398 SHORT COMMUNICATIONS

LITERATURE CITED grationConf. for Hawk Migration Associationof North America. BENT, A. C. 1938. Life historiesof North American birds DORST,J. 1974. The life ofbirds. Vol. 1. Columbia Univ. of prey. U.S. Nat. Mus. Bull. 170. Press,New York. BOND, R. M. 1936. Speedand eyesightof a pigeonhawk. MEINERTZHAGEN,R. 1955. The speedand altitude ofbird Condor 38:85. flight (with notes on other animals). Ibis 97:81-l 17. COCHRAN,W. W. 1972. Long distancetracking of birds. NEWTON,I. 1985. Birds of prey-the raptors, p. 102- NASA SP 262~39-59. 121. In C. M. Penins and A.L.A. Middleton [eds.], COCHRAN,W. W. 1975. Following a migrating peregrine The encyclopediaof birds. Facts-On-File, New York. from Wisconsin to Mexico. Hawk Chalk 14:28-37. STORER,J. H. 1952. Bird aerodynamics. Sci. Am. 186: COCHRAN,W. W. 1985. Ocean migration of Peregrine 24-29. raicons:=_1.. ~~ IST- .I._me aaun-J--II male_--1. pelagic:.--1--:_” r-rot.n_-- r-nawnTT_...,.ivu-%I:

TheCondor 88:398-401 0 The CooperOmlthological Society I986

OBSERVATIONS ON THE BEHAVIOR AND ECOLOGY OF THE MARIANA CROW ’

DIANA F. TOMBACK~ Department of Zoology and Entomology, Colorado State University,Fort Collins, CO 80523

Key words: Mariana Crow; Mariana Islands; Guam: Guam’s endangeredavifauna is currently in preparation, Rota; endangeredavifauna; Corvidae. much of the anecdotal information available for these specieshas come primarily through the efforts of J. Mark The Mariana Crow (Corvus kubaryz?is found on the is- Jenkins (1983). lands of Guam and Rota in the Mariana Islands of Mi- I observed the Mariana Crow on Guam and Rota for cronesia.An attractive, fairly tame island inhabitant, the 15 days during July 1980. Although the study was brief, Mariana Crow is small with short, broad wings, a green the observations supplement the information in Jenkins glossto the black head plumage, and a blue glossto the (1983), presenting additional details on the ecology and black body plumage. Patches of gray feather basesin the behavior of the crow. body plumage and white feather basesin the neck region STUDY AREAS AND METHODS are often exposed. The bill and feet are black, eyes are brown, and nasal bristles are prominent. As in other cor- The Mariana Islands lie above the Equator, midway be- vids, the juveniles have short, flutTy head plumage. In tween Japan and New Guinea and about 2000 km east of addition, young Mariana Crows are disheveled in ap- the Philippines; Guam and Rota are situated about 14”N pearance;exposed feather baseson the breast show aslight and 145”E. Guam, a United States trust territory, is the patches, the black plumage has a brown cast, and flight largestand best known of the Mariana Islands. Rota, ad- feathers may appear worn at the tips. As suggestedby ministered by the Commonwealth of the Northern Mar- Baker (195 I), C. kubaryi may be closelyrelated to C. enca iana Islands, lies about 59 km northeast of Guam. For of the Malayan and Philippine regions. He further spec- both islands the year-round climate is humid, with day- ulatesthat “C. kubarvi is an isolated and modified suecies time temperaturesabout 30°C or higher. Details concem- of crow, which probably has been living at Guam and ing the geology, weather, flora, and fauna of Guam are Rota for a considerablelength of time.” Features which available in Stone (1970) Moore and McMakin (1979) differ from the candidate ancestral speciesinclude small and Eldredge(1983). Plant communities referred to in this size, slender bill, and dull coloration. Goodwin (1976) report are found on both Guam and Rota and have been believes, however, that aside from the geographicprox- described in detail by Moore and McMakin (1979) and imity of C. enca, any Corvusspecies could be ancestral. Jenkins (1983). The Mariana Crow is a little-known specieswhose pop- I observed the Mariana Crow on Guam from 4 to 11 ulations have suffereddeclines over the past 25 years. In and 25 to 27 July 1980, in limestone forest and coastal 1975, the crow was placed on the Trust Territory Endan- gered Specieslist; in 1984, it was placed on the U.S. Fed- eral EndangeredSpecies List alongwith the Bridled White- TABLE 1. Flock sizes observed for the Mariana Crow. eye (Zosteropsc. conspicillatus),Guam Flycatcher (Myi- agra freycineti), Micronesian Kingfisher (Halcyon c. cin- No. of times namomina), Vanikoro Swiftlet (Aerodramusvanikorensis Flock sire observed bartschii),Common Moorhen (Gallinula chloropusguamI>, and Guam Rail (Rallus owstoni; revised nomenclature, Guam 1 5 Pyle and Engbring 1985). Although a recovery plan for 2 5 4 1 5 2 Rota 1 1 ’ Received 19 November 1985. Final acceptance3 April : 104 1986. 4 3 2 Presentaddress: Department of Biology, University of 6 1 Colorado at Denver, Denver, CO 80202. SHORT COMMUNICATIONS 399

TABLE 2. Vegetation usedby Mariana Crows on Guam in forests. Occasionallythe flight pattern was like that of and Rota. woodpeckers,with wings held to the sides during a glide. For fast flying, the flappingwas of the pumping type typical No. of times of other crows. Crows flew above the forest canopy during Plant species observed most sightingson Guam. In contrast to the observations of J. Engbring, F. L. Ramsey, and V. Wildman (unpubl.), Guam Pandanus sp. 1 I observed crows flying above the forest canopy on Rota 1 Leucaena leucocephala as well, but primarily on the southwestside of the island. Cocosnucifera 2 Total 4 FAMILY GROUPS Rota Cocosnucifera 5 Several of the crow groups observed on Guam and Rota Scaevolataccada 4 consistedof two parents and one, two, or more offspring. Casuarina equisetifolia 3 On Rota, family groupsappeared to occupya defined home Hernandia nymphaelfolia rangealong the coastalstrand, one group was found in the Maytenus thompsonii : same 2-km stretch for three consecutive mornings. The Elaeocarpussphaericus juveniles were particularly tame and occasionallyperched Artocarpusmariannensis : no more than 1 m from me in the understory. As in other Pandanus sp. 2 corvid species,the juveniles must have a long period of Ficus prolixa 2 dependency, as several observed at close range did not Messerschmidtiaargentea 1 have rictal flanges. Neisospermaoppositifolia 1 Squalling, persistent food-begging calls were common Mangifera indica 1 throughoutthe coastalstrand on Rota and were occasion- ally heard on Guam. When most intense, the calls were Total 29 accompaniedby wing flutters and bill gaping. On Rota I observedthat the beggingjuveniles were fed intermittently by a parent, and as the juveniles swallowedthe food, they emitted a choking, yelping sound as do young of other strand on the northern third of the island, which is now crow species.In between feedings or when their parents the only area where crowscommonly occur(Jenkins 1983). did not respond, the juveniles often perched quietly or Field transportation and logistic support were provided moved alone through the dense Scaevola understory. I by the Division ofAquatic and Wildlife Resources(DAWR) observed one juvenile preen and then scratchits neck by on Guam. From 22 to 25 July 1980, I studied crows on lifting a leg over the wing. While juveniles traveled through Rota in more detail than I could on Guam, working pri- the understory, moving from perch to perch or searching marily in coastal strand association 1 to 3 km northeast the ground, they would encounter and manipulate a va- of the village of Songsong.Crow vocalizations were re- riety of objects, apparently not consuming them. They corded at close range with a Sony T-55 cassetterecorder snippedor ripped leavesfrom trees,hammered on branch- and cardioid microphone and later transferred to reel-to- es, and played with seeds,snail shells,twigs, live and dead reel tape. In the Department of Zoology and Entomology leaves. One juvenile was observed shreddingbark from a at Colorado State University, I made sonogramson a Kay twig by sliding the twig between the mandibles. Another ElemetricsSona-Graph, Model 7029A, for roughanalysis. juvenile placed a seed in a cup-like hollow, where two OBSERVATIONS large branchesjoined in a Hernandia tree, and removed and replaced the seed in the hollow at least twice. FLOCK SIZE AND FLIGHT I obtained flock size countson a total of 32 crow sightings FORAGING BEHAVIOR AND with an overall range of 1 to 6 crows per flock (Table 1). VEGETATION USE The mean flock sizeson Guam and Rota were 2.2 and 2.7 The Mariana Crow is clearly an omnivore, feeding both crows, respectively.Compared by t-test, the flock sizeson on plant and animal foods (Marshall 1949, Jenkins 1983). Guam and Rota did not differ significantly (t = -0.972, Crows commonly foragein the forest canopy, in the dense df= 30, P= 0.170). Jenkins(1983)reportsflocksonGuam understory, and on the forest floor. On Rota I observed typically of 2 to 5 individuals but mentions one flock of crows foraging on the fruit of Artocarpus mariannensis. 14 and another flock of 7, both in flight. J. Engbring, F. Crows alsosearched the leavesof Cocosnucijkra and Ficus L. Ramsey, and V. Wildman observed larger flocks on prolixa and the bark of Scaevola taccada, presumably average on Guam than on Rota. looking for insects or lizards. On one occasion, a crow Sightingsof crowsin flight were common on both Guam pulled vigorously at the dead leaves of a Pandanus sp., and Rota. The shapeof the crows’ wings appearedshorter which Jenkins(1983) has observedcrows eating. In Table and broader in comparisonwith American Crows (C. bra- 2, I list all speciesof plants that I observed crows using chyrhynchos),perhaps permitting more maneuverability on Guam and Rota for foraging as well as other activities

TABLE 3. Characteristicsof vocalizations of the Mariana Crow, based on rough sonagraphicanalysis.

Dominant freq. Syllableduration Vocalization Samplesize KHZ No. of syllables set

Adult Locational call 4 2-3, 5-7 l-3 0.35-0.46 Squalling locational call 1 1.8-3, 5-6.5 2 0.50 Alarm call 2 1.8-3,4-7 3-4 0.23-0.35 Juvenile Locational call 1.8-3,4.5-7.5 2-4 0.23-0.35 Hunger call 2-3, 5-7.5* l-3 0.31-0.77

* Strongharmonics, down-slurred syllables. 400 SHORT COMMUNICATIONS suchas vocalizing, playing, resting, and feeding young. Of the junction between two large branchesabout 8 m up in the 13 plant speciesused by crows, 11 were native and a large Hernandia nymphaeifolia, which bore abundant typical of coastal strand and limestone forest. The excep- fruit. Located near the dirt road parallel to the strand, the t&s were the introduced Leucaenaand Mangifera (man- nest tree was approximately 2.9 km northeastof Songsong go). Similar foraging habits and behavior were described and next to a small Chamorro cemetery about 20 m from for the endangeredHawaiian Crow (C. tropicus)by Sakai the beach. At an early stageof construction,the nest con- and Ralph (1980). sisted of a skeletal platform and ring of twigs, about 50 cm in diameter. Details from my field notes follow. At 0800 one crow flew to the nest carrying a long strand VOCALIZATIONS of Casuarina foliage, vocalizing both the locational calls The vocalizationsof adult crows heard on Guam and Rota and the squallinglocational calls. After placing the foliage were of four general types. (1) Locational calls, noted on on the nest platform, the crow perched in a nearby Casu- eight occasions,consisted of high pitched cawsor hi sounds arina for 5 min and then flew off. Bv 0820 the crow and given one to three times in succession.As the name in- its mate returned,vocalizing squalling-locational calls while dicates, locational calls were probably used to maintain flying to the nest tree. Both birds engagedin monolog calls contactamong individuals of a pair, family group,or flock. for severalminutes, while breakingoff and dropping leaves Usually two or more crows were involved; sometimesone and twigs and tapping branches in woodpecker fashion. crow respondedto a locational call with a locational call. One crow stood on the nest and cawed in an agitated The accompanyingbehaviors included perching,foraging, manner. It is possible that both were disturbed by our flying through or above the forest, and one crow leaving presence(two observers);such reaction to human distur- a nest using locational calls, followed shortly by its mate, bance has also been observed by Jenkins (pers. comm.). also usinglocational calls. (2) Squally locational calls were However, the crow at the nest abruptly flew into the Her- nasal caws with an aaa sound used by a nesting pair on nandia foliage above the nest, picked up a twig, and re- six different occasions.I heard the vocalization in the fol- turned to the nest. lowing contexts: given by a crow on the nest as another After placing the twig on the platform, the crow on the flew in, by both crows at the nest, and by a crow setting nest vocalized squalling locational calls while the other a twig in place on the nest (two occasions).This variation crow Aew off. The secondcrow returned at 0845, and both of the locational call may be used between mated crows. crowsexchanged squalling locational calls. Both crowsleft (3) The vocalizations under the categorymonolog include the vicinity of the nest and returned togetherat 0905. One a variety of quiet or excited guttural and squallingsounds crow silently entered the nest while the other perched,gave that were given for several consecutiveminutes by mem- squallinglocational calls, and hammered on a limb. The bers of a pair at the nest. While vocalizing, the crows crow on the nest flew into the Hernandia canopy, broke ripped nearby leaves from a branch and dropped them off a twig about 0.3 m long with its bill, returned to the from the tree or hammered on a limb. These calls may nest platform, and set the twig in place; the crow repeated serve some courtship or pair-bond function. Similar calls this behavior twice again. Pausing, it gave squalling lo- and playful behavior are described by Kilham (1985) for cational calls, which were answered by the other crow, American Crows. The “song” describedby Brown (1985) perched nearby. The crow on the nest entered the Her- for C. brachyrhynchosseems homologous in sound struc- nandia foliage again and found a fourth twig, but it then ture and context to the monolog in Mariana Crows. (4) flew off (at 09 18), twig in bill, vocalizing the high-pitched Alarm calls consisted of a series of sharp, rapid caws. I locational calls. The other crow followed immediately, heard this vocalization from two crowsperched in a Casu- vocalizing similarly. Since the crows did not return to the arina tree when they were startled by the sudden appear- nest by 1000, I continued my observationselsewhere. ance of several people. Measurements taken from sona- When I returned to the nest tree at 1115, one crow was gramsof three typesof adult vocalizationsare summarized on the nest and the other was perched in a nearby tree. I in Table 3. The sonagraphicpatterns are very similar to could not tell whether the crow on the nest was the same thoseof C. brachyrhynchos(e.g., Richards and Thompson individual observedthere earlier. However, the crow soon 1978, Brown 1985), but of higher frequencies. As de- moved into the Hernandia canopy, broke off a slender scribedby Brown (1985) for caws,the locational callsshow twig, and dropped it. Quickly, it then broke off a long, dominant frequencies(defined by loudness)and also har- sturdy branch, about 0.5 m long, and arranged it in the monic structure. nest platform. The other crow softly vocalized monolog The vocalizationsof juvenile crows are similar to those calls. At 1115 both crows flew off. Unfortunately, I had of adults. (1) Locational calls (heard on three occasions) to terminate observationsthat afternoon. were somewhat shorter in duration than those of adults Marshall (1949) suggestedthat Mariana Crows may have but given in similar contexts. (2) Monolog consistedof a a limited breeding season;he stated that the crows do not seriesof squalling,whining calls of varying inflection, giv- breed in May, June,or July. This suggestionwas supported en on three different occasionsby a juvenile alone in the bv Baker (195 1) who did not find crows in breedina con- understory or with other crows perched nearby. During dition from May to September. However, DAWN staff one observation,the juvenile simultaneouslypreened and observedfledglings with adults in May, June, September, scratched.(3) Hunger calls, heard on three occasions,con- and October, which indicated that the crows breed year- sisted of a series of insistent aaa sounds with a rising round (Jenkins 1983). The observationspresented here of inflection, often accompanied by wing flutters. The calls nest building on Rota in July substantiatethat possibility. increased in intensity until the young bird was fed by a Although I report that only one crow appearedto engage nearby adult. While gulping food, the juvenile made a in nest construction, DAWR staff have observed both gurglingsound. Table 3 summarizes measurementsfrom members of a pair carry nest material (Jenkins 1983; R. sonagramsfor two of the juvenile vocalizations. The hun- E. Beck, pers. comm.). Additional observationsat an ac- ger calls showed a pronounced harmonic structure;on 3 tive crow’s nest confirm that both mates incubate eggsand of 5 sonagrams,dominant frequencies were not readily feed the vounz (Jenkins 1983). Kilham (1984) reoorted apparent. that both-members of pairs of American‘Crows engaged in nest construction;however, only the female incubated NEST CONSTRUCTION AND NESTING the eggsand brooded the nestlings. BEHAVIOR I am grateful to the Division of Aquatic and Wildlife On 25 July, I observeda pair of crows constructinga nest Resourcesof the Guam Department of Agriculture for in coastalstrand vegetation on Rota. The nest site was at encouragingthis study, providing field vehicles, and ar- SHORT COMMUNICATIONS 401 rangingaccess to military installations.The guidance,sup- GOODWIN,D. 1976. Crows of the world. Comstock Pub- port, and assistanceof Anne F. Maben and Robert D. lishing Associates,Ithaca, NY. Anderson were particularly appreciated.I am also grateful JENKINS,J. M. 1983. The native forest birds of Guam. to the Department of Agriculture of the Mariana District Ornithological Monograph No. 3 1. of the Trust Territory of the Pacific Islands for providing KILHAM, L. 1984. Cooperative breeding of American a field vehicle and assistanceon Rota. I thank C. John crows. J. Field Omithol. 55:349-356. Ralph for preliminary information on Guam and Rota KILHAM, L. 1985. Behavior of American Crows in the and CharlesT. Collins for his crow observationsand com- early part of the breeding cycle. Fl. Field Nat. 13:25- panionship in the field. Helpful comments on an earlier 31. draft of this paper were provided by Robert Beck, John MARSHALL,J. T., JR. 1949. The endemic avifauna of Engbring,Mark Jenkins,Gary Michael, and Julie Savidge. Saipan, Tinian, Guam, and Palau. Condor 51:200- The study of the Mariana Crow was supportedby a grant 221. from the American Philosophical Society. MOORE,P. H., AND P. D. MCMAKIN. 1979. Plants of Guam. Univ. of Guam Collegeof Agriculture and Life LITERATURE CITED SciencesCooperative Extension Service, Mangilao. AND BAKER, R. H. 1951. The avifauna of Micronesia, its or- PYLE,P., J. ENGBRING.1985. Checklist of birds of igin, evolution, and distribution. Univ. Kans. Publ. Micronesia. ‘Elepaio 46:57-68. Mus. Nat. Hist. 3: l-359. RICHARDS.D. B.. AND N. S. THOMPSON.1978. Critical BROWN,E. D. 1985. The role of songand vocal imitation prope’tiies of the assemblycall of the common Amer- among Common Crows (C. bruchyrhynchos).Z. Tier- ican Crow. Behaviour 64: 184-203. psychol. 68: 115-l 36. SAKAI. H. F.. AND C. J. RALPH. 1980. Observations on ELDREDGE,L. G. 1983. Summary of environmental and tde Hawaiian Crow in south Kona, Hawaii. ‘Elepaio fishinginformation on Guam and the Commonwealth 40:133-138. of the Northern Mariana Islands. Historical back- STONE,B. C. 1970. The flora of Guam. Micronesica 6: ground, description of the islands, and review of the l-659. climate, oceanography and submarine topography. NOAA Technical Memorandum NMFS, No. 40.

ERRATUM In our May 1986 issue, the authors’ names for the article “Costs of adoption in Western Gulls” (Condor 88:253-256) appeared incorrectly. The authors are Harry R. Carter and Larry B. Spear.