do not migrate very long distances (HedenstroÈm and Alerstam 1998). Manuscript received 21 October 2003; accepted 5 April 2004. Several factors affect the length of avian de- 1 E-mail: [email protected] velopmental periods. For example,

[540] CONDOR Tuesday May 25 2004 01:37 PM cond 106_309 Mp_541 Allen Press • DTPro System GALLEY File # 09TQ

ONTOGENY OF LARGE BIRDS 541

(Picidae) and other cavity nesters usually have a larger threshold for ``large'' birds would have shorter incubation and longer ¯edging times prevented us from controlling for phylogeny in compared to similar-sized noncavity nesters comparing migrant and sedentary species. It (Yom-Tov and Ar 1993). Similarly, shearwaters should be noted, however, that we do not mean (Procellariidae) have a very long ¯edging peri- to imply that 2 kg is the precise threshold for od, whereas game birds () ¯edge differences in developmental periods between early in relation to other birds (Carrier and Au- migrant and sedentary species. riemma 1992). Body-mass data (of the larger sex where Long-distance migrants have small clutches available) were obtained from Dunning (1993) and raise fewer broods per year than do resi- and del Hoyo et al. (1992). We followed the tax- dents, while short-distance migrants have the onomy of del Hoyo et al. (1992). We calculated most broods per year and the smallest clutches development time as the sum of incubation and among these three groups (BoÈhning-Gaese et al. ¯edging periods (in days). Some large birds 2000). Large birds usually lay only one clutch have a long post¯edging period, leading to an per year, and are therefore unable to shorten underestimate of actual developmental periods. their breeding cycle by having fewer clutches. Our data show that such post¯edging periods They can, however, shorten it by shortening the (which can be up to a year long, especially in periods of courtship behavior, nest building, egg large raptors such as the Martial or the laying, incubation and ¯edging. Since the latter Great , del Hoyo et al. 1992) are two periods are generally the longest in the almost invariably restricted to sedentary species. breeding cycle, it is reasonable to assume that Because precise data on post¯edging care are plasticity in length of the breeding cycle would often lacking we have taken a conservative ap- be re¯ected mostly through changes in these pa- proach, and added such data to the calculated rameters. Therefore we designed this study to developmental times only for species of Pele- examine whether large migratory birds respond canus, the only in our data in which post- to the selective pressure to complete their ¯edging care exists in migrants. growth quickly by accelerating their nesting cy- Another factor in¯uencing the length of the cle relative to similar-sized nonmigratory spe- nesting period is that of time taken to lay the cies. entire clutch. In to examine whether laying time differs between migrant and sedentary spe- METHODS cies (as was found for clutch size, BoÈhning-Gaese We obtained the data for mean clutch size, et al. 2000), clutch size must be multiplied by length of incubation and ¯edging periods, and the interval between the laying of successive migratory habits of large (over 2 kg) ¯ying birds eggs. However, to the best of our knowledge from del Hoyo et al. (1992). Data for the Greater data on the latter variable are nearly nonexistent Spotted Eagle are from Graszynski et al. (2002; for the species in our sample. We therefore used scienti®c names appear in Table 1). Members of clutch size only as a surrogate for the length of the Galliformes and Otididae, whose chicks are the laying period. Because clutch size is highly capable of ¯ying well before they attain adult dependent on phylogenetic relationships, we an- size (Carrier and Auriemma 1992, Collar 1996), alyzed this effect by using only families con- were not considered. Species in which some taining both migrant and sedentary species. We populations migrate whereas others do not (such used a two-way ANOVA, with clutch size as the as and Canada ) were not dependent variable, and both and migra- used, as the developmental periods listed for tion as the categorical predictors. these species do not note whether they refer to Species may not constitute independent data migratory or nonmigratory populations. points, owing to shared phylogeny (Smith The choice of 2 kg for classifying a bird as 1994); thus comparing all large birds regardless large was somewhat arbitrary, considering that of phylogeny runs the risk of in¯ating the num- the modal size in the class Aves is much lower ber of actual degrees of freedom. This is not (33 g, Maurer 1998). However, we expected that merely a statistical problem: different bird taxa birds with body mass smaller than 2 kg would occupy different positions along the altricial± have developmental periods far shorter than the precocial continuum, which greatly in¯uences length of the respective breeding season. Using developmental rates and times (Ricklefs and CONDOR Tuesday May 25 2004 01:37 PM cond 106_309 Mp_542 Allen Press • DTPro System GALLEY File # 09TQ

542 SHAI MEIRI AND YORAM YOM-TOV low del yes no no no no no no yes yes yes yes yes yes yes yes yes no no yes yes no no yes no no no no no 53 77 77 73 84 68 42 56 45 55 40 68 87 70 70 55 61 60 66 67 85 63 55 60 160 102 135 105 Fledging period (days) Migratory? 29 24 29 36 30 38 29 25 27 24 25 28 35 31 35 36 32 29 34 34 31 31 35 33 30 30 30 30 Incubation period (days) 2 kg) birds with known development times (del Ͼ 5 8.5 4 5 7.5 6.5 6 4 4.5 6 8.5 5.5 4 4.5 5 5 4 4 3 3 3.5 2.5 2.5 3 2.5 3.5 size 10 13 Clutch 2600 2050 2000 3550 2900 6200 7100 9350 3550 2600 44506700 3 2 27 30 62 100* no no 2100 2700 2750 3450 2750 3200 5000 3450 2200 4200 3000 6400 7250 2700 2400 3200 11 400 11 800 2 1 4 3 3 1 2 4 Cygnus buccinator Cygnus columbianus Anser indicus Anseranas semipalmata Branta sandvicensis Ardea goliath Balaeniceps rex Ciconia boyciana Ciconia ciconia Anser brachyrhynchus Anser caerulescens Anser canagicus Anser fabalis Cereopsis novaehollandiae Chloephaga melanoptera Cygnus atratus Cygnus Cygnus Cygnus olor Plectropterus gambensis Sarkidiornis melanotos Ciconia episcopus Ciconia maguari Ciconia nigra Ephippiorhynchus senegalensis Leptoptilos crumeniferus Mycteria americana Mycteria Mycteria leucocephala Alopochen aegyptiacus Anser albifrons Common name Scienti®c name Mass (g) Trumpeter Swan Tundra Swan Whooper Swan Mute Swan Spur-winged Goose Comb Bar-headed Goose Magpie Goose Hawaiian Goose Cape Barren Goose Andean Goose Black Swan Goliath Shoebill Oriental White European White Stork Wooly-necked Stork Maguari Stork Black Stork Saddlebill Marabou Wood Stork Yellow-billed Stork Painted Stork Egyptian Goose Greater White-fronted Goose Pink-footed Goose Snow Goose Emperor Goose Bean Goose Ardeidae Balaenicipididae Ciconiidae Hoyo et al. 1992).Hoyo Species et with al. precocial (1992). ¯ight Superscripts and denote species migrant-resident with pairs both of migratory closely and related sedentary species, populations used are for excluded. pairwise comparisons and to common names account fol for phylogeny. TABLE 1. Body mass, clutch size, length of incubation and ¯edging periods, and migratory habits of all large ( Anatidae CONDOR Tuesday May 25 2004 01:37 PM cond 106_309 Mp_543 Allen Press • DTPro System GALLEY File # 09TQ

ONTOGENY OF LARGE BIRDS 543 no no yes yes no no no no no no no yes no no no no no no no no no no no no no no yes no no no no no no 75 63 69 90* 68 60 81 94 70 70* 65 70 84 79 85 86 90 87* 97.5* 105 118* 125 140 150 120 160* 130* 118 100 109.5* 164.5* 102.5* 143.5* Fledging period (days) Migratory? 44 45 43 43 49 56 56 45 55 55 40 45 42 45 42 44 41 56 39 29 51 34 39 42 39 40 40 61 49 50 56 55 56 Incubation period (days) 2.5 2 2 2.5 1 1.5 1 1 1 1 2 2 2 2 2 2 2 2 1.5 1.5 1 2.5 2 1.5 2 2 2 1 1 2 1 1 1 size Clutch 3000 3500 2700 3400 2050 5700 5300 4750 8200 7400 2650 3150 3400 4600 7750 3400 2500 7600 2500 2500 2050 2450 3000 2700 4000 4800 2150 4050 4250 3650 2450 7500 5900 8 8 7 5 5 6 6 7 nipalensis Haliaeetus pelagicus cinereus Gypaetus barbatus africanus Gyps bengalensis Gyps coprotheres Gyps rueppellii Haliaeetus leucogaster Aquila audax Aquila Aquila heliaca Aquila rapax Aquila verreauxii Haliaeetus vocifer Haliaeetus vociferoides Harpia harpyja fasciatus Ichthyophaga ichthyaetus Necrosyrtes monachus Aceros cassidix bicornis Buceros rhinoceros Bucorvus abyssinicus Bucorvus leadbeateri undulatus Aquila adalberti Pithecophaga jefferyi Polemaetus bellicosus Stephanoaetus coronatus Terathopius ecaudatus Torgos tracheliotus Trigonoceps occipitalis Common name Scienti®c name Mass (g) Eagle Verreaux's Eagle Steller's African Eagle Madagascar Fish Eagle Harpy Eagle Bonelli's Eagle Grey-headed Fishing Eagle Hooded Brown Eagle African White-backed Vulture Indian White-backed Vulture Ruppell's Griffon White-bellied Sea Eagle Knobbed Northern Spanish Imperial Eagle Wedge-tailed Eagle Great Philippine Eagle Crowned Hawk Eagle Lappet-faced Vulture White-headed Vulture TABLE 1. Continued Bucerotidae CONDOR Tuesday May 25 2004 01:37 PM cond 106_309 Mp_544 Allen Press • DTPro System GALLEY File # 09TQ

544 SHAI MEIRI AND YORAM YOM-TOV no yes no no no yes no yes yes yes yes no yes no no no yes yes no yes no no yes yes no no no no 85 60 80 78 85 93 68 73 75 90 73 90 80* 75 70 53* 65 90 90* 63 74 110 100 107* 120 105* 180* 180* Fledging period (days) Migratory? 32 28 30 30 35 30 33 30 29 29 32 30 30 34 32 29 29 33 30 30 31 30 42 58 56 59 29 25 Incubation period (days) 2 2 2.5 2.5 1.5 2 2 2 2 2 2 2 2 2 2 2 3 2 3.5 2 3 2.5 2 1 1 1 2 2 size Clutch 5650 2300 3600 3750 8150 5850 8850 5500 6400 3950 6000 6400 4650 5000 9000 7000 3700 5000 5200 2000 2900 2150 3400 38003350 4150 2 44 86 no 11 000 10 100 12 500 12 12 9 13 14 14 9 10 13 11 10 11 Grus grus Grus leucogeranus Pelecanus conspicillatus aristotelis Phalacrocorax georgianus Anthropoides virgo Balerica pavonina Balerica regulorum Bugeranus carunculatus Grus americana Grus antigone Grus monacha Grus nigricollis Grus rubicunda Grus vipio Pelecanus crispus Pelecanus erythrorhynchos Pelecanus occidentalis Pelecanus onocrotelus Pelecanus philippensis Pelecanus rufescens Coragyps atratus Gymnogyps californicus Sarcoramphus papa Vultur gryphus Anthropoides paradisea Sagittarius serpentarius Gavia arctica Gavia immer Common name Scienti®c name Mass (g) Eurasian Crane Siberian Crane Hooded Crane Black Necked Crane Brolga White-naped Crane Australian Pelican Dalmantian Pelican American White Pelican Brown Pelican Great White Pelican Spot-billed Pelican Pink-backed Pelican American Black Vulture California Condor King Vulture Andean condor Blue Crane Demoiselle Crane Black-crowned Crane Grey-crowned Crane Wattled Crane Whooping Crane Sarus Crane Secretary bird Black-thorated Diver Great Northern Diver Pelecanidae Phalacrocoracidae TABLE 1. Continued Cathartidae Gruidae Sagittariidae Gaviidae CONDOR Tuesday May 25 2004 01:37 PM cond 106_309 Mp_545 Allen Press • DTPro System GALLEY File # 09TQ

ONTOGENY OF LARGE BIRDS 545 no no yes yes no no no no no yes yes no no yes no no no no no no no no no no no no 54 90 97 75 78 61 49 63 50 69 56.5* 120* 102* 102 235 145 120 141 240 278 165 167 120 140 160 156 Fledging period (days) Migratory? 44 44 44 41 44 29 29 79 72 72 72 79 78 65 60 71 65 70 69 34 35 36 32 33 30 29 Incubation only, to make the migrant-resident comparisons period (days) Plecanus 1 1 1 1 1 1 1 1 2 2 1 1 1 1 1 1 1 2 3 1.5 7.5 1.5 1 1 2.5 3 size Clutch 6250 2750 4100 3750 8200 8400 3050 2700 2100 2000 2350 3250 3750 3900 3150 2500 2800 40002300 3000 12600 2300 2200 60 113 no 2300 3550 5200 1 61 118 no 2200 2200 15 15 16 16 Sula bassama Sula capensis Diomedea amsterdamensis Diomedea bulleri Diomedea cauta Diomedea chrysostoma Diomedea epomophora Diomedea exulans Diomedea immutabilis Diomedea irrorata Diomedea melanophris Diomedea nigripes Diomedea fusca Phoebetria palpebrata Macronectes halli Bubo ascalaphus Bubo bubo Bubo lacetus Nyctea scandica Scotopelia peli Phoenicopterus chilensis Phoenicopterus ruber Macronectes giganetus Sula dactylard Sula nebouxii Sula serrator Phalacrocorax melanogenis Phalacrocorax varius Common name Scienti®c name Mass (g) Northern Masked Blue-footed Booby Amsterdam Albatross Buller's Albatross Shy Albatross Grey-headed Albatross Royal Albatross Wandering Albatross Pharaoh Eagle- Eurasian Eagle-Owl Verreaux's Eagle-Owl Snowy Owl Pel's Fishing-Owl Laysan Albatross Waved Albatross Black-browed Albatross Black-footed Albatross Sooty Albatross Light-mantled Albatross Northern Giant Petrel Chilean Greater Flamingo Southern Giant Petrel Crozet Shag Pied * Species with a signi®cant period of post¯edging care. This period was added to the ¯edging period for Diomedeidae Strigidae Phoenicopteridae Procellariidae TABLE 1. Continued conservative. CONDOR Tuesday May 25 2004 01:37 PM cond 106_309 Mp_546 Allen Press • DTPro System GALLEY File # 09TQ

546 SHAI MEIRI AND YORAM YOM-TOV

Starck 1998a, Starck and Ricklefs 1998). Al- nonmigratory species. Although this approach though this problem can be dealt with using a ignores the effects of phylogeny, we believe that phylogenetic correction, such as independent the use of all species added an aspect lacking in contrasts (Felsenstein 1985), we avoided using the other methods described above: we were in- this method for several reasons: ®rst, a complete terested not only in relative developmental pe- phylogeny of all relevant taxa is unavailable; riods of closely related taxa, but also in absolute and second, the high lability of developmental values. These are biologically distinct questions times could seriously impair the results of such (Smith 1994): with this analysis we asked corrections (Felsenstein 1985, Gittleman et al. whether migrants can possess a developmental 1996, Cunningham et al. 1998, Losos and Glor period as long as that of the slowest-developing 2003). As we show below, developmental peri- sedentary species, regardless of phylogenetic af- ods are highly variable even between closely re- ®nity. All statistical tests were run using STA- lated species. TISTICA 6 (StatSoft 2003). We used two separate analyses to control for phylogenetic differences in developmental peri- RESULTS ods between distantly related taxa. We thus Our analyses indicated that sedentary species avoided in¯ating the number of degrees of free- probably tend to lay larger clutches than do mi- dom without sacri®cing too much data. We ®rst ϭ gratory ones in the same families (F1,68 3.2, P compared the residuals of a regression of devel- ϭ 0.08; Table 1). The variation in clutch size opmental periods on body mass of migrants and between families (F ϭ 46.8, P Ͻ 0.001) and nonmigrants within orders and also within fam- 5,68 the signi®cant interaction between factors (F ilies in which some species migrate while others 5,68 ϭ 6. 8, P Ͻ 0.001) prevented an unequivocal do not. Because most of the variation in incu- conclusion on the in¯uence of migratory behav- bation periods resides at these levels (Starck and ior on clutch size. Ricklefs 1998) it is reasonable to assume that Comparing the residuals of a regression of de- this method removes most of the phylogeny ef- velopmental periods on body size in migrants fect related to developmental strategies (Ricklefs and nonmigrants within orders and families in and Starck 1998b). which both modes exist, reveals that migrants We also applied a conservative method pro- ϭ ϭ indeed develop faster (orders: F1,81 9.8, P posed by Felsenstein (1985), and paired species ϭ ϭ of large, closely related birds in which one 0.002; families: F1,68 9.1, P 0.004). Ordinal member migrates while the other does not. We or familial af®nity in itself did not affect the ϭ ϭ chose taxa for this comparison primarily from residuals (orders: F4,81 2.0, P 0.10; families: ϭ ϭ estimates of their phylogenetic relationships F5,68 1.5, P 0.21). An interaction with mi- (Ricklefs and Starck 1996). If a phylogenetic hy- gratory status was apparent in both cases (or- ϭ ϭ ϭ pothesis was lacking, or either all migrants or all ders: F4,81 4.3, P 0.003; families: F5,68 ϭ sedentary species formed a monophyletic clade, 2.3, P 0.05). Analyzing the residuals of in- we ranked migrant and nonmigrant congeners cubation and ¯edging times within families con- separately by mass, and compared pairs of equal taining both migrants and resident species re- rank. We used the phylogeny of Donne-Gousse vealed that incubation periods did not differ be- ϭ et al. (2002) for Anserinae (with Branta as sister tween migrants and nonmigrants (F1,68 1.1, P ϭ ϭ group for Anas), Mooers et al. (1999) for Grui- 0.30), but that migrants ¯edged faster (F1,68 nae, Roulin and Wink (2004) for Accipitridae, 11.8, P ϭ 0.001). Familial af®nity did not in¯u- ϭ and Slikas (1997) for Ciconiidae. We ran a Wil- ence the residuals of either incubation (F5,68 ϭ ϭ ϭ coxon signed-ranks paired test (Sokal and Rohlf 0.6, P 0.69) or ¯edging (F5,68 1.5, P 0.19) 1995) in order to determine whether migrants times. There were no interactions with family ϭ ϭ ϭ develop faster than nonmigrants. (incubation F5,68 1.7, P 0.16; ¯edging F5,68 In addition, we analyzed all species of birds 2.1, P ϭ 0.08). in our sample, including members of orders in ANCOVA showed that large migrants devel- which all species have a uniform pattern of mi- op faster than large sedentary birds. This was ϭ gration. We used an ANCOVA (Sokal and Rohlf true whether we used all large species (F1,117 1995), with body mass as the covariate, in order 30.7, P Ͻ 0.001), or only species from orders ϭ Ͻ ϭ to compare development time of migratory and (F1,88 26.6, P 0.001) and families (F1,77 CONDOR Tuesday May 25 2004 01:37 PM cond 106_309 Mp_547 Allen Press • DTPro System GALLEY File # 09TQ

ONTOGENY OF LARGE BIRDS 547

25.4, P Ͻ 0.001) in which there are both mi- city of food. Geffen and Yom-Tov (2000) found grants and sedentary species. no differences in developmental periods of trop- In pairs of large closely related species, mi- ical vs. temperate . It is not clear if grants develop faster (Wilcoxon signed-ranks this situation is different for other orders. test, n ϭ 16 pairs, T ϭ 9, P ϭ 0.002). This is An interesting question that remains to be an- achieved through faster ¯edging (T ϭ 3, P ϭ swered is whether migration distance affects de- 0.002), whereas incubation periods did not differ velopment time. Species with long migration (T ϭ 49.5, P ϭ 0.55). routes may have to accelerate their development relative to closely related species that are short- DISCUSSION range migrants. Similarly, species migrating Our results show that large migratory birds com- from high latitudes might be predicted to have plete their development faster than do similar shorter development than closely related species sized, closely related nonmigrants, and also tend migrating from lower latitudes. We do not, at to have smaller clutches. We suggest that the present, have suf®cient data to explore this ques- difference between the patterns for migratory tion, but the three following examples indicate and sedentary birds results from the selective the probable validity of this prediction. The Tun- pressure on large migrants to complete their de- dra Swan, migrating from the highest latitudes velopment before the fall migration commences. and covering the greatest distance of all migra- The longest development time we found for a tory swans, has the shortest developmental pe- migrant was 171 days for the Mute Swan, one riod of all its congeners. In migratory geese, of the largest of all migrants. Longer develop- species migrating over long distances (Snow mental periods may not enable juveniles of mi- Goose, White-fronted Goose, Bean Goose) seem grant species to complete their growth before to have developmental periods over 15% shorter fall migration commences. This might be the than those of species migrating short distances reason why no migrant birds are larger than the (Bar-headed Goose, Pink-footed Goose, Emper- present-day Cygnus, whereas Palmqvist and or Goose), although the Pink-footed Goose and VizcaõÂno (2003) deduced that the 80-kg teratorn Emperor Goose also breed at very high latitudes Argentavis magni®cens had an extremely long (del Hoyo et al. 1992). The Black-necked Crane development, and must have been sedentary in (Grus nigricollis), the slowest developer among habits. migratory species of its genus, migrates between A longer developmental period may enhance high and low altitudes in the Himalaya region the probability of juvenile survival (Ricklefs (del Hoyo 1992), whereas other cranes often un- 1992, Martin 1996). If migrants have shorter de- dergo very long migrations. Clearly these pat- velopmental periods, this can be seen as incur- terns deserve further investigation. ring an additional cost for them, on top of lower In conclusion, we suggest that the shorter in- annual fecundity (BoÈhning-Gaese et al. 2000) cubation and ¯edging periods of large-bodied and survival (Cox 1985). migrants are a result of constraints imposed Another possible explanation for the shorter upon them by onset of fall migration. The length developmental periods of migrants is that during of summer at high latitudes may limit the length the breeding season at high latitude, migrants are of the developmental periods of birds breeding able to obtain more resources due to high food there, and thus the size of migratory birds breed- availability and longer foraging hours in com- ing in these regions. parison with the more equatorial breeders (Cox 1985, Safriel 1995). Most migrants in our sam- ACKNOWLEDGMENTS ple breed in high latitudes, whereas most sed- We thank Uri Roll, Amos Ar, and two anonymous re- entary species breed in more equatorial ones. viewers for valuable suggestions and discussion, and Therefore the pattern we describe may result Naomi Paz for many useful editorial comments. from differences in the amount of food available LITERATURE CITED for low vs. high-latitude breeding birds. Hence it is possible that migrant (high latitude) species BOÈ HNING-GAESE, K., B. HALBE,N.LEMOINE, AND R. OBERRATH. 2000. Factors in¯uencing the clutch are not forced to shorten their developmental pe- size, number of broods and annual fecundity of riods, but rather that sedentary (low latitude) North American and European land birds. Evo- species may be constrained by the relative scar- lutionary Ecology Research 2:823±839. CONDOR Tuesday May 25 2004 01:37 PM cond 106_309 Mp_548 Allen Press • DTPro System GALLEY File # 09TQ

548 SHAI MEIRI AND YORAM YOM-TOV

CALDER, W. A., III. 1984. Size, function and life his- PALMQVIST,P.,AND S. F. VIZCAÂõNO. 2003. Ecological tory. Harvard University Press, Cambridge, MA. and reproductive constraints of body size in the CARRIER, D. R., AND J. AURIEMMA. 1992. A develop- gigantic Argentavis magni®cens (Aves, Therator- mental constraint on the ¯edging time of birds. nithidae) from the of Argentina. Ameg- Biological Journal of the Linnean Society 47:61± hiniana 40:379±385. 77. RAHN, H., C. V. PAGANELLI, AND A. AR. 1975. Relation COLLAR, N. J. 1996. Otididae, p. 240±275. In J. del of avian egg weight to body weight. Auk 92:750± Hoyo, A. Elliott, and J. Sargatal [EDS.], Handbook 765. of the birds of the world. Vol. 3.Lynx Edicions, REILLY, P. 1994. Penguins of the world. Oxford Uni- Barcelona. versity Press, Oxford, UK. COX, G. W. 1985. The evolution of avian migration RICKLEFS, R. E. 1973. Patterns of growth in birds. II. systems between temperate and tropical regions of Growth rates and mode of development. Ibis 115: the New World. American Naturalist 126:451± 177±201. 474. RICKLEFS, R. E. 1992. Embryonic development period CUNNINGHAM, C. W., K. E. OMLAND, AND T. H. OAK- and the prevalence of avian blood parasites. Pro- LEY. 1998. Reconstructing ancestral character ceedings of the National Academy of Sciences 89: states: a critical reappraisal. Trends in Ecology & 4722±4725. Evolution 13:361±366. RICKLEFS, R. E., AND J. M. STARCK. 1998a. Avian em- DEL HOYO, J., A. ELLIOTT, AND J. SARGATAL. 1992. bryonic growth, p. 31±58. In J. M. Starck and R. Handbook of the birds of the world. Lynx Edi- E. Ricklefs [EDS.], Avian growth and develop- cions, Barcelona. ment. Evolution within the altricial-precocial DONNE-GOUSSEÂ , C., V. LAUDET, AND C. HAÈ NNI. 2002. spectrum.Oxford University Press, New York. A molecular phylogeny of Anseriformes based on RICKLEFS, R. E., AND J. M. STARCK. 1998b. The evo- mitochondrial DNA analysis. Molecular Phylo- lution of the developmental mode in birds, p. 366± and Evolution 23:339±356. 380. In J. M. Starck and R. E. Ricklefs [EDS.], DUNNING, J. B., JR. 1993. CRC handbook of avian Avian growth and development. Evolution within body masses. CRC Press, London. the altricial-precocial spectrum.Oxford University FELSENSTEIN, J. 1985. Phylogenies and the comparative Press, New York. method. American Naturalist 70:1±15. ROULIN,A.AND M. WINK. 2004. Predator-prey rela- GEFFEN, E., AND Y. Y OM-TOV. 2000. Are incubation tionships and the evolution of colour polymor- and ¯edging periods longer in the tropics? Journal phism: a comparative analysis in diurnal raptors. of Ecology 69:59±73. Biological Journal of the Linnean Society 81:565± GITTLEMAN, J. L., C. G. ANDERSON,M.KOT, AND H. 578. K. LUH. 1996. Phylogenetic lability and rates of SAFRIEL, U. 1995. The evolution of Palearctic migra- evolution: a comparison of behavioral, morpho- tionÐthe case for Southern ancestry. Jour- logical and life history traits, p. 166±205. In E. P. nal of Zoology 41:417±431. Martins [ED.], Phylogenies and the comparative SCHMIDT-NIELSEN, K. 1984. Scaling. Why is animal method in animal behavior.Oxford University size so important? Cambridge University Press, Press, Oxford, UK. Cambridge, UK. GRASZYNSKI, K., B. KOMISCHKE, AND B.-U. MEYBURG. SLIKAS, B. 1997. Phylogeny of the avian family Ci- 2002. On the biology of the Greater Spotted Ea- coniidae () based on cytochrome b sequenc- gle, (Aquila clanga Pallas 1811), p. 62±75. In R. es and DNA±DNA hybridization distances. Mo- Yosef, M. L. Miller, and D. Pepler [EDS.], Raptors lecular Phylogenetics and Evolution 8:275±300. in the new millennium. Israel International Bird- SMITH, R. J. 1994. Degrees of freedom in interspeci®c ing and Research Centre in Eilat,Eilat, Israel. allometry: an adjustment for the effects of phy- HEDENSTROÈ M, A., AND T. A LERSTAM. 1998. How fast logenetic constraint. American Journal of Physical can birds migrate? Journal of Avian Biology 29: Anthropology 93:95±107. 424±432. SOKAL, R. R., AND F. J. ROHLF. 1995. Biometry. 3rd LOSOS, J. B., AND R. E. GLOR. 2003. Phylogenetic com- ed. W. H. Freeman, New York. parative methods and the geography of speciation. STARCK,J.M.,AND R. E. RICKLEFS. 1998. Patterns of Trends in Ecology & Evolution 18:220±227. development: the altricial precocial spectrum, p. MARTIN, T. E. 1996. Life history evolution in tropical 3±30. In J. M. Starck and R. E. Ricklefs [EDS.], and south temperate birds: what do we really Avian growth and development. Evolution within know? Journal of Avian Biology 27:1±10. the altricial-precocial spectrum.Oxford University MAURER, B. A. 1998. The evolution of body size in Press, New York. birds. II. The role of reproductive power. Evolu- STATSOFT. 2003. STATISTICA (data analysis software tionary Ecology 12:935±944. system) version 6. StatSoft Israel Ltd., Kefar Saba, MOOERS, A. é., S. M. VAMOSI, AND D. SCHLUTER. Israel. 1999. Using phylogenies to test macroevolution- YOM-TOV,Y.,AND A. AR. 1993. Incubation and ¯edg- ary hypotheses of trait evolution in cranes (Grui- ing durations of woodpeckers. Condor 95:282± nae). American Naturalist 154:249±259. 287.