The Condor 92:772-775 0 The CooperOrnithological Society 1990

EGG SIZE AND LAYING ORDER OF SNOWY EGRETS, GREAT EGRETS, AND BLACK-CROWNED NIGHT- ’

THOMAS W. CUSTER U.S. and Wildlife Service,Patuxent Wildlfe ResearchCenter, Gulf CoastResearch Station, P.O. Box 2506 Victoria, TX 77902 PETERC. FREDERICK Department of Wildlfe and Range Science,University of Florida, 118 Newins-Ziegler Hall, Gainesville,FL 3261 I

Key words: Black-crownedNight-; Nyctico- nests were visited daily during the incubation period. rax ;Great Egret; Casmerodiusalbus; Snowy Nests were marked with flags, and eggswere individ- Egret: Egretta thula; size. ually marked with a felt-tipped permanent ink pen when first observed. Laying order was determined for The nesting biology of the family Ardeidae (bitterns. 179 in two- and three-egg clutches by hatching herons, and egrets) has been intensively studied (e.g.; order (n = 10 eggsin two-eggclutches, 92 eggsin three- Owen 1960. Milstein et al. 1970. Werschkul 1979). but egg clutches)and order of appearance(n = 22 eggsin egg size in ‘relation to laying order has not received two-egg clutches, 55 eggsin three-eggclutches). attention. The last egg laid in gull and tern clutchesis In Texas, the length and width of each egg(n = 179) generallysmaller than precedingeggs (e.g., Parsons 1970, were measuredat least twice with a caliper to 0.0 1 mm Nisbet 1978). The relative size of the final eaa in a and the values averaged.If the measurementsdiffered clutch decreaseswith increasedbody size among by more than 0.2 mm, one or more additional mea- species and this relationship may be correlated with surementswere taken and either one discarded(n = 7) an increasedbrood-reduction strategy(Slagsvold et al. or an average calculated (n = 14). Egg volume was 1984). Relative egg size could be an important com- determined in the field using the laboratory technique ponent to brood reduction, becauseegg size can affect of subtractingthe mass of an egg suspendedin water subsequentsurvival of young (Parsons 1970, Nisbet from its mass in air (Evans 1969, Hoyt 1979). A bat- 1978, Lundberg and V&islnen 1979). tery-powered electronic balance accurateto 0.1 g was Asynchronoushatching can facilitate size differences used to weigh eggs.Each egg’s volume was determined among nestlings and assist in brood reduction (Lack twice on separatedays. If the estimate of volume dif- 1954, Ricklefs 1965). In herons and egrets, asynchro- fered by more than 0.1 ml, a third measurement was nous hatching results in senior broodmates who are taken and either one discarded (n = 3) or an average superiorto their youngersiblings in both food handling calculated(n = 6). and aggressiveinteractions (Fujioka 1985, Inoue 1985, Nesting Great Egretsalso were studied from 13 Feb- Mock and Parker 1986). ruary to 2 May 1987 at a large mixed-species colonv Our objective was to describeegg size in relation to located in the central Evergladesmarshes near Andy- laying order for Great Egrets (Casmerodiusalbus), town. Florida (26”ll’N. 80”3 1’Wl. This colonv was SnowyEgrets (Egretta thula), and Black-crownedNight- visited every 4-6 days during incubation and hatching Herons (Nycticorax nycticorax) in a southern Texas periods. Eggswere numbered with indelible ink. Laying colony and Great Egretsin a southernFlorida colony. order was determined by order of appearance(n = 28) Basedon egg-sizepatterns in other colonial waterbirds and hatching order (n = 18). and the occurrence of brood reduction in egrets and Length and width of Great Egret eggsin Florida were herons (see above), we predicted that the final egglaid in a clutch would be smaller than those laid earlier. METHODS TABLE 1. Frequency distribution in clutch size of Nesting Snowy Egrets,Great Egrets,and Black-crowned , Great Egret, and Black-crowned Night- Night-Herons were studied on a dredge-materialisland Heron nests in Texas and Florida. in Lavaca Bay near Port Lavaca, Texas (28”36’N, 96”34’W, colony 609- 12 1, Texas Colonial Waterbird Number of eggs Society 1982) from 14 April to 8 July 1988. The colony Species Location 2 3 4 5 was visited every 2-3 davs durina the incubation ve- riod and every day during the hatching period, some Snowy Egret TX 9 35 2 0 Great Egret TX 11 33 2 1 Great Egret FL 11 8 0 0 Black-crowned ’ Received 11 October 1989. Final acceptance 6 Night-Heron TX 9 42 1 0 March 1990. SHORT COMMUNICATIONS 113

TABLE 2. Egg volume, length, and width in two- and three-eggclutches of Snowy Egrets, Great Egrets, and Black-crowned Night-Herons.

Volume (ml) Length (mm) Width (mm) Clutch wh size Species (location) order n K SD x SD R SD

2 eggs Snowy Egret A 22.0 A’ 2.4 40.9 A 2.5 32.4 A 1.0 (Texas) B 21.5 A 1.1 40.8 A 2.8 31.9A 1.5 Great Egret A 46.6 A 3.1 55.5 A 2.4 40.4 A 1.3 (Texas) B : 46.4 A 2.2 56.6 A 3.1 40.3 A 1.2 Great Egret A 11 45.4 A 2.4 55.6 A 1.5 40.1 A 0.8 (Florida) B 11 43.6 B 3.6 54.9 A 2.6 39.5 A 1.1 Black-crowned A 5 35.3 A 4.2 50.5 A 2.8 31.0 A 1.6 Night-Heron B 5 33.4 B 3.5 50.3 A 1.4 36.1 B 1.5 (Texas) 3 eggs Snowy Egret A 12 22.0 A 1.1 42.8 A 1.3 31.8 A 0.9 (Texas) B 12 21.5 AB 1.4 42.1 AB 1.5 31.1 A 0.9 C 12 20.9 B 1.1 41.1 B 1.1 31.3 A 0.9 Great Egret A 20 45.6 A 3.9 55.3 A 2.1 40.2 A 1.3 (Texas) B 20 46.2 A 4.1 55.1 A 1.8 40.3 A 1.4 C 20 43.5 B 4.2 55.1 A 2.1 39.2 B 1.4 Great Egret A 8 45.0AB 4.1 55.2 A 3.2 40.0 AB 1.2 (Florida) 55.9 A 3.2 40.8 A 0.9 : : 41.243.1 AB 2.52.1 53.9 A 1.9 39.6 B 1.0 Black-crowned A 11 31.2 A 2.8 52.6 A 2.5 31.3 A 1.2 Night-Heron 52.4 A 2.0 31.2 A 1.4 (Texas) : 11 36.935.0 AB 2.93.0 51.8A 2.0 36.4 B 1.5

’ Means between or among laying orders by clutch size and species not sharing the same letter are significantly different from one another (two- way ANOVA, nest x laying order, P < 0.05). measured with a caliper to 0.5 mm. Egg volume was accounted for 68.7%, 79.2%, and 78.6% of the total estimatedby the equation(Hoyt 1919),volume = 0.509 variation (R*) in egg volume of Snowy Egrets, Texas x length (cm) x width* (cm’). The constant 0.509 was Great Egrets, and Black-crowned Night-Herons, re- obtained from Texas Great Egret eggs(T. W. Custer, spectively(one-way ANOVA by nest, P < 0.01); laying unpubl. observ.) and is similar to that found earlier for order only accountedfor 11.3%. 1.9%. and 1 1.1%. re- Florida Great Egret eggs(0.508, Loftin and Bowman spectively (one-way ANOVA by laying order, Snowy 1918). Egret:P = 0.138; Great Egret:P = 0.096; Black-crowned Relative size of the final egg laid was calculated as Night-Heron: P = 0.60). For Florida Great Egrets, the percent deviation from mean volume of all eggsin the interclutchdifferences for three-eggclutches accounted clutch (Slagsvold et al. 1984). Analysis of variance for 40.3% of the variation (P = 0.223) and laying order (ANOVA) techniques were used to compare means. only accountedfor 23.8% (P = 0.057). Means were separated using Bonferroni mean sepa- Eggvolume, length, and width were not significantly ration tests (Neter and Wasserman 1914). different between A and B eggsin two-egg clutchesof Snowy Egrets and Texas Great Egrets (Table 2). In RESULTS contrast, the B egg in two-egg Black-crowned Night- Eggsgenerally hatched in the same order that they were Heron and Florida Great Egret clutches had signifi- laid. For nests where one or more eggswere of known cantly smaller volume than the A egg;width also was laying order, hatching order was the same in Texas in significantly smaller in the B than A egg of Black- 29 of 30 cases(Snowy Egrets:eight of eight nests;Great crowned Night-Herons. Egrets: 12 of 13 nests; Black-crowned Night-Herons: The C eggin three-eggclutches of Texas Great Egrets nine of nine nests). In one Great Egret nest the A egg and Black-crowned Night-Herons was significantly (first egg laid) hatched secondin a three-eggclutch. In smaller in volume and width than the A and B eggs: Florida, hatching order was identical to laying order the A and B eggs,however, were not different from one in all eggsof 14 Great Egret nests. another (Table 2). The C eaa for Florida Great Earets The modal clutch size was three eggsfor all three was significantlysmaller involume and width than the speciesin Texas and two eggsfor Great Egretsin Flor- B egg;however, there were no other significantdiffer- ida; only a few clutches had four or more eggs(Table encesamong laying order. The C eggfor Snowy Egrets 1). Thus, the analysis is concentrated on three-egg was significantlysmaller in volume than the A ea, but clutchesand secondarilyon two-egg clutches. there were no differencesin width among layingorder Most of the size variation in the egret and heron eggs (P = 0.057). Egg length was not significantly different arose from interclutch rather than intraclutch differ- among laying order for Texas Great Egrets(P = 0.454) ences. For three-egg clutches, interclutch differences Florida Great Egrets (P = 0.8 13), and Black-crowned 774 SHORT COMMUNICATIONS

Night-Herons (P = 0.300); for the Snowy Egret, how- becauseegg size is negatively correlated to the length ever, the C eggwas significantlyshorter than the A egg. of incubation, small final eggs may be an adaptation Mean percent deviation in volume of the final egg in to reduce the degree of hatching asynchrony (Parsons two- and three-e= clutcheswas - 1.2% and - 2.6% for 1972). Snowy Egrets, ?J.2% and -3.5% foir Texas Great mcI ne- --w”rK --‘m m‘- “oridal-1’ was supportedby a grant from Egrets, -2.0% and -4.4% for Florida Great Egrets, the U.S. Army Ccorps of Engineers.We thank R. Will and -2.6% and -3.8% for Black-crownedNight-Her- Roach, Ki rke ’ A.A Ring, and Patricia S. McDonald for ons. field assistante in Texas; Reed Bowman and Susan Some individual C eggswere the largest or second Fitzgerald for field assistancein Florida; R. Will Roach largestin three-eggclutches. One C eggwas the largest for develo ping the method to measure egg volume in egg in the clutch for each species and location. For the field alnd for statistical assistance:Clementine M. Snowy Egrets, Texas Great Egrets, and Florida Great Glenn for l?___r..rl-_^_ n c_.._f- -- > .l__^_K- _--_ _--_--r~..-l.. typing the manuscript; Douglas Mock for cgit;rs, Llllt-t: L, l”Ul L, all” LllItx L eggs, respecuveiy, identifying critical referencesand reviewing the manu- were the second largest eggsin the clutches. scrint: and Lawrence Blus. Christine M. Bunck. Kirke DISCUSSION A. I&g, Christine A. Mitchell, R. D. Morris,’ D. N. Nettleship, and Jeffrey Spendelow for reviewing the This study and an earlier one support the use of hatch- manuscript. ing order to predict laying order in egretsand herons. In this study, hatching order and laying order were LITERATURE CITED identical in 43 of 44 egret and heron clutches;the A and B egg hatched in reverse order in one three-egg BIRKHEAD,T. R., AND D. N. NETTLESHIP.1982. The clutch. Hatching order and laying order were also iden- adaptive significanceof eggsize and laying date in tical in nine of 10 Little Egret (Egret& gurzetta) clutch- Thick-billed Murres Uria lomvia. Ecoloav_. 63:300- es (total 47 eggs);an A and B eggalso hatchedin reverse 306. order in one clutch (Inoue 1985). Coutso~, J. C. 1963. Egg size and shape in the Kit- Our resultsdemonstrate that the last egglaid in egret tiwake (Rissa tridactvla) and their use in estimat- and heron clutchesis generally smaller than other eggs ing age‘ compositions of populations. Proc. Zool. in the clutch, a pattern observed in other colonial wa- Sot. Lond. 140:21l-227. terbirds (e.g., Parsons1970, Nisbet 1978). In this study COULSON, J. C., G. R. Porrs, ANDJ. HOROBIN. 1969. the C egg in three-eggclutches of Snowy Egrets,Great Variation in eggsof the Shag(Phalacrocorax aris- Egrets, and Black-crowned Night-Herons was smaller totelis). Auk 86~232-245. in volume than either or both the A or B egg. The B EVANS,R. M. 1969. Specificgravity of White Pelican eggin two-eggclutches of Black-crownedNight-Herons eggs.Auk 86:560-561. and Florida Great Egrets, but not Snowy Egrets and FUJIOKA,M. 1985. Sibling competition and siblicide Texas Great Egrets, was also smaller than the A egg. in asynchronously-hatchingbroods of the Cattle Our data supportthe hypothesisthat small final eggs Egret Bubulcusibis. Anim. Behav. 33:1228-1242. are associatedwith brood reduction and that within HOYT, D. F. 1979. Practical methods of estimating speciesthere is a negative correlation between relative volume and fresh weight of bird eggs.Auk 96:73- size of the final egg laid and clutch size (Slagsvoldet 77. al. 1984). It is argued that if the clutch is increasedby INOUE,Y. 1985. The processof asynchronoushatch- one egg, it is important that this egg is small so that ing and sibling competition in the Little Egret the brood reduction will operate effectively (Slagsvold Egretta garzetta. Colonial Waterbirds 8: 1-12. et al. 1984). Brood reduction through asynchronous LACK, D. 1954. The natural regulation of hatching of eggshas been documented in egrets and numbers. Clarendon Press, Oxford. herons (Owen 1960, Werschkul 1979, Fujioka 1985, Lomnr, R. W., AND R. D. BOWMAN. 1978. A device Inoue 1985, Mock and Parker 1986). In our study, the for measuringegg volume. Auk 95: 190-192. last egg laid in egret and heron clutches was smaller LUNDBERG,C. A., AND R. A. VKISXNEX 1979. Se- than earlier eggs.In addition, for all three speciesin lective correlation of eggsize with chick mortality our study, the percent deviation in volume of the last in the Black-headedGull (Larus ridibundus).Con- egglaid from the mean volume of all eggsin the clutch dor 81:146-156. was greater in three- than two-egg clutches. MILSTEM,P. LE S., I. Pmm-r, AND A. A. BELL. 1970. The importance of egg size in egrets and herons is The breedina cvcle of the Grev Heron. Ardea 58: unknown, but based on other studies of colonial wa- 171-257. - . terbirds deserves further attention. Chick survival is MOCK, D. W., ANDG. A. PARKER. 1986. Advantages positively correlated with egg size in gulls and terns and disadvantagesof egret and heron brood re- (Parsons 1970, Nisbet 1978, Lundberg and Vaisiinen duction. Evolution 40:459+70. 1979), and although female age and time of laying NETER,J., ANDW. W~ss~at+s~~.1974. Applied linear influence eggsize (Coulson 1963, Coulson et al. 1969), statisticalmodels. Richard D. Irwin, Homewood, egg exchangeexperiments demonstrate a relationship IL. between egg size and survival independent of parental NISBET,I.C.T. 1978. Dependence of fledging success care (Parsons 1975, Nisbet 1978). Females may be on egg-size, parental performance and egg-com- nutrient limited and produce a small final egg early position among Common and Roseateterns, Ster- rather than delay laying further to produce a larger egg na hirundo and S. dougallii. Ibis 120:207-2 15. (Birkhead and Nettleship 1982). On the other hand, OWEN,D. F. 1960. The nesting successof the heron SHORT COMMUNICATIONS 775

Ardea cinerea in relation to the availability offood. SLAGSVOLD,T., J. SANDVIK, G. ROFSTAD,0. LOR- Proc. Zool. Sot. Lond. 133:597-617. ENSTEN,AND M. HUSBY. 1984. On the adaptive PARSONS,J. 1970. Relationship between eggsize and value of intraclutch egg-size variation in . post-hatchingchick mortality in the Herring Gull Auk 101:685-697. (Larus argentatus). Nature 228:1221-1222. TEXASCOLONIAL WATERBIRD Socm. 1982. An at- PARSONS,J. 1972. Egg size, laying date and incuba- las and censusof Texas waterbird colonies 1973- tion period in the Herring Gull. Ibis 114:536-541. 1980. CaesarKleberg Wildlife ResearchInstitute, PARSONS,J. 1975. Asynchronoushatching and chick Texas A&I Univ., Kingsville, TX. mortality in the Herring Gull Larus argentatus. WERSCHKUL,D. F. 1979. Nestling mortality and the Ibis 117:517-520. adaptive significanceof early locomotion in the RKKLEFS,R. E. 1965. Brood reduction in the Curve- Little Blue Heron. Auk 96: 116-l 30. billed Thrasher. Condor 67:505-510.

The Condor 925’15-776 0 The CooperOmitholo&al Society1990

INFANTICIDE IN THE EURASIAN DIPPER’

SONJA I. YOERG* Department of Psychology, University of Massachusetts, Amherst, MA 01003

Key words: Eurasian Dipper; infanticide: polygyny; area: in addition to helping to provision the brood in sexual selection; brood reduction. question, he was building a nest with another female, and helping a third female feed five nestlingsdue to Infanticide has been reported in several avian species fledge in 2 days. (see Mock 1984, for a review). Direct observationsof At 09: 10 the female was brooding in the nest. The infanticide among individually marked are, male from the downstream territory flew upstream however, scarce. Here I describe one definite case of landing 5 m downstream of the nest. This male was infanticide by a male Eurasian Dipper (Cinclus cin- also polygynous.Both of his females were incubating clus), and three suspectedcases. eggs.For approximately 1 min he engagedin a solic- Dippers are passerinesclosely associatedwith fast- iting display, dipping deeply with his wings low and flowing rivers in which they forage for invertebrates. neck outstretched. He then landed directly below the Domed nests, made primarily of moss, are always lo- nest, where he called. (When the nestlings are small cated over water. Linear territories along rivers are and the female likely to be brooding, a provisioning defended vigorously by pairs during the breeding sea- male will often call below the nest before attempting son, and maintained less assiduouslyby individuals to deliver food.) After calling, the intruding male flew the rest of the year. Though traditionally considereda up and perched on the threshold of the nest opening. monogamousspecies, polygyny has been reported (e.g., The female immediately chased the intruder down- Galbraith 1979). Indeed, within my study area half of stream. the males were polygynous(Yoerg et al., unpubl.). He returned 2 min later, before the female reap- Thirteen contiguousterritories along a tributary of peared, and flew directly to an old dipper nest adjacent the river Wye in mid-Wales comprised the study area. to the one in use. After looking inside, he flew down, All adults (13 females and eight males) were individ- then up to the active nest out of which he pulled a ually color-marked with leg bands during February and nestling.He descendedto the river edgebelow the nest March of 1989, and were observed from February and dropped the nestlingin the water. Flying up to the through June in a study of foraging and breeding be- nest again, he graspedanother nestlingwith his bill and havior. Observations were made from inside a hide. drowned it in the same manner. As he was poised at On 28 April 1989 I was 7 m from a nest where adults the threshold of the nest, presumablyto retrieve a third were provisioningyoung. The five nestlingswere 5 days nestling,the resident male appearedbelow the nest and old, dippers typically fledge at 23 days of age. The gave chase downstream. The female returned from resident male was the only trigamousmale in my study downstream 15 set later and immediately entered the nest to brood. Nine minutes later the resident male returned to a site near the nest where he habitually perched, presumably to guard. He flew downstream I Received 8 November 1989. Final acceptance28 again after 10 min. February 1990. At 11:15 the same morning the intruder male ap- 2 Present address: Llysdinam Field Centre, New- proachedthe nest again. The female, who was foraging bridge-on-Wye, Llandrindod Wells, Powys, Wales LDl 10 m downstream, gave chase,joined by the resident 6NB, United Kingdom. male as they passedhim upstreamof the nest. All three