ISRAEL JOURNAL OF ZOOLOGY, Vol. 49, 2003, pp. 159–173

*Author to whom correspondence should be addressed. E-mail: izhaki@research.haifa.ac.ilAccepted October 2003.

DEVELOPMENTAL RATES AND MORPHOMETRICS OF THESYMPATRIC PYGMY CORMORANT (PHALACROCORAX PYGMEUS) AND

GREAT CORMORANT (P. CARBO SINENSIS)

MARVA SHMUELI,a ZEEV ARAD,a GADI KATZIR,b AND IDO IZHAKIb,*aDepartment of Biology, Technion—Israel Institute of Technology, Haifa 32000, Israel

bDepartment of Biology, Faculty of Science and Science Education,University of Haifa—Oranim, Qiryat Tiv’on 36006, Israel

ABSTRACT

We compared growth rates and adult morphological traits in two sympatriccormorant species, the pygmy cormorant (Phalacrocorax pygmeus) and thegreat cormorant (P. carbo sinensis), in Israel. The smaller P. pygmeus exhib-ited higher developmental rates than P. carbo sinensis, as expressed in thegrowth rate constant (K) of body mass and of various body parts (bill, wing,tarsus, primaries, and tibia lengths). The consequences of the higher develop-mental rate of P. pygmeus are early fledging and a relatively low body mass offledglings. We suggest that several proximate ecological and developmentalfactors such as risk of nest predation, body temperature regulation, andhydrodynamics act in concert to promote rapid development in chicks ofP. pygmeus. However, the four-fold lower body mass of the adult P. pygmeusis probably the most important physiological constraint that might explain itsrapid growth rate in comparison with P. carbo sinensis. The body mass andthe size of various body parts of adult P. carbo sinensis are much higher thanthose of adult P. pygmeus. This difference in adult morphological attributes,together with the marked differences in growth rates between the two species,should be reflected in different ecological functions that promote ecologicalsegregation between them. Therefore, the conservation policies and futurepractical solutions of the cormorant–fisheries conflict should be species-specific.

INTRODUCTION

The pygmy cormorant (Phalacrocorax pygmeus) and the great cormorant (P. carbosinensis) are sympatric in certain southeastern regions of the western Palearctic(Johnsgard, 1993). In Israel, P. pygmeus is an all-year resident, whereas P. carbosinensis is a wintering species (Johnsgard, 1993; Shy and Frankenberg, 1995; Koshelevet al., 1997). Both species became almost extinct in Israel during the 1950s due todrainage of the Hula swamps and to massive hunting (Shy and Frankenberg, 1995).P. pygmeus is currently considered an endangered species (Crivelli et al., 2000). The

160 SHMUELI ET AL. Isr. J. Zool.

number of its nests in Israel has increased, however, from 60 in 1998 to between 110 and155 in 2001 (Nemtzov, 2001). Its few breeding colonies in Israel are located mainlyalong the Jordan Valley. Recently established new colonies have expanded this species’breeding zone into northern Israel. The number of P. carbo sinensis wintering in Israelhas increased from ~5000 in 1990 up to 16,000–20,000 in recent years (Shy and Geva,1998), concurrent with its dramatic increase in number in Europe over the last decade(Koshelev et al., 1997).

Foraging sites of the two species in Israel largely overlap as they both feed in LakeKinneret (Sea of Galilee) and in the same artificial fishponds. Fishermen in Israel claimthat their fish yields from artificial ponds are being severely damaged by these twospecies, and they therefore use a variety of techniques to deter both cormorants andpelicans (Izhaki, 1994; Shy and Geva, 1998). The conflict between cormorants andfishermen is escalating, and finding practical solutions to prevent further harm to thesespecies in Israel largely depends upon our better understanding their ecology. The aim ofthis study was to evaluate the possible ecological overlap between the two cormorantspecies based on body growth patterns and morphological traits.

Because the rates of development of the young of many bird species are relativelyconstant, it has been suggested that they are determined mainly by hereditary factors(Lack, 1968). Different patterns of body growth among bird species are mainly attrib-uted to differing developmental strategies along the altricial–precocial spectrum (Starckand Ricklefs, 1998a). It was argued that avian body development represents a compro-mise between environmental selective factors and internal constraints, and thus hasimplications on fitness (Lack, 1968; but also see Ricklefs et al., 1998). A rapid bodygrowth was predicted where chicks are under severe risk of mortality or when theyintensively compete with each other, while a slow growth was predicted where foodsupply is limited (Lack, 1968). We assume that differences in the patterns of bodygrowth of two congeneric species should represent, at least partially, ecological differ-ences between them, although recent comprehensive analyses have revealed the com-plexity of the relationships between growth patterns and fitness (Ricklefs et al., 1998;Royle et al., 1999). Furthermore, we assume that differences in growth patterns amongbody parts (e.g., tarsus, bill, wing, etc.) represent ecological adaptations. Resourceallocation to different body parts during growth depends on functional priorities, whichmay differ among species. Here we compared the body growth rates of the twocormorant species in captivity under similar environmental conditions, thus eliminatingecological factors that may affect the growth pattern of a species, such as food availabil-ity and predation risks, (Kirkham and Montevecchi, 1982; Starck and Ricklefs,1998a,b,c). We interpreted differences in growth patterns between species as differencesin their adaptations and ecological function.

The species’ common ancestry should result in morphological similarities (Starckand Ricklefs, 1998b). Morphological traits may represent aspects of the relationshipsbetween the birds and their specific environment and therefore reflect their ecology(Leisler and Winkler, 1985). Many studies have demonstrated correlations betweenmorphological traits and habitat use, diet breadth, foraging behavior, and locomotory

Vol. 49, 2003 DEVELOPMENTAL RATES AND MORPHOMETRICS OF CORMORANT 161

performance (Ricklefs and Miles, 1994). Hence, we may understand the ecologicalfunctioning of a species and of individuals within the species (e.g., different sexes) in aspecific habitat by analyzing their morphological traits (e.g., Leisler and Winkler, 1985;Ricklefs and Miles, 1994). Here we compared the adult morphological traits (body mass,length of wing, bill, tarsus, tibia, and primaries) of the two cormorant species and ofmales and females as an indication of their ecological similarity. High ecologicalsimilarity may limit their ecological and geographical distribution and their coexistence(Ricklefs and Miles, 1994).

Specifically, the aims of the present study were: (a) To evaluate the interspecificdifferences in terms of growth patterns, including growth rate constants and the age atthe inflection point of the growth curves of body mass and of various body parts; (b) Toevaluate the interspecific and intersexual differences of adults in morphological traits(bill, wing, tarsus, primaries, and tibia lengths and body mass); (c) To correlate morpho-logical traits within each species and to compare these correlations between species; (d)To determine the morphological traits that best discriminate between the two species andbetween sexes within each species; and (e) Based on the integration of the above, toassess the ecological similarity between these two species.

METHODS

CAPTIVITY CONDITIONSChicks of P. pygmeus were collected, under license from the Israel Authority for

Nature Conservation, from breeding colonies in the Jordan and Bet She’an valleys. Inaddition, we used P. pygmeus from successful breeding in our own aviaries. P. carbosinensis eggs and chicks of unknown age were obtained from the Jerusalem Biblical Zooand from the Safari Park in Ramat Gan, Israel. The eggs and nestlings were placed in anincubator for up to one week and subsequently transferred to padded boxes inside theoutdoor cages, which were heated at night. The nestlings of each species were thentransferred to separate outdoor enclosures (12 × 5 m) at Kefar Yehoshua that includedwater pools. The cormorants were hand-fed twice a day on small fish (Tilapia sp., Carpsp., and cultured species such as comet, koi, and shibunkin). In total, we kept 20P. pygmeus and 17 P. carbo sinensis during the three years of study.

MORPHOLOGICAL TRAITSDuring the first 7 days following hatching or collection of nestlings we measured

body mass (digital scales; Acculab 5001; ± 1 g), bill length (upper bill), wing length(carpometacarpus), tarsal length (tarsometatarsus), tibia length (tibiotarsus), and prima-ries length, daily (ruler; ± 1 mm). We subsequently measured the birds every 2–3 daysfor the first 2 months and once a week thereafter up to the age of 5–6 months. Body masswas followed up to 2–3 years of age. Asymptotic values of the morphometric traits andof body mass were determined at the age of 35–60 days in P. pygmeus and at the age of50–70 days in P. carbo sinensis, following which no further change in these traits was

162 SHMUELI ET AL. Isr. J. Zool.

observed during a period of at least 30 days. Sex was determined based on the breedingbehavior of the birds upon reaching sexual maturity.

AGE STANDARDIZATIONFor age standardization of nestlings that were collected from nests at various un-

known precise ages, we calculated regression equations of wing and tarsus length foreach species as a function of age, based on individuals that had been measured sincehatching. We chose (a) wing length as the best parameter for age prediction in altricialnestlings due to its slow growth (Kirkham and Montevecchi, 1982) and (b) tarsus lengthas it demonstrated the most uniform growth rate. Both these traits were shown to be thebest predictors of age in the semi-precocial herring gull (Larus argentatus) because theirgrowth curves are relatively linear (Dunn and Brisbin, 1980). Our equations were basedon eight P. pygmeus and five P. carbo sinensis nestlings and calculated for the firstgrowth period, where the growth curve was linear. The R2 values for all four equations (2species × 2 traits) were >0.92 (Table 1). We subsequently estimated the age of theunknown-aged nestlings of each species by averaging the values obtained from thesetwo equations (Table 1).

GROWTH RATESThe logistic growth rate constant (K) and the age (days) of the inflection point of the

growth curve (ti) for body mass and each of the morphometric traits were determined

according to the widely-used nonlinear logistic equation for birds (Starck and Ricklefs,1998c). The equation, Y = A / (1 + e (–K * (t – ti))) [Y is the measured growth trait (bodymass, bill length, etc.), A is the asymptotic size, K is the rate constant of the equation,and t is the age of the nestling in days] was found to be especially accurate forcormorants (Dunn, 1975; Goutner et al., 1997; Ostnes et al., 2001). Our relatively smallsample size did not allow us to calculate a separate growth curve for each sex. Therefore,the growth curves were calculated for pooled data of all individuals.

To determine the nestlings’ age at which asymptotic sizes were reached, we followedPinshow et al. (1976) and calculated, for each morphological trait, the convergence pointof the two best-fitted linear lines: the upward increase of the first part of the growth

Table 1Wing length (WL) and tarsus length (TL) as a function of nestling age (AGE) within the lineargrowth range of both cormorant species (sexes pooled). Calculations were based on nestlings of

known age that had hatched in captivity

Phalacrocorax pygmeus (n = 8) Phalacrocorax carbo sinensis (n = 5)

Trait Range Growth R2 Range Growth R2

(mm) (days) equation (days) equation

WL 0–43 WL = 5.26 × AGE – 3.60 0.96*** 2–52 WL = 6.80 × AGE – 35.0 0.97***TL 0–19 TL = 1.95 × AGE + 6.23 0.92*** 2–29 TL = 2.31 × AGE + 3.90 0.96***

*** = p < 0.001.

Vol. 49, 2003 DEVELOPMENTAL RATES AND MORPHOMETRICS OF CORMORANT 163

curve (for all parameters and body mass R2 > 0.92) and the minimal incline (for allparameters and body mass R2 < 0.07) of the second part of the growth curve.

STATISTICAL ANALYSISWe used Student’s t-test to compare morphological traits between adults of the two

sexes in each species. Pearson coefficient (r) was used to correlate morphologicalparameters with each other for each species. Because we conducted 15 pairwisecorrelations we kept the overall experiment error rate to 5% using the Bonfferoniadjustment by evaluating each of our pairwise correlations against p < 0.003 (0.05/15).A discriminant analysis was used to determine the most parsimonious morphometricvariables to distinguish between the two species, and then between sexes within species.Wilks’ lambda was used to indicate a linear discriminant function (equation predictive)providing the best separation between groups. We also calculated the percentage ofcorrect classifications yielded from the equation. Statistical analyses were performedusing SYSTAT (SYSTAT, 1998). Data are presented as means ± s.e.

RESULTS

GROWTH PATTERNS: COMPARISON BETWEEN SPECIESThe different morphological traits of both species had different growth periods

(Table 2). P. pygmeus reached its tarsal asymptotic level (thus completing its growth)

Table 2Calculated growth values from the logistic equation (± s.e.) of the morphometric variables of thepygmy cormorant P. pygmeus (n = 18) and the great cormorant P. carbo sinensis (n = 11) (sexespooled). A, asymptotic value; T

(A), asymptotic age (days); K, growth rate constant (day–1); T

i, age

(days) at the inflection point. R2 of each equation is also included

Species A T(A)

K s.e. Ti

s.e. R2

Body mass (g) Pygmy 484.9 25.2 0.231 0.11 13.6 0.204 0.95Great 2124.3 42.7 0.163 0.005 24.5 0.204 0.97

Tarsus length Pygmy 39.4 16.9 0.268 0.011 6.5 0.152 0.95 (mm) Great 70.2 26.9 0.176 0.007 13.6 0.231 0.96

Tibia length Pygmy 77.9 17.3 0.261 0.011 6.7 0.157 0.95 (mm) Great 134.7 29.9 0.161 0.004 14.2 0.175 0.98

Bill length Pygmy 29.5 26.9 0.150 0.005 10.8 0.199 0.96 (mm) Great 68.2 37.1 0.122 0.004 16.5 0.253 0.97

Wing length Pygmy 203.8 36.8 0.145 0.003 18.9 0.151 0.98 (mm) Great 339.7 54.4 0.115 0.002 30.1 0.174 0.99

Primaries Pygmy 129.1 48.7 0.176 0.006 28.9 0.214 0.98 length (mm) Great 190.0 58.1 0.151 0.005 40.5 0.255 0.98

164 SHMUELI ET AL. Isr. J. Zool.

Fig. 1. Logistic growth curves of body mass, tarsus length, and tibia length in relation to age (days)in the P. pygmeus (n = 18) and in the P. carbo sinensis (n = 11).

Vol. 49, 2003 DEVELOPMENTAL RATES AND MORPHOMETRICS OF CORMORANT 165

Fig. 2. Logistic growth curves of bill length, wing length, and primaries length in relation to age(days) in the P. pygmeus (n = 18) and in the P. carbo sinensis (n = 11).

166 SHMUELI ET AL. Isr. J. Zool.

after 16.9 days but reached its asymptotic level of primaries growth after 48.7 days, withall other traits being in-between (Table 2). P. carbo sinensis completed the growth of allbody parts much later (27–58 days, Table 2). The growth rate coefficient (K) of thetarsus was the highest of all the measured parameters in both species (sexes pooled;Fig. 1 and Table 2). Following in order were the growth rate constants of tibia length,body mass, primaries length, bill length, and finally wing length (Fig. 2 and Table 2).

Growth rate coefficients of all morphological traits were 17–62% higher in P. pygmeusthan those in P. carbo sinensis (Table 2). The asymptotic body mass of P. pygmeus(484.9 g) was 38 times its body mass at hatching (12.9 ± 0.7 g, n = 6), while the asymptoticbody mass of P. carbo sinensis (2124.3 g) was 55 times its body mass at hatching (39.0 ±1.7 g, n = 6). But since P. pygmeus reached its asymptotic body mass after 25.2 days, whileP. carbo sinensis reached it after 42.7 days, the growth rate coefficient was higher inP. pygmeus than in P. carbo sinensis (K = 0.231 and K = 0.163, respectively).

MORPHOMETRY: COMPARISON BETWEEN SPECIES AND SEXESValues obtained for all morphological traits of adults were lower in P. pygmeus than

in P. carbo sinensis. However, while the asymptotic body mass of P. carbo sinensis was4.4 times higher than that of P. pygmeus the ratio of the other traits was much lower: 2.3for bill length, 1.7–1.8 for tarsus, tibia, and wing length, and 1.5 for primaries (Table 2,sexes pooled). Thus, P. pygmeus is characterized by larger extremities relative to itsbody mass. A discriminant analysis of the morphological traits between the two speciesrevealed that tarsus length was the perfect indicator of species classification with overall100% correct classification (Wilks’ lambda = 0.014, n = 37, p < 0.001).

Only one significant correlation (between tibia and primaries length) was foundamong asymptotic morphological traits in P. pygmeus, whereas in P. carbo sinensis, allcorrelations were significant (Table 3). Thus, in P. carbo sinensis the ratio between thesizes of different body parts was consistent among individuals, while this was not thecase in P. pygmeus.

P. pygmeus males were 10% heavier than the females and their wings were only 3%larger, whereas all other morphometric traits were similar in size in both sexes (Table 4).According to the discrimination analysis, the best predictor of sex in P. pygmeus was thebody mass with 100% correct classification (Wilks’ lambda = 0.674, n = 20, p < 0.05).The males of P. carbo sinensis were 34% heavier than the females and their wing was8% larger. Male’s bill, tarsus, tibia, and primaries were 13%, 6%, 10%, and 15% larger,respectively (Table 4). Body mass and primaries length were the best predictors of sex inP. carbo sinensis, with an overall 100% correct classification (Wilks’ lambda = 0.072,n = 15, p < 0.001).

DISCUSSION

DEVELOPMENTAL RATES OF THE TWO CORMORANT SPECIESThe growth rate constants (K) of P. pygmeus and P. carbo sinensis found in this study

were, respectively, 70% and 92% higher than expected from the allometric equation of

Vol. 49, 2003 DEVELOPMENTAL RATES AND MORPHOMETRICS OF CORMORANT 167

Tab

le 3

Pear

son

corr

elat

ion

coef

fici

ents

(r)

bet

wee

n m

orph

olog

ical

tra

its o

f py

gmy

corm

oran

ts (

P. p

ygm

eus)

at

the

age

of 5

0 da

ys a

nd o

f gr

eat

corm

oran

ts (

P.

carb

o si

nens

is)

at t

he a

ge o

f 70

day

s (s

exes

poo

led)

. O

nly

sign

ific

ant

resu

lts (

p <

0.0

5) a

fter

a B

onff

eron

i ad

just

men

t ar

epr

esen

ted.

Sam

ple

size

is in

dica

ted

in p

aren

thes

es

Bod

y m

ass

Bill

leng

thW

ing

leng

thT

arsu

s le

ngth

Tib

ia le

ngth

Pygm

yG

reat

Pygm

yG

reat

Pygm

yG

reat

Pygm

yG

reat

Pygm

yG

reat

Bill

leng

th0.

88 (

17)

Win

g le

ngth

0.67

(16

)0.

72 (

16)

Tar

sus

leng

th0.

78 (

17)

0.78

(17

)0.

58 (

16)

Tib

ia le

ngth

0.86

(17

)0.

85 (

17)

0.64

(18

)0.

85 (

16)

0.86

(17

)Pr

imar

ies

leng

th0.

67 (

15)

0.66

(17

)0.

83 (

15)

0.61

(17

)0.

83 (

15)

0.83

(17

)0.

84 (

15)

168 SHMUELI ET AL. Isr. J. Zool.

K as a function of asymptotic mass (A, in grams), Log K = –0.316 log A – 0.017 (Starckand Ricklefs, 1998b), in accordance with previous findings in other cormorant species(Dunn, 1975; Ostnes et al., 2001). Moreover, the growth rate of body mass inP. pygmeus (K = 0.231 day–1) was higher than that of P. carbo sinensis (K = 0.163 day–1)and other larger cormorant species (Dunn, 1975; Ostnes et al., 2001). Likewise, thedevelopmental rates of all body parts were much higher in P. pygmeus than in P. carbosinensis. Tarsus length in both species increased faster than all other morphologicaltraits, and much faster in P. pygmeus (K

tarsus = 0.268 day–1) than in P. carbo sinensis

Table 4Asymptotic values of morphometric parameters (age 50–70 days) of both cormorant species in

captivity (means ± s.e.)

Phalacrocorax pygmeus Phalacrocorax carbo sinensis

Males Females t Males Females t

Body mass 524.9 ± 10.2 476.3 ± 8.2 3.71** 2270.6 ± 48.7 1698.3 ± 80.5 6.08*** (g)Range (n) 472–567 430–520 2070–2550 1495–2000

(8) (12) (10) (7)

Bill length 29.4 ± 0.4 28.7 ± 0.4 1.21ns 71.0 ± 0.7 62.6 ± 0.8 8.3*** (mm)Range (n) 28–31 26–30 68–75 60–65

(8) (10) (10) (7)

Wing length 203.8 ± 1.6 197.7 ± 1.8 2.55* 347.0 ± 3.7 321.7 ± 1.3 6.29*** (mm)Range (n) 200–210 185–205 330–360 320–330

(8) (10) (10) (6)

Tarsus length 39.8 ± 0.4 39.0 ± 0.3 1.7ns 71.1 ± 0.5 66.9 ± 0.8 4.32*** (mm)Range (n) 38–41 38–41 68–73 64–70

(8) (12) (10) (7)

Tibia length 79.8 ± 0.3 77.3 ± 1.1 1.78ns 138.5 ± 0.8 126.0 ± 1.5 7.21*** (mm)Range (n) 78–81 68–80 135–142 120–130

(8) (12) (10) (7)

Primaries 128.6 ± 1.4 123.5 ± 3.0 1.34ns 208.3 ± 2.4 180.8 ± 3.8 6.21*** length (mm)Range (n) 120–130 105–130 195–220 165–190

(7) (10) (9) (6)

* = p < 0.05. ** = 0.05 > p > 0.01. *** = p < 0.001. ns = not significant. Sample size is indicated inparentheses.

Vol. 49, 2003 DEVELOPMENTAL RATES AND MORPHOMETRICS OF CORMORANT 169

(Ktarsus

= 0.176 day–1). A faster tarsus growth and a slower wing growth relative to otherorgans were also found in other altricial birds (Kirkham and Montevecchi, 1982; Ostneset al., 2001). O’Connor (1984) suggested that resource allocation to different body partsis timed in relation to their functional importance (see below).

The high developmental rates of P. pygmeus are the result of (a) a lower asymptoticbody mass of its chicks at fledging relative to the adult body mass: the asymptotic bodymass of fledgling P. pygmeus was 13% lower than that of the adults (485 g vs. 555 g)compared with 8% in P. carbo sinensis (2124 g vs. 2310 g), and (b) early fledging of thechicks: assuming that fledging time of cormorants is twice that of the inflection point (t

i)

of the growth curve (Platteeuw et al., 1995), we can estimate the fledging time in thepresent study as 27 days in P. pygmeus and 49 days in P. carbo sinensis.

Several, non-exclusive, proximate ecological and developmental factors have beenoffered to explain interspecific differences in developmental rates. We next examinedthe possibility that some of these factors are responsible for the higher developmentalrate of P. pygmeus relative to P. carbo sinensis.

Sibling competition. Lack (1968) suggested that because parental ability to supplyfood to the chicks is limited, they should either raise only a few nestlings that willdevelop fast or raise many nestlings that will develop slowly. Although the clutch size(4–6) of the P. pygmeus (Johnsgard, 1993) is larger than that of P. carbo sinensis(between 2.6 in Poland and 3.83 in Italy: Pajkert and Górski, 1996; Volponi, 1999), theformer develop much faster. Thus, our present findings are in accord with the positivecorrelation between clutch size and growth rate that was found among 40 bird species(Royle et al., 1999). It was also suggested that strong intra-brood competition forresources exerts a major selection pressure for the evolution of rapid growth rates inbirds (Werschkul and Jackson, 1979). However, to date, there is no direct evidence thatsibling competition is more intensive in P. pygmeus than in P. carbo sinensis.

Nest predation. Differential nest predation among bird species may also explain differ-ential growth rates because selection should favor rapid growth rates to minimize the periodthat eggs and nestling are exposed to predators (Lack, 1968; Bosque and Bosque, 1995).The rate of nest predation in altricial birds decreases with increasing adult body size becausesmall prey species may be taken by a wider variety of predators than larger species (Bosqueand Bosque, 1995). Furthermore, clutch size is positively correlated with nest predation(Lack, 1968). We therefore expect nestlings of P. pygmeus to be more vulnerable topredation than those of P. carbo sinensis. Hence, the rapid development of overall body sizeand legs of P. pygmeus could be a consequence of strong selection to reduce the time inwhich the nestlings are exposed to predators. Although eggs and chick predation has alreadybeen documented in both species (Johnsgard, 1993; Volponi, 1999), no study has yetidentified heavier nest predation in P. pygmeus than in P. carbo sinensis.

Body temperature regulation. The newly-hatched, heterothermic, altricial nestlingsof the cormorants are typically naked, with limited physical and sensory capacities.Therefore, they are completely dependant on their parents for maintenance of bodytemperature (Kirkham and Montevecchi, 1982; Starck and Ricklefs, 1998a,c). It is

170 SHMUELI ET AL. Isr. J. Zool.

expected that smaller species will be under stronger selection to reduce the period ofexposure in the nest because their nestlings are more vulnerable to mortality caused byadverse weather (Bosque and Bosque, 1995). Rapid body development may, therefore,promote an early acquisition of homeostatic capabilities (Björklund, 1996). Thus, well-developed legs enable the chicks to push their body upwards in the nest and performgular fluttering for effective dissipation of body heat when exposed to high ambienttemperatures (Kirkham and Montevecchi, 1982), and increased shivering heat productionof the accompanying musculature in the cold (Montevecchi and Vaughan, 1989). Wesuggest that the rapid development of overall body size and legs is especially critical forP. pygmeus, which breeds in warmer climates, rather than for the larger P. carbo sinensis.

Body size. It was suggested that there is a trade-off between growth rate and asymp-totic body size (Charnov, 1993). Indeed, body mass in birds is negatively correlated withgrowth rate, probably because the ability of a nestling to sustain a given growth ratedepends on adult body mass (O’Connor, 1984). Hence, the fourfold higher body mass ofP. carbo sinensis is probably the most important physiological constraint that mightexplain its lower growth rate in comparison with P. pygmeus. Furthermore, several ofthe other factors that may affect growth rate are related to body size, such as nestpredation and environmental stress.

Hydrodynamics. Cormorants are feet-propelled during underwater swimming. As theforces driving swimming are proportional to the foot area, and as the cost of transportincreases with decreasing body mass (Schmidt-Nielsen, 1972), the larger feet (relative tobody mass) of P. pygmeus may contribute to its hydrodynamic performance. This mayalso explain the lower correlation coefficient of the relations between feet morphometryand body size that we demonstrated in P. pygmeus, compared to P. carbo sinensis.

Other factors. Other factors that have been suggested as affecting developmental rates inbirds probably do not apply here. Among them are the altricial–precocial continuum (Starckand Ricklefs, 1998a), as both species are altricial; different mating systems (Royle et al.,1999), as both species are probably monogamous (Johnsgard, 1993); different parental careamong sexes (but see Royle et al., 1999), as both parents in both species provide for theyoung (Johnsgard, 1993); and different nutritional quality of the food provided to thenestlings, as both species feed their young mainly on fish (Johnsgard, 1993).

ADULT ECO-MORPHOLOGY OF THE TWO CORMORANT SPECIES

Morphological traits and sexual size dimorphism. Sexual size dimorphism is knownin cormorants (Cooper, 1985; Koffijberg and Van Eerden, 1995; Kato et al., 1999,2000). The results of our study indicate that sexual size dimorphism is far greater inP. carbo sinensis than in P. pygmeus. A common viewpoint is that sexual dimorphismfacilitates the separation of feeding niches between sexes, thus reducing intraspecificfood competition. In cormorants, large males can dive deeper than females (Kato et al.,1998, 1999, 2000), maintain longer dive durations (Kato et al., 1998, 1999), and feed onlarger fish (Koffijberg and Van Eerden, 1995). From our results on morphological traits

Vol. 49, 2003 DEVELOPMENTAL RATES AND MORPHOMETRICS OF CORMORANT 171

between males and females, we predict that the above intersexual differences fitP. carbo sinensis to a much greater extent than they do P. pygmeus. It was argued thatdifferences between social mating systems and differences between sexes in parentalcare are related to the size dimorphism differences between species (Owens and Hartley,1998). However, this is not the case here, as both species are monogamous and bothparents similarly provide parental care (Johnsgard, 1993).

Comparison between species. Based on the large differences in morphological attributesbetween the two species, we predict a low niche overlap between them. Indeed, P. pygmeusfeed frequently on invertebrates (J.B. Kiss, personal communication) and on fish muchsmaller than those taken by P. carbo sinensis (Johnsgard, 1993; Crivelli et al., 2000), so thatonly a small proportion of interspecific diet overlap occurs. Furthermore, it seems that theirfeeding sites only partially overlap, as the P. carbo sinensis prefers brackish and saltywaters while P. pygmeus prefers non-brackish water (Crivelli et al., 2000). In cases wherebrackish water is unavailable, as in Lake Kinneret, P. pygmeus prefers to feed in shallowwater while P. carbo sinensis feeds in the deep water.

CONCLUSIONS

The different developmental rates and adult morphological traits of the two cormorantspecies reflect their distinct ecological functioning and the possibility that they willcoexist (see also Crivelli et al., 2000). Therefore, any future solution of the conflictbetween the cormorants and the fisheries should consider the different ecological needsof the two species. Moreover, any such solution will have to consider a possible collapseof the ecological system due to human activities, which might interfere with the currentecological segregation of the two species.

ACKNOWLEDGMENTS

This research was supported by grants from the Israeli Ministry of Science, from theTechnion-Haifa University Fund, from the Technion SAFRAM Fund, and from theForchheimer Foundation (to M.S.). We are grateful to Prof. Daniel Weihs for his helpfulcomments, Ofer Shmueli, Zvi Shmueli, Abraham Dagan, Yahali Dagan, and RuthAlmon for their help in obtaining and raising the cormorant nestlings during the study; toShoshana Goldenberg for her technical assistance; to the staffs of the Safari Park inRamat Gan and the Biblical Zoo in Jerusalem for providing us with nestlings and eggs;and to the fishermen in the kibbutzim Hazorea, Maoz-Haim, Geva, Ma’agan-Mikhael,Baram, and Kefar Yehoshua for the fish supply during the study.

REFERENCES

Björklund, M. 1996. Similarity of growth among Great tits (Parus major) and Blue tits(P. caeruleus). Biol. J. Linn. Soc. 58: 343–355.

Bosque, C., Bosque, M.T. 1995. Nest predation as a selection factor in the evolution of

172 SHMUELI ET AL. Isr. J. Zool.

developmental rates in altricial birds. Am. Nat. 145: 234–260.Charnov, E.L. 1993. Life history invariants: some explorations of symmetry in evolutionary

ecology. Oxford Series in Ecology and Evolution. Oxford University Press, Oxford, 167 pp.Cooper, J. 1985. Biology of the Bank Cormorant, Part 2: Morphometrics, plumage, bare parts and

moult. Ostrich 56: 79–85.Crivelli, A.J., Nazirides, T., Catsadorakis, G., Hulea, D., Malakou, M., Marinov, M., Shogolev, I.

2000. Status and population development of Pygmy cormorant Phalacrocorax pygmeus breed-ing in the Palearctic. Medmaravis 5th Symposium on Marine Wildlife. Malta: EnvironmentProtection Department, pp. 49–60.

Dunn, E.H. 1975. Growth, body components and energy content of nestling double-crestedcormorants. Condor 77: 431–438.

Dunn, E.H., Brisbin, L.L.Jr. 1980. Age-specific changes in the major body components andcaloric values of herring gull chicks. Condor 82: 398–401.

Goutner, V., Papakostas, G., Economidis, P.S. 1997. Diet and growth of great cormorant(Phalacrocorax carbo) nestlings in a Mediterranean estuarine environment (Axios Delta,Greece). Isr. J. Zool. 43: 133–148.

Izhaki, I. 1994. Preliminary data on the importance of Israel for the conservation of the whitepelican Pelecanus onocrotalus L. Ostrich 65: 213–217.

Johnsgard, P.A. 1993. Cormorants, darters, and pelicans of the world. Smithonian InstitutionPress, Washington, 445 pp.

Kato, A., Watanuki, Y., Naito Y. 1998. Benthic and pelagic foraging of two Japanese Cormorants,determined by simultaneous recording of location and diving activity. J. Yamashina Inst.Ornithol. 30: 101–108.

Kato, A., Watanuki, Y., Shaughnessy, P., Le Maho, Y., Naito, Y. 1999. Intersexual differences inthe diving behaviour of foraging Subantarctic Cormorant (Phalacrocorax albiventer) andJapanese cormorant (P. filamentosus). Life Sci. 322: 557–562.

Kato, A., Watanuki, Y., Nishiumi, I., Kuroki, M., Shaughnessy, P., Naito, Y. 2000. Variation inforaging and parental behavior of king cormorant. Auk 117: 718–730.

Kirkham, I.R., Montevecchi, W.A. 1982. Growth and thermal development of northern gannets(Sula bassanus) in Atlantic Canada. Colonial Waterbirds 5: 66–72.

Koffijberg, K., Van Eerden, M.R. 1995. Sexual dimorphism in the cormorant Phalacrocoraxcarbo sinensis: possible implications for differences in structural size. Ardea 83: 37–46.

Koshelev, A., Chaban, B., Pokusa, R. 1997. Seasonal distribution and mortality of cormorant fromthe northern Azov Sea. Suppl. Ric. Biol. Selvaggina XXVI: 153–157.

Lack, D. 1968. Ecological adaptation for breeding in Birds. Methuen, London, 409 pp.Leisler, B., Winkler, H. 1985. Ecomorphology. In: Johnston, R.F., ed. Current Ornithology, Vol.

2. Plenum Press, New York, pp. 155–186.Montevecchi, W.A., Vaughan, R.B. 1989. The ontogeny of thermal independence in nestling

Gannets. Ornis Scand. 20: 161–168.Nemtzov, S.C. 2001. Summary of nesting success of the pygmy cormorant, Phalacrocorax

pygmeus in spring 2001 in Israel. Internal report of the Nature and Parks Authority (inHebrew).

O’Connor, R.J. 1984. The growth and development of birds. Wiley, New York, 315 pp.Ostnes, J.E., Jenssen, B.M., Bech, C. 2001. Growth and development of homeothermy in nestling

European shags (Phalacrocorax aristotelis). Auk 118: 983–995.Owens, I.P.F., Hartley, I.R. 1998. Sexual dimorphism in birds: why are there so many different

forms of dimorphism? Proc. R. Soc. London B 265: 397–407.

Vol. 49, 2003 DEVELOPMENTAL RATES AND MORPHOMETRICS OF CORMORANT 173

Pajkert, Z., Górski, W. 1996. Breeding ecology of great cormorant Phalacrocorax carbo sinensisin the Slowinski National Park (NW Poland). Cormorant Research Group Bulletin 2: 6–10.

Pinshow, B., Fedak, M.A., Battles, D.R., Schmidt-Nielsen, K. 1976. Energy expenditure forthermoregulation and locomotion in emperor penguins. Am. J. Physiol. 231: 903–912.

Platteeuw, M., Koffijberg, K., Dubbeldam, W. 1995. Growth of cormorant Phalacrocorax carbosinensis chicks in relation to brood size, age ranking and parental fishing effort. Ardea 83: 235–245.

Ricklefs, R.E., Miles, D.B. 1994. Ecological and evolutionary inferences from morphology: anecological perspective. In: Wainwright, P.C., Reilly, S.M., eds. Ecological morphology.University of Chicago Press, Chicago, pp. 13–41.

Ricklefs, R.E., Starck, J.M., Konarewski, M. 1998. Internal constraints on growth in birds. In:Starck, J.M., Ricklefs, R.E., eds. Avian growth and development. Oxford University Press,Oxford, pp. 266–287.

Royle, N.J., Hartley, I.R., Owens, I.P.F., Parker, G.A. 1999. Sibling competition and the evolutionof growth rates in birds. Proc. R. Soc. London B: 266: 923–932.

Schmidt-Nielsen, K. 1972. Locomotion: energy cost of swimming, flying and running. Science177: 222–226.

Shy, E., Frankenberg, E. 1995. The conflict between cormorants and fishery in Israel. CormorantResearch Group Bulletin 1: 45–46.

Shy, E., Geva, A. 1998. Final report of the great cormorants wintering in Israel and their damagesto fish ponds (1995–1998). Nature Authority and Ministry of Agriculture (in Hebrew).

Starck, M., Ricklefs, R.E. 1998a. Pattern of development: the altricial-precocial spectrum. In:Starck, J.M., Ricklefs, R.E., eds. Avian growth and development. Oxford University Press,Oxford, pp. 3–30.

Starck, M., Ricklefs, R.E. 1998b. Variation, constraint and phylogeny: comparative analysis ofvariation in growth. In: Starck, J.M., Ricklefs, R.E., eds. Avian growth and development.Oxford University Press, Oxford, pp. 247–260.

Starck, M., Ricklefs, R.E. 1998c. Avian growth rate data set. In: Starck, J.M., Ricklefs, R.E., eds.Avian Growth and Development. Oxford University Press, Oxford, pp. 381–423.

SYSTAT. 1998. SYSTAT for Windows, Statistics. Version 8.0 edition. SYSTAT, Evanston, IL.Volponi, S. 1999. Reproduction of a newly-established population of the great cormorant in north-

eastern Italy. Waterbirds 22: 263–273.Werschkul, D.F., Jackson J.A. 1979. Sibling competition and avian growth rates. Ibis 121:

97–102.