Pacific is found on Christmas Island (15,000,000 at highest 

 count), and 3,000,000 have been recorded on Starbuck (Perry, 

 1980). 



Distribution and Habitat Preference : To date, 19 colonies 

 from 10 islets are known for the years 1965, 1988, 1989, and 

 1990 (Fig. 11). In September 1988, we found three colonies, 

 two on the northern half of Long and one on Bo' sun Bird Islet; 

 all fit the general habitat description in Clapp & Sibley ( 1 97 1 a). 

 Colony A, nearly square, was 210 m on a side. Eggs were 

 placed under a savannah-type Tournefortia scrub, from 1-4 m 

 tall with approximately 60% canopy cover (Subchapter 1.1, 

 PI. 59). The substrate was coral rubble mixed with sand, 

 covered by Heliotropium (5%), Portulaca (1%), Laportea 

 (< 1 % ), and Lepturus (< 1 %), typical of old interisland channels. 

 Colony 1 was located in a broad sandy corridor with two large 

 "groves" of Tournefortia. The northern subpopulation extended 

 116 m along the windward beach, but 248 m along the lagoon. 

 The southern subpopulation began 28 m further south along the 

 beach, fronted the seaward reef for 86 m, and was shaped like 

 a blunt triangle, its apex pointing toward the lagoon. Most 

 chicks were under Tournefortia, which consisted of shrubs 

 2-4 m high with 80% canopy cover. The substrate was also 

 older beach sands mixed with coral rubble and covered with 

 Portulaca (40% cover), Lepturus (<5%), and Heliotropium 

 (<5%). The Bo'sun Bird colony, a rough oval approximately 

 55 m wide by 70 m long, was under 2-3 m high Tournefortia 

 with 75% cover, on coral rubble/sand sparsely carpeted with 

 Portulaca and Heliotropium. 



Numbers : Populations were determined by measuring 

 colony dimensions, then counting eggs and/or chicks in 9-m 2 

 sample plots located at random points along a compass line. 

 Because juveniles moved as we approached, they were counted 

 6 m ahead of us in estimated 3 x 6 m plots. The Colony I 

 subcolonies (North, South) were treated separately. 



Colony size (rounded) in 1988 ranged from 

 1 27,000 ± 30,000 "nests" (Colony A) to 1,500 ±750 new eggs 

 on Bo'sun Bird Islet (Table 6). There were an additional 

 6,900 + 1 ,600 nearly-fledged chicks in the Bo'sun Bird colony, 

 resulting from eggs laid three months earlier. 



The total number of eggs and chicks was 188,000 ±21%. 

 Actual numbers of adults are difficult to estimate but in other 

 studies have exceeded the number of eggs and young by factors 

 of more than two because innumerable eggs and chicks were 

 lost, colonies often overlapped, and many nonbreeding adults 

 joined the prebreeding swarms or associated with breeding 

 birds. Schreiber & Ashmole (1970), relying on POBSP data 

 from Johnston Atoll (north-central Pacific), estimated that four 

 adults were present for each egg laid. Pacific Ocean Biological 

 Survey Program data from Johnston (Amerson & Shelton, 

 1976) indicated that about 600,000 adults were present in a 

 colony with 105,000 eggs, or approximately 5.7 adults/egg. If 

 we assume that real numbers of terns in our colonies lay 

 midway between 4 and 5.7 times the number of eggs and 

 chicks, then the number of sooty terns using Caroline Atoll 

 would have ranged between 720,000 and 1,100.000 birds 

 (91 1.800±21%). This is twice the estimate provided by Clapp 

 & Sibley (1971a), even though we found fewer colonies. 



In March 1990, laying was just beginning in two colonies 

 on Long Island, (625 x 1 50-3 1 5 m wide and 1 80 x 1 60 m wide). 

 Enormous numbers of birds, both on the ground in densities up 

 to 9 or 1 pairs/nr and in the air, made it impossible to calculate 

 a reasonable population figure. According to Anne Falconer, 

 these two colonies were very successful. Similarly, counting 

 was difficult in May 1990 when six large prebreeding swirls 

 hovered like huge clouds of gnats over discrete islets and islet 

 groups (Fig. 1 1). Our 1988 estimate of approximately one 

 million birds is probably a conservative count for the atoll as a 

 whole on an annual basis. 



Phenology : The incubation period in sooty terns is about 

 4 weeks (Dinsmore, 1972). Young fledged 7-8 weeks after 

 hatching, although fledgling ages, dependent upon food supply 

 (Schreiber & Ashmole, 1970), are highly variable. 



Four separate sooty tern colonies had been started over the 

 12-week period prior to our study in 1988 (Table 6). On Bo'sun 

 Bird Islet a new wave of laying was just beginning in an open 

 area immediately southwest of most of the colony, while nearly 

 fledged chicks scurried about beneath the Tournefortia. 

 Undoubtedly many young had already fledged, so many more 

 eggs would have been laid in early July by this colony than 

 indicated (Fig. 12). The two colonies on Long were established 

 at different times: the short-tailed juveniles in Colony 1 

 preceded the large number of eggs, hatching eggs, and downy 

 chicks of Colony A by 3-4 weeks. 



The July-September laying period on Caroline in 1988 is 

 very different from the bimodal breeding (May-June, 

 December-January) reported from Christmas Island, Pacific 

 Ocean (Schreiber & Ashmole, 1 970), and the May laying dates 

 noted for Caroline by the POBSP in 1965. Additional data 

 (Anne Falconer, personal communication) indicate that sooty 

 terns may lay any time (Fig. 11), certainly January through 

 September (1988 to 1990). Severe storms, which destroyed 

 large Long Island colonies in February 1990, were perhaps 

 responsible for reinitiating breeding activities on the leeward 

 side of the atoll within the next few months. A great deal more 

 research will be needed on Caroline before the breeding seasons 

 for this species are fully understood. 



Brown Noddy (Anous stolidus) (Fig. 13) 



This tern, primarily a tree nester, is widely distributed 

 throughout the warm oceans of the world. It is abundant in the 

 Line and Phoenix Groups, with an estimated total population 

 exceeding 40,000 birds. Brown noddies are most abundant on 

 Palmyra Island (10,000 birds). 



Distribution and Habitat Preference : The brown noddy is 

 second only to the white tern in the number of motus (28) upon 

 which it is known to breed (Fig. 13). It utilized the smallest 

 (Noddy Rock, 0.02 ha) and largest (South, 104.41 ha) motus, 

 nesting upon coral rubble and in plant communities ranging 

 from the simplest herb mats to Tournefortia. Pisonia, Cordia, 

 Cocos, and the mixed anthropogenic forests of South and 

 Nake. Most pairs were well dispersed, nesting from the outer 

 edges of Tournefortia to the central, inner branches of Pisonia, 

 and from the ground to the crowns of 25-m Cocos. When 

 nesting sympatrically with black noddies in Pisonia. the brown 



145 



