ORNITOWGIA NEOTROPICAL 4: 43-54, 1993@ The Neotropical Ornithological Society.

NATURAL RANGE EXPANSION ANO LOCALEXTIRPATION OF AN EXOTIC PSITTACINE -AN UNSUCCESSFUL

COLONIZATION ATTEMPT

James W. Wiley.

Patuxent Wildlife Research Center, U.S. Fish and Wildlife Service, Laurel, Maryland 20708, U.S.A

Resumen. El Perico Cara Sucia Aratinga pertinax es oriundo de Panamá, norte de Suramérica e islas al norte dela costa venezolana. Se cree que este perico fue introducido a Santo Tomás, Islas Virgenes, desde Curazao antesde los mediados del siglo XIX, aunque no existen registros para indicar cuando y como esto pudo haber ocurrido.El perico no se había reportado para la isla de Puerto Rico, Vieques o Culebra antes de 1975. Sin embargo, enese año A. pertinax sostuvo una aparente expansión natural en su distribución de Santo Tomás a cada una de estaslocalidades. Desde su descubrimiento en abril de 1975 las 5 aves de la población del este de Puerto Rico anidaronunas 5 veces y produjeron unos 11 pichones que volaron. La población creció a un máximo de 10 pájaros en juniode 1979 pero para julio de 1980 sólo quedaba un pájaro y la población ya no existia al comienw de lo que hubierasido la época reproductiva de 1982. La poblaciones en las islas de Culebra y Vieques aparentemente fueronextirpadas antes de 1976. Las pérdidas en la población de! este de Puerto Rico fueron bajas durante los primeroscuatro años de la colonización, cuando la reproducción fue adecuada y la mortandad baja. En contraste, entre 1978y el tiempo de la desaparición de la población en 1982, la extirpación esperada fue de 0.999. Se describen el usodel hábitat, comportamiento general, biologia reproductiva, competidores y depredadores de la población del estede Puerto Rico. Se discuten implicaciones biogeográficas, evolutivas y conservacionistas.

Abstract. The Brown-throated Parakeet Aratinga pertinax is native to Panama, northern South America, and theislands off the northern coast of Venezuela. It is widely believed that the parakeet was introduced to St. Thomas,Virgin Islands, from Cura~ao before the mid-19th century, although no records exist as to when and how thismay have occurred. The parakeet was not recorded from Vieques or Culebra islands or mainland Puerto Rico be-fore 1975. However, in that year A. pertinax underwent an apparent natural range expansion from St. Thomasinto each of these sites. The 5 birds in the eastern Puerto Rico population made 5 breeding attempts and f!edged11 young since their discovery in April1975. That population grew to a maximum size of 10 birds in June 1979.However there was only 1 bird left by July 1980, and the population no longer existed at the beginning of whatwould have been the 1982 breeding season. The populations on Culebra and Vieques islands were apparently extir-pated by 1976. Losses in the eastern Puerto Rico population were low during the first four years of colonization,when reproduction was adequate and the mortality was low. However, between 1978 and the time of populationdisappearance in 1982, the expected extirpation was 0.999. Habitat use, general behavior, breeding biology, compet-itors and predators of the colonizing population in eastern Puerto Rico are described. Biogeographic, evolutionary,and conservation implications of these observations are discussed. Accepted 7 September 1992.

Key words: Aratinga pertinax, breeding biology, Brown-throated Parakeet, Caribbean, colonization, Culebra Island,Cura"ao, extinction, extiryation, parakeet, Puerto Rico, range expansion, St. john, St. Thomas, 1Ortola, Vieques Island,Virgin Islands, West Indies.

exhibit ecological release: their populations in-crease greatly and spread into habitats not occu-pied by the parent population (MacArthur &Wilson 1963, 1967; Cox & Ricklefs 1977). Fur-thermore, invading populations may be able toexploit niches not occupied by native species.After the invaders become established, however,their competitive ability appears to wane, thendistribution among habitats becomes restricted,and local population density decr~ses. Thesetrends may eventually lead to extirpation. Fur-thermore, the probability of extirpation in-

INTRODUCTION

Source populations of colonizing species areusually abundant and widespread in the originlocality. Immigrants to islands can initiallyappear to be excellent competitors and may

* Present address: Cooperative Research Unit Center,

U.S. Fish and Wildlife Service, Washington, D.C.,U.S.A.Mailing address: Grambling Cooperative Wildlife Pro-ject, P.O. Box 4290, Grambling State University,Grambling, Louisiana 71245, U.S.A.

43

WILEY

(2,730ha, Fig. 1) as follows: 30 October-1November 1974, 19-21 February 1975, 22-23April 1975, 8-9 November 1977, 18-19 July1978, 31 July-1 August 1978, and 20-23 June1980. Culebra Island is 27km northeast ofPuerto Rico and 25 km west of St. Thomas. Thenatural vegetation is subtropical dry forest (Ewel& Whitmore 1973).

I visited Vieques Island (13,766ha, Fig. 1) on30 October 1974, 19 February 1975, 22 April1975,8 November 1977, 18 July 1978,31 July1978,.14 October 1978, and 31 October 1978.Vieques Island lies 9.7 km east of Puerto Ricoand 35km southwest of St. Thomas. The vegeta-tion of Vieques Island is classified as subtropicaldry forest and subtropical moist forest (Ewel &Whitmore 1973).

The Roosevelt Roads Naval Station (3,260ha,Fig. 1) is within the subtropical dry forest zone(Ewel & Whitmore 1973) of easternmost PuertoRico and is characterized by salt pannes andmangrove forests in the lowlands (dominated byBlack Mangrove Avicennia germinans, Red Man-grove Rhizophora mangle, White Mangrove La-guncularia racemosa, Button Mangrove Conocar-pus erectus) and scrub dominated by the exoticleadtree l.i?ucaena leucocephala in the dry areas(Wiley & Wiley 1979). Hardwood canyon for-ests, dominated by Royal Palm Roystonea borin-quena, Red Manjack Cordia nitida, Mango Man-gifera indica, and Shortleaf Fig Ficus citrifolia,lead into the mangrove bottoms. I visited theRoosevelt Roads Naval Station at least each 2-3days from March through July in 1974-1975and again in 1977-1981, and at 1-2 week inter-vals during August-February in 1974-1978 andin 1980-1981. I continued to visit the RooseveltRoads study area at least every other week from1982 through 1986.

creases as population size decreases (MacArthur& Wilson 1967, Piek>u 1969). Extirpation ofsmall local populations may occur when localclimatic conditions become severe, when an effi-cient predatot is introduced into the community,or as a result of competition with ecologicallysimilar species (DeBach 1966, Ricklefs 1979). Co-lonizing populations on small islands are partic-ularly vulnerable to extirpation because they arerestricted geographically and are infrequently

augme~ted by immigration.I observed an incipient colonization of sev-

eral West Indian islands by an exotic psittacine,the Brown-throated Parakeet, A ratinga pertinax.The species is native to Panama, northern SouthAmerica, and islands off the northern coast ofVenezuela (including Curas:ao, Bonaire, andAruba of the Netherlands Antilles; AmericanOrnithologists' Union 1983, Forshaw 1989).The parakeet was recently introduced to Domini-ca, formerly occurred on Martinique (now extir-pated), and persists in St. Thomas, U.S. Virgin Is-lands (Clark 1905, Norton 1983, Forshaw 1989).Marien & Koopman (1955) suggested three possi-ble scenarios for the source of the disjunct popu-lation of Aratinga pertinax in St. Thomas: (1) in-troduction by humans, (2) direct natural over-water transportation, perhaps storm-aided, and(3) natural range extension by island-hoppingthrough the Lesser Antilles, followed by local ex-tirpations. Most authorities favor the hypothesisthat the species was introduced by man (Grafvon Berlepsch in Hartert 1893, Marien & Koop-man 1955, Leopold 1963).

I review the status of A. pertinax in St.Thomas and other Virgin Islands, and I reporton a recent unsuccessful attempt at colonizingthe nearby islands of Vieques, Culebra, and Puer-to Rico. I use these data to discuss the origin ofthe St. Thomas population and possible biogeo-graphical, ecological, and conservation im-plications for other species of West Indian

parakeets.

OBSERVATIONS AT NESTS

Observations of breeding parakeets were madefrom blinds placed on the ground or in a treeabout 10 m from the nest. Individual birds wereidentifiable in the short term (i.e., seasonally) bymolt and in the long-term (year to year) by facialplumage and color of soft parts. Nine chicks (3in 1978, 6 in 1979) were marked with stainlesssteel leg bands. At four nests I installed a door

HABITATS VISITED

I observed parakeets in subtropical moist andsubtropical dry forests of eastern and western St.Thomas during 31 May-3 June 1977, and Julyof both 1978 and 1979. I visited Culebra Island

44

PARAKEET RANGE EXTENSION AND EXTINCTION

o 20 40 Km

FIG. 1. Northwestern Virgin Islands, Culebra Island, Vieques Island, and eastern Puerto Rico.

large enough to allow access to the nest chamberfor regular inspections. Nests were inspected forcontents at one-week intervals until about oneweek before chicks fledged. Nests were not in-spected again until about one week after theyoungest chick was expected to fledge. Fledgingwas assumed if chicks reached the full-featheredstage and no chick remains or other evidence ofpredation were found in the nest chamber.

NEST SITE AVAILABILITY

At Roosevelt Roads, I surveyed habitat for para-keet nest cavities in 1.64ha of white and blackmangrove-dominated forests and adjacent hard-wood canyon forest. Characteristics of trees con-taining A. pertinax nest cavities were determinedfrom active nests and from published records andunpublished data in museum collections of eggs.From these data, I derived the following conser-vative measurements, which I used in my surveysof nest site availability: diameter at breast height(dbh) of nest cavity-bearing tree -~ 40cm,height of cavity above ground -3 m, diametercavity entrance -7.0 cm, interior chamber dia-meter -20cm, cavity depth -40cm.

The minimum diameter of a termitarium wasbased on observations of active termite nests usedby A. pertinax in St. Thomas and Puerto Ricoand from unpublished data on the similar-sizedHispaniolan Parakeet (A. chloroptera, length-32cmyersus 25cm for A. pertinax) in Hispan-iola and Orange-fronted Parakeet (A. canicularis,length -24cm) in Guatemala (Wiley, unpubl.data). from these observations, I judged theminimum diameter of a termitarium suitable fora parakeet nest to be 55cm, the minimum dbh ofa termitarium-bearing tree to be 20cm, and theminimum height above the ground of a suitablecavity 3 m.

Trees of suitable size were searched forcavities from the ground or, if sections of thetrunk were not visible from that perspective, byclimbing. All cavities were inspected whileclimbing the trees.

CALCULATIONS

Expected extirpation rates (P¡OV of the pOpU-lation were calculated following Pielou's (1969)formula. I followed Ricklefs (1979) in deter-mining feeding niche overlap among species.

45

STATUS AND POPULATION

CHANGES ON s.T. THOMAS AND

OfHER VIRGIN ISLANDS

Because of morphological resemblances, mostauthorities have considered the St. Thomas pop-ulation of A. pertinax to represent the same raceas is endemic to Cura~ao, A. p. pertimax (Hartert1893, von Berlepsch in Hartert 1893, Ridgway1916, Wetmore 1927, ~pold 1963, Forshaw1989; but see Marien & Koopman 1955). If theparake~t has been introduced to St. Thomas, itmust have been before the mid-nineteenth cen-tury, for several naturalists reported parakeets(probably A. pertinax) in St. Thomas during thatperiod (Ledru 1810, Knox 1852, Newton 1859,Sclater 1859). The earliest known specimen fromthe Virgin Islands was collected in 1860 (Forshaw

1989).Although generally regarded as common on

St. Thomas from the mid-nineteenth centurythrough the early part of the twentieth century,the Brown-throated Parakeet population under-went substantial fluctuations in numbers, in-cluding near-extirpation by the hurricanes of1926 and 1928 (Knox 1852, Newton & Newton1859, Eggers 1878, Hartert 1893, Nicoll 1904,Mortensen [1909] in Wetmore 1927, Nichols1943). It subsequently increased in numbers but,like before the 1926 and 1928 storms, remainedlargely confined to the eastern half of the island(Nichols 1943). By the early 1960s, the parakeethad extended its range to most parts of St.Thomas, and the population was estimated at a-bout 400 birds (~pold 1963). Murray (1969) re-ported it as a common resident nesting on St.Thomas. In recent years, it has also been recordedfrom St. John (3.8 km east of St. Thomas, Fig. 1),where apparently it has not become established(American Ornithologists' Union 1983, Raffaele1983, Norton 1985; Fig. 1). Mirechi et al. (1977)reported that occasional1y groups of A. pertinaxstrayed from St. Thomas to Tortola 15km north-east of St. Thomas (Fig. 1) and that possibly asmall flock resided there.

Kepler & Kepler 1978). However, Gundlach(1878) heard of parakeets (species unknown) onVieques and Wetmore (1916a, 1916b) was toldthey sometimes were seen on that island in theJune to August rainy season. Sorrie (1975) didnot find them on Vieques Island during his workfrom June 1970 to July 1971, nor did colleaguesand I detect the parakeet during surveys of Vie-ques and Culebra in 1974. However, on 22 April1975 I found 3-7 A. pertinax at EnsenadaHonda (Yanuel Lagoon) and 12 unidentifiedparakeets near Puerto Ferro, Vieques. J. Declet(in Pérez-Rivera & Vélez 1980) also observedBrown-throated Parakeets in Vieques Island.

On Culebra Island, several contemporaryobservations were made of unidentified para-keets. Cameron Kepler (pers. comm.) was toldby a Culebra resident that parakeets (A ratingachloroptera?) had regularly visited Cayo Norte tofeed on Coccoloba uvifera in December forseveral years in the late 1960s and early 1970s.Herbert Raffaele (pers. comm.) was told of aflock of parakeets (species unknown) at PlayaLarga in Culebra in 1972. A Culebra residentalso told Raffaele of a flock of about 8 parakeetsat Playa Grande on 31 May 1972. IobservedAra.tinga pertinax in the forested hillsides northeastof Playa Flamenco on Culebra Island on 21 Feb-ruary 1975, when I estimated a population of4-8 individuals. On 23 Apri11975, I saw a flockof 25 A. pertinax at Playa Grande on CulebraIsland.

I did not find pertinax on either island duringvisits in subsequent years, nor were other reportsof the parakeet made after 1975 (Duffield & Car-dona 1978, Furniss & Collazo 1983, Marc Weit-

zel, pers. comm.).

HISTORY OF ARATINGA PERTINAX

COLONIZATION OF

EASTERN PUERTO RICO

Historically, Aratinga pertinax was not con-firmed from Puerto Rico (Chaplain 1859, Wet-more 1916a, 1927), although Chaplain (1859)noted parakeets there, which Clark presumed we-re introduced Brown-throated Parakeets fromCura~ao. Sorrie (pers. comm.) did not findAratinga pertinax on the Roosevelt Roads NavalStation during his residence there from Novem-ber 1969 to August 1971. I also did not find the

SPREAD OF THE PARAKEET TOPUERTO RICO'S EASTERN SATELUTES

The Brown-throated Parakeet was not previouslyrecorded from Vieques or Culebra islands (e.g.,Wetmore 1916a, 1916b, 1917,1927; Sorrie 1975;

PARAKEET RANGE EXTENSION AND EXTINCTION

larly in fruit plantations, manchineel (Hippoma-ne mancinella) thickets, mangroves, and semi-de-serts with cactus and acacias (Voous 1983). In St.Thomas, the parakeet prefers wooded thickets inthe hills, although it descends at times to feed onfruits in the lowlands (Nichols 1943, Raffaele1983, pers. obs.). Norton (in Raffaele 1983) re-ported it as uncommon in mangrove habitat onSt. Thomas. Habitat use by the parakeet inPuerto Rico and its satellites was similar to thatin St. Thomas. Parakeets were observed in man-grove forests and open woodlands on hillsides onCulebra Island, mangrove forest edge and openscrub on Vieques Island, and mangrove forestand adjacent hardwood canyon-leadtree scrubedge in eastern Puerto Rico. Parakeets roosted

parakeet there during extensive field work in1974 and early 1975. Mary Hickman (pers.comm), a Base resident and avid bird watcher,first observed the parakeet at Roosevelt Roads inApril 1975. The population at that time con-sisted of a minimum of S individuals (2 pairs and1 single), but grew to a maximum size of 10 birdsin June 1979 (Table 1). I found no evidence ofadditional immigration into the area after 1975.Eleven chicks fledged from five known nestingattempts, but the young birds suffered highmortality during their first year (Table 1). Thepopulation declined to 1 bird by July 1980, andwas extirpated by the beginning of the 1982breeding season (Table 1).

Expected extirpation rate of the population(p[OV was low during the first years of the colo-nization when reproduction was good and mor-tality was low (Fig. 2). However, thereafter p[O]quickly approached unity (xP[O] 1979-1980 =

0.999). :2:o.."9

.01

001HABITAT USE OF THE COLONIZING

POPULATION

A ratinga pertinax uses a wide variety of habitat

types rangewide, including mangroves, savan-nahs, open woodland, dry thorn and cactusscrublands, cultivated farmlands, fruit planta-tions, and gardens (Lowe 1907, Haverschmidt1968, Wetmore 1968, Meyer de Schauensee 1970,Voous 1983). The Curas:ao race is found invirtually all of that island's habitats, but particu-

0001

o1976 1977 1978 1979 1980 1981 1982

YEAR

FIG. 2. Expected extirpation rate (P¡O}) of Aratinga per-tinax population at Roosevelt Roads Naval Station,eastern Puerto Rico, 1975-1981.

47

and 15cm in height (Voous 1983). In St. Tho-mas, Nichols (1943) reported a nest burrowabout 1 m deep in a live termite nest, 5.5 m fromthe ground. Clutch size is generally reported as4-7 eggs (Forshaw 1989).

and nested primarily in the mangrove forest, butforaged in adjacent c~nyons and, occasionally,suburban areas.

BREEDING BIOLOGY IN EASTERN

PUERTO RICO

Chronology. I found 5 nests in eastern PuertoRico from 1977 to 1980 (Table 2). Nests con-tained eggs in mid-March through late April,which coincides with the onset of the rainyseason in eastern Puerto Rico (Wiley & Wiley1979). Chicks fledged from mid-May to late Juneand loosely associated with adults at least untilthe next breeding season.

Nest Site Characteristics. Four of the five parakeetnests were in active termitaria of Nasutitermescostalis in trees at the inland edge of white man-grove-dominated forest and in a narrow hard-wood canyon. The insects sealed off the birds'breeding chamber with layers of cellulose, but Iobserved no other obvious interactions betweenthe birds and termites. One of the termitarianests was in a Mango (Mangifera indica), two we-re in White Mangroves, and another was in aBlack Mangrove (Table 2). The fifth nest was11.9m high in the broken trunk of a dead RoyalPalm (Roystonea borinquena). Entrances of nestsin termitaria averaged 9.8 :I: 1.71 (S.D., Range =8-12) cm in diameter and 5.3 :I: 1.13 (R =

4.3-6.9) m above the ground (Table 3). Nestchambers averaged 57.0 :I: 14.58 (R = 40-75)

GENERAL ECOIOGY AND BEHAVIOR

Aratinga pertinax feeds and roosts in pairs orsmall to 1arge flocks (Wetmore 1968, Voous 1983,Forshaw 1989). The parakeet is a food generalist,feeding on a variety of seeds, fruits, nuts, blos-soms, and possibly insects and their larvae (Har-tert 1893, Voous 1983, Forshaw 1989). It showsno definite breeding season in most parts of itsrange (Schafer & Phelps 1954, Haverschmidt1968, Wetmore 1968). Aratinga pertinax onCura~ao has been reported as breeding through-out the year, but primarily following a rainyperiod (Voous 1957, 1983). On St. Thomas, abird collected on 22 March 1892 had an egg in itsoviduct, and Nichols (1943) took 2 eggs from anest on 23 March (in collection of Western Foun-dation of Vertebrate Zoology, ws Angeles).

Breeding pairs are usually dispersed, but birdshave been reported nesting in small colonies ofup to seven pairs (Voous 1957). Aratinga pertinaxnests in holes, often excavated by the birds indecayed trunks of date palms, in earthen andsand walls, and in crevices and holes in rock(Hartert 1893, Voous 1983, Forshaw 1989). Mostfrequently, nesting holes are dug in large treenests of termites (Myers 1935, Hindwood 1959).Nests of Brown-throated Parakeets using termi-taria in the Netherlands Antilles had an entrancetunnel of up to 50cm in length and 7-10cm inwidth, with a nest chamber of 25 cm in diameter

TABLE 2. Productivity of Aratinga pertinax at the Roosevelt Roads Naval Station, eastern Puerto Rico,1977-1980.

11

2.6 :t 1.67 2.2 :i: 30

a AII eggs were fertile. b Nests were not watched through fledging. Fledging was assumed if chicks reached the completely

feathered stage (i.e., within 1 week of fledging), and no chick remains or evidence of predation were found in the nest chamber.

'ARAKEET RANGE EXTENSION AND EXTINC

TABLE 3. Dimensions of Aratinga pertinax nest cavities in termitaria and a Roya! Palm, Roosevelt Roads NavalStation, eastern Puerto Rico, 1977-1980.

Nest chamber

InsideHeight above

ground (m)

Entrancediameter (cm) diameter (cm) depth (cm)Nest number

5.3

4.8

4.3

6.9

5.3 :I: 1.1

10

9

12

8

25

27

23

27

25.5 :t 1.91

75

40

60

53

57.0 :t 14.589.8 :t

Nest sites in termitaria1345

Mean :t SD

Nest site in dead Royal Palm2

All nest sitesMean :t SD 6.6 :t 3.10

11.9 48 40 92

17.4 :t 17.17 28.4 :t 6.69 64.0 :t 20.11

Thrashers were observed on the termitaria usedby the parakeets, but they only perched momen-tarily and made no attempt to enter the cavities.

FOOD HABITS

In Puerto Rico, I observed Aratinga pertinaxfeeding on fruits and seeds of 14 plant species,especially Mango, Red Manjack, Royal Palm,Brisselet (Erythroxylum rotundifolium), FloridaBoxwood (SchaejJeria frutescens), Coco Plum( Chrysobalanus icaco ), Mountain Immortale(Erythrina poeppigiana), and Short-leaf Fig (Table4). Voous (1957) reported that crop and stomach

TABLE 4. Food items observed being eaten by Ara-tinga pertinax at Roosevelt Roads Naval Station,eastern Puerto Rico, 1975-1981.

Mangifera indicaCordia nitida

Roystonea borinquenaErythroxylum rotundifoliumSchaefferia frutescensChrysobalanus icaco

Erythrina poeppigianaFicus citrijoliaAvicennia germinansCoccoloba unifera

Prosopis julijloraT amarindus indicaAnnona muricataCitus aurantium

cm deep and 25.5 :t 1.91 (R = 23-27) cm inside

diameter. I observed A. pertinax modifying thetermitaria cavities, but did not see birds selectingor initiating excavation of the cavities.

Cavity Availability. Of the 164 suitably sized(dbh ~ 0.40m) trees in 1.64ha surveyed atRoosevelt Roads, only 5.5% (n = 9) contained

cavities of adequate dimensions for nesting para-keets. Of these cavities, 56% (n = 5) were at

heights (~ 3 m) suitable for parakeet nests. Tree-borne termitaria were more common than suita-ble cavities; 11.5% of larger trees (dbh ~ 20cm;n = 1,844) had full-sized termitaria (i.e., mini-

mum 55cm diameter). Of these termitaria, 79%(n = 168) were at a height suitable for nesting

parakeets.

Productivity and Nest Success of the ColonizingPopulation. Clutches at Roosevelt Roads averaged5.4 :t 1.14 (R = 4-7) eggs, all of which were

fertile, 66.7% hatched, and 40.7% fledged (Table2). Of those nests that hatched eggs, broodsaveraged 4.5 :t 0.58 (R = 4-5) first-week chicks;3.3 :t 0.96 (R = 2-4) chicks survived to themid-nestling period; and 2.8 :t 0.50 (R = 2-3)

chicks fledged (i.e., at least reached the last weekof the brooding period).

One of the five nests in eastern Puerto Ricofailed, apparently after the adults deserted for un-determined causes. All other nests were success-ful. Pearly-eyed Thrashers (Margarops fuscatus)are aggressive predators of the eggs and youngchicks of cavity-nesting birds (Snyder et al. 1987).

49

WILE'

of adequate dimensions and condition withinthe 1.64ha I sampled, 5 contained honeybeecolonies and another a rat nest. Furthermore, ofthese cavities, only 5 were at a height suitable fornesting parrots, and of these, 3 were occupied bybees and rats.

contents of birds collected in the Netherland An-tilles were comprised of fruits and seeds; Caesal-pinia, Acacia, Prosopi~, and Organ Pipe Cactus(Cereus repandus) were the most numerous. Birdswere also seen eating the fruits of Malpighia andflowers of Gliricidia sepium (Voous 1957).

ORIGIN OF THE ST. THOMAS

POPULATION

Evidence presented here suggests that Aratingapertinax could have, but probably did not,colonize St. Thomas, perhaps through a series ofisland hops originating in Curas:ao. Althoughthe distances among St. Thomas, Puerto Rico,and the intervening islands are not nearly sogreat as that between St. Thomas and Curas:ao(ca. 770km), my observations suggest that A.pertinax is capable of at least short over-water dis-persal. Such movements may have been aided bytropical storms, which can be powerful agents ofdispersal (Wiley, in press). One or more such dis-

persions, perhaps supplemented by subsequentevents, could have landed sufficient birds on St.Thomas to establish a viable population.

However, A. pertinax does not commonlymove over water to new islands, as evidenced bythe fact that each of the Dutch West Indianislands with A. pertinax populations has aunique subspecies: Aruba -A. p. arubensis,Bonaire -A. p. xanthogenius, Curas:ao -A. p.pertinax. Distances among these islands are notgreat: Aruba-Curas:ao = 78 km, Curas:ao-Bo-naire = 52 km, Bonaire-Aruba = ca. 135 km.

The distances between Venezuela and Aruba(30km), Curas:ao (70km), and Bonaire (87km)are also not great. These distances are about thesame as those among the Virgin and PuertoRican islands: Culebra Island to St. Thomas-25 km, Culebra Island to Vieques Island-15km, Vieques Island to St. Thomas -35km,eastern Puerto Rico to St. Thomas -60 km.

Perhaps a former effective barrier against theestablishment of A. pertinax on other islands wasthe extensive A ratinga faunas on many of theLesser Antilles (Fig. 3). Historically, endemicspecies of Aratinga occurred on Martinique,Guadeloupe, Dominica, and St. Lucia. Formercompetition with these native species may havebeen sufficient to prevent A. pertinax fromestablishing populations on these islands. No

PorENTIAL COMPETITORS

Pigeons and doves were the avian species perhapsclosest to A. pertinax in food habits. I observedchases (parakeet at dove, and vice versa), appar-ently related to food, among A. pertinax andScaly-naped Pigeons (Columba squamosa), White-crowned Pigeons (C. leucocephala), Zenaida Doves(anaida aurita), and White-winged Doves (Z.asiatica). I found the parakeet had a moderatelyhigh feeding niche overlap (Ojk,) with some spe-cies of columbids: notably Ojk(White-crownedPigeon versus Aratinga pertinax) = 0.814,Ojk(Zenaida Dove versus A. pertinax) = 0.732.

Four common native bird species breed incavities in eastern Puerto Rico. The Puerto RicanScreech-Owl ( Otus nudipes) and Puerto RicanWoodpecker (Melanerpes portoricensis) were un-common in the area (perhaps because of low nestcavity availability). The Pearly-eyed Thrasherand American Kestrel (Falco sparverius) werecommon at Roosevelt Roads and nested incavities in hardwoods and palms. However, thekestrel preferred more open habitat (mixed palm-savannah) than the closed mangrove and canyonforests frequented by the parakeets. Pearly-eyedThrashers were among the most common birdsin the mangrove forest and were frequent ne~tersthere. None of these cavity-nesting species usedtermitaria for breeding.

Both native psittacine species ( endangeredPuerto Rican Parrot Amazona vittata and ex-tinct Puerto Rican Parakeet Aratinga maugei) areno longer in the lowland habitat in eastern Puer-to Rico (Snyder et al. 1987). During the mid tolate 1970s, 2 other exotic psittacines appeared ineastern Puerto Rico. Two Yellow-headed Parrots(Amazona ochrocephala) and one Rose-ringedParakeet (Psittacula krameri) resided in the man-groves and uplands at Roosevelt Roads, but didnot nest in the termitaria or tree cavities.

Exotic honeybees (Apis melifera) and roofrats (Rattus rattus) nested and roosted in treecavities at Roosevelt Roads. Of the 9 tree cavities

50

PARAKEET RANGE EXTENSION AND EXTINCTION

\FlDADA } ~""')) RA1-1A~AA

"\1 -~ ~A~~~~ .--¡1 ~ o~ ~ ~SLANDS

, \

tN

A TLANTIC OCEAN

~~- .~" ~~

JAMAICA

?A. nana

~ ...-6"""".

df'A. euop

~o "

.,. .A. Isbatl*o /

GUADELOUPE ~

"' \) OOMIN",AUnnsmea-

~ST WCIA

Unnsmed*--"'6 BAABAOOS

~ ¡tIO-' Unn.mea"

,;7

..~iJ

, HISPANIOLA~ PUERTO ~ RICO ~ (--"'-"] ..' "

J' ~lf ,.. "'

A. chtoropters A. msuget.

-~CENmAL

, AMERICA f'00

CARIBBEAN SEA km

~\ ~~

FIG. 3. Historic and present populations of native species of Aratinga in the West Indies. Extinct species andpopulations indicated by an asterisk.

race, it is likely that more extensive differen-tiation would have occurred in the time it tookfor the species to island hop to St. Thomas, andthen disappear from intervening islands.

In summary, lack of records, differentiation,and intervening populations suggest that A. per-tinax did not naturally spread to the Virgin Is-lands from a southern source population. TheBrown-throated Parakeet was apparently a popu-lar cage bird. It was harvested from Cura9ao inlarge numbers (Hartert 1893) and wascommonin the pet trade in the West Indies and Europe(Russ 1895, Tavistock 1929, Seth-Smith 1903,Norton 1983). The popularity of the species asa cage bird and its regularity in the trade giveplausible support for the hypothesis of introduc-tion to St. Thomas.

native Aratinga has been reported from theVirgin Islands and Puerto Rico's endemic spe-cies, A ratinga maugei, is extinct.

The St. Thomas population of A. pertinaxhas been assumed by most authors to have origi-nated through introductions from Cura~ao stock(von Berlepsch in Hartert 1893, Hartert 1893,Nichols 1943, Marien & Koopman 1955, For-shaw 1989). Consistent with this hypothesis isthat none of the islands between Cura~ao and St.Thomas, including those closest to the Dutch is-land, has A. pertinax populations. Also, thehistorical and fossil records, though fragmentary,suggest A. pertinax was not present on theseislands or the Virgin Islands in the past (Wet-more 1918).

Regardless of the means of dispersal, A. perti.nax is likely a recent arrival in St. Thomas. Thespecies seems quite plastic, and populations onother islands and the mainland show distinctracial differentiation. However, the St. Thomaspopulation is considered identical with theCura~ao race (Hartert 1893). Although Marien& Koopman (1955) thought the St. Thomaspopulation averaged smaller than the Cura~ao

RECENT EXPANSION TOPUERTO RICO

The range expansion reported here occurred at atime when the St. Thomas population hadgreatly increased in numbers since the hurricane-caused low numbers in the mid- to late-1920s.

for most other avian species, and extensivelyused exotic fruits abundant in natural and resi-dential areas.

Predation is the probable majorsource of lossof the young and adult parakeets. Nest successand productivity were reasonably high, perhapspartly because bird nests in termitaria were pro-tected from most forms of local predation by theincidental defenses of the host insects. However,mortality rates of free-flying birds were apparent-ly high. Red-tailed Hawks (Buteo jamaicensis) arethe most important predators of medium-sizedbirds, including parrots, in eastern Puerto Rico(Wiley & Wiley 1979, Snyder et al. 1987). Whenpopulations consist of few individuals and areconfined to small areas, the stochastic loss ofbirds, even by low predation pressure, has sub-stantial effect on population viability. Also, par-rots, including A. pertinax, are flocking species,which derive survival benefits from the vigilanceprovided by the several individuals comprisingthe flock (Snyder et al. 1987, Westcott & Cock-burn 1988). If numbers of individuals in a pOpU-lation are reduced to the level where insufficientbirds are available for adequate predator vigi-lance, the risk of capture increases substantially.

An alternative explanation to the hypothesisof predator-caused extirpation of the parakeets atRoosevelt Roads may be that the birds dispersedto other, perhaps more favorable, sites. AlthoughI failed to find A. pertinax populations insuitable habitat adjacent to the Roosevelt Roadsstudy area, dispersed populations of A. pertinaxhave been reported from other parts of PuertoRico since the observations recorded here. No-where has it become numerous (Raffaele 1983,Pérez-Rivera & Vélez 1980; Raul A. Pérez-Rive-ra, pers. comm.); the largest population is at Ca-boSanJuan, where small groups persist (about8 birds/group; Pérez-Rivera, pers. comm.). Thesource of this and other populations in PuertoRico have been suggested as escaped cage birds

(Pérez-Rivera 1980).

Also, the parakeet had increased its range fromprimarily eastern St. Thomas to throughout theisland. Recently, the .species has been recordedfrom the adjacent islands of St. John and Tortola(Mirechi et al. 1977, Raffaele 1983).

The expansion of A. pertinax westward in1975 was apparently not storm-aided. No severestorms o"ccurred in this part of the West Indies inthat year or in 1974 (National Climatic Center

1974, 1975)., Population characteristics of increase in num-bers a~d range are favorable to natural rangeexpansions. I believe the St. Thomas population,already expanding throughout that island andwith small satellite populations appearing on St.John and Tortola, briefly expanded to Culebra,Vieques, and Puerto Rico through a series of is-land hops. Colonization was unsuccessful in allof these islands.

FACTORS INVOLVED IN THE

EXTIRPATION OF THE EASTERN

PUERTO RICO POPULATION

The probability of a colonizing population becom-ing established is higher when there is a largenumber of founders, a rapid rate of populationincrease, and minimum competition (MacArthur& Wilson 1967). The failure of Aratinga pertinaxto colonize eastern Puerto Rico in the late 1970smay be the result of several interactive factors.The small size of the original innoculum withno further outside immigration into the popula-tion ensured a low probability of establishment.Competition with native and introduced ani-mals may have also been a factor, but to an un-known degree. Although tree cavities for nestingevidently were limited in availability, termitariaof sufficient size for A. pertinax nests were abun-dant and, apparently, A. pertinax had no compe-tition for these nest sites. Competition for foodwith other species is an unlikely reason for theparakeet's local extirpation. The food speciesused by parakeets and potential competitors(especially columbids) were plentiful in easternPuerto Rico and available year-round through asuccession of fruiting. Also, populations ofpotential competitors were depressed relative tohistoric times because of over-hunting and habi-tat degradation (Wiley & Wiley 1979, Wiley1985). Furthermore, A. pertinax was able to feedon large fruits (e.g., mango) that were unsuitable

IMPLICATIONS FOR CONSERVATION

PROGRAMS

At the time of Columbus "discovery" of theWest Indies, the region's parrot fauna consistedof no less than 28 species, including 7 macaws, 12Amawn parrots, and 9 parakeets (Fig. 3). Lessthan 500 years later, that impressive array of uni-

52

PARAKEI RANGE EXTENSION AND EXTINCTION

que varieties has been reduced to only 10 species,including but 4 parak@ets. Whereas all endemicspecies of Aratinga in the Lesser Antilles areextinct and most Greater Antillean species aredeclining or extinct, A. pertinax has expanded itsrange and increased its numbers on St. Thomas(Wiley \991). These population changes arelikely a result of improved protection fromshooting and exploitation, extensive planting ofexotic vegetation, which provides food and shel-ter for the parakeet, and reduced competitionfrom native frugivores. Examples of other psitta-cine species, threatened or extinct in their nativerange, yet flourishing in their adopted home-lands of suburban and urban environments, arecommon (Wiley et al. 1991).

Conservation programs for West Indian para-keets must incorporate strategies for managingthe native species in the predominantly exotic,yet protected, suburban habitats -especially inthis era of rapidly disappearing native habitats inthe region. Conservation efforts must also ad-dress the threats posed by exotic psittacines.Introduced parrots may affect native speciesthrough competition for limited resources, bythe spread of diseases, or through hybridization.

ACKNOWLEDGEMENTS

I thank Mary Hickman, Elizabeth Litovich, thelate Hector O'Neil, and Beth Wiley for theirhelp in observing parakeets. Herbert A. Raffaeleand Cameron B. Kepler provided data on theparakeet from their unpublished field notes, forwhich I thank them. Raul A. Pérez-Rivera gene-rously permitted me to use his observations onthe current status of A. pertinax in Puerto Rico.I am grateful to Lloyd F. Kiff and Jim Jennings(Western Foundation of Vertebrate Zoology) andEugene C;ardiff (San Bernardino Museum ofNatural History) for allowing me access to theiregg collections. I thank Wylie C. Barrow, Jr.,Gerald Lindsey, Oliver H. Pattee, and MarciaWilson for suggestions which improved the

manuscript.

LITERATURE CITED

American Ornithologist' Union. 1983. Check-list ofNorth American birds, 6th ed. Lawrence.

Chaplain, s. 1859. Narrative on a voyage to the WestIndies and Mexico in the years 1599-1602. Hak-lauyt Soc.

Clark, A. H. 1905. The genus Conurus in the WestIndies. Auk 22: 310-312.

Cox, G. W., & R.E. Ricklefs. 1977. Species diversity,ecologiCaI release, and community structuring inCaribbean land bird faunas. Oikos 29: 60-66.

DeBach, P. 1966. The competitive displacement andcoexistence principies. Ann. Rev. Entomol. 11:183-212.

Duffield, J., & J. Cardona. 1978. Estimated avian pOpU-lation density and diversity indices of ViequesIsland. Puerto Rico Dept. Nat. Res., San Juan.

Eggers, H. F. A. 1878. Naturen paa de dansk-vestindiskeger. Tidsskr. pop. Fremst. Naturrid. pp. 1-34.

Ewel, J.J., & J.L. Whitmore. 1973. Las zonas de vidade Puerto Rico y las Islas V¡rgenes Americanas:una sinopsis. Bol. Invest. Servicio Forestal ITF-18A.

Forshaw, J. M. 1989. Parrots of the world. 3rd ed. Mel-bourne.

Furniss, S., & J. Collazo. 1983. Recent avian recordsfor the Culebra Archipelago, Puerto Rico. Pp.115-121 in Memorias del Cuarto Simposio sobrela Fauna de Puerto Rico y el Caribe, Univ. PuertoRico, Humacao.

Gundlach, J. 1878. Neue Beitriige zur Ornithologie derInsel Porto Rico. J. Orn. 26: 157-194.

Hartert, E. 1893. On the birds of the islands of Aruba,Cura~ao and Bonaire. Ibis 5: 289-338.

Haverschmidt, F. 1968. Birds of Surinam. Edinburgh.

Hindwood, K.A. 1959. The nesting of birds inthenests of social insects. Emu 59: 1-36.

Kepler, C.B., & A.K. Kepler. 1978. The sea birds ofCulebra and its adjacent islands, Puerto Rico.Living Bird 16: 21-50.

Knox, J.P. 1852. A historical account of St. Thomas,W.I. New York.

Ledru, A.P. 1810. Voyage aux iles de Tenerife, La Tri-nité, Sainte Croix et Porto Rico, etc. Arthur Ber-trand, Libraire, Paris, 2 vols.

Leopold, N. F. 1963. Check-Iist of birds of Puerto Ricoand the Virgin Islands. Bull. Puerto Rico Agric.Exp. Sta. No.168.

Lowe, P.R. 1907. On the birds of Margarita Island,Venezuela. Ibis 1: 547-570.

MacArthur, R.H., & E.O. Wilson. 1963. An equili-brium theory of insular zoogeography. Evolution17: 373-387.

MacArthur, R.H., & E.O. Wilson. 1967. The theoryof island biogeography. Princeton.

Marien, D., & K.F. Koopman. 1955. The relationshipsof the West Indian species of Aratinga (Aves, Psitta-cidae). Am. Mus. Novitates 1712.

Meyer de Schauensee, R. 1970. A guide to the birds ofSouth America. Wynnewood.

53

WILEY

Sclater, P.L. 1859. Descriptions of two new species ofAmerican parrots. Ann. Mag. Nat. Hist., Ser. 3,4:224-226.

Seth-Smith, D. 1903. Parrakeets. A handbook to theimported species. London.

Snyder, N.F.R.,J. W. Wiley & C.B. Kepler. 1987. Theparrots of Luquillo: natural history and conserva-tion of the Puerto Rican parrot. West. Found. Vert.Zool., Los Angeles, California.

Sorrie, B.A. 1975. Observations on the birds ofVieques Island, Puerto Rico. Caribb. J. Sci. 15:

89-103.Tavistock, Marquess of. 1929. Parrots and parrot-like

birds in aviculture. London.Voous, K.H. 1957. The birds of Aruba, Cu~ao, and

Bonaire. Stud. Fauna Cura~o Other Caribb. IsI. 7:

1-260.Voous, K.H. 1983. Birds of the Netherlands Antilles.

2nd ed. Zutphen, Utrecht.Westcott, D.A., & A. Cockburn. 1988. Flock size and

vigilance in parrots. Aust. J. Zool. 36: 335-349.Wetmore, A. 1916a. Birds of Porto Rico. U.S. Dept.

Agric. Bull. 326.Wetmore, A. 1916 b. The birds of Vieques Island, Porto

Rico. Auk 33: 403-419.Wetmore, A. 1917. The birds of Culebra Island, Porto

Rico. Auk 34: 51-62.Wetmore, A. 1918. Bones of birds collected by Theo-

door de Booy from kitchen midden deposits in theislands of St. Thomas and St. Croix. Proc. U.S.

Nat. Mus. 54: 513-522.Wetmore, A. 1927. The birds of Porto Rico and the

Virgin Islands. Sci. Surv. Puerto Rico VirginIslands 9: 409-598.

Wetmore, A. 1968. The birds of the Republic of Pana-ma, part 2, Columbidae (Pigeons) to Picidae(Woodpeckers). Washington, D.C.

Wiley, J. W 1985. Bird conservation in the UnitedStates Caribbean. Bird Conserv. 2: 107-159.

Wiley, J. W 1991. Status and conservation of parrotsand parakeets in the Greater Antilles, BahamaIslands, and Cayman Islands. Bird Conserv. Inter-

nat. 1: 187-214.Wiley, J. W. In press. The effects of hurricanes on island

fauna. J. Caribb. Ornithol.Wiley, J. W., & B.N. Wiley. 1979. The biology of the

white-crowned pigeon. Wildl. Monogr. 64.Wiley, J. W., N.F.R. Snyder and R. Gnam. 1991. Rein-

troduction as a conservation strategy for parrots.Pp. 165-200 in Beisinger, S., & N.F.R. Snyder(eds.). Crisis in New World Parrot Conservation.Washington, D.C.

Mirechi, D.N.,J.M. Hutton, C.M. Pannell, T.J. Stowe& R. W. Unite. 1977. ~port of the Cambridge Or-nithological Expedition to the British Virgin Is-lands 1976. Churchill College, Cambridge, U.K.

Murray, D. 1969. Birds of the Virgin Islands. Holly-wood, Florida.

Myers, J.G. 1935. Nesting associations of birds withsocial .nsects. Trans. Royal Ent. Soc. London 83:

11-22.National Climatic Center. 1974. Puerto Rico and

Virgin Islands. Climatological Data 20.National Climatic Center. 1975. Puerto Rico and

Virgin Islands. Climatological Data 21.Newton, A. 1859. Psittaci novae Speciei ad Conurum

Genus pertinentis Descriptio. Priv. publ. by author.Newton, A., & E. Newton. 1859. Observations on the

birds of St. Croix, West Indies, made betweenFebruary 2Oth and August 6th, 1857 by AlfredNewton, and, between March 4th and September28th, 1858 by Edward Newton. Ibis 1: 59-69,138-150, 252-264, 365-379.

Nichols, R.A. 1943. The breeding birds of St. Thomasand St. John, Virgin Islands. Mem. Soc. CubanaHist. Nat. Felipe Poey 17: 23-37.

Nicoll, M.J. 1904. On a collection of birds madeduring the cruise of the 'Valhalla', R.Y.S., in theWest Indies (1903-4). Ibis 4: 555-591.

Norton, R.L. 1983. West Indies region. Am. Birds. 37:

916-917.Norton, R.L. 1985. West Indies region. Am. Birds 39:

965-966.Pérez-Rivera, R. A. 1980. Clave para la identificaci6n de

Psittaciformes en Puerto Rico y algunos nuevosrecords. Science-Ciéncia 7: 59-62.

Pérez-Rivera, R.A., & M.J. Vélez. 1980. La prolifera-ci6n de Psittaciformes en Puerto Rico y el proble-ma que esos representan. Pp. 115-127 in VI Simpo-sio de Recursos Naturales, Puerto Rico Dept. Nat.Res., Puerto Rico.

Pielou, E.C. 1969. Mathematical ecology. New York.Raffaele, H.A. 1983. A guide to the birds of Puerto

Rico and the Virgin Islands. Fondo Educativo In-teramericano, San Juan, Puerto Rico.

Ricklefs, R.E. 1979. Ecology. 2nd ed. New York.Ridgway, R. 1916. The birds of North and Middle

America. Part VII. Bull. U.S. Nat. Mus. 50.Russ, K. 1895. The speaking parrots: a scientific

manual. London.Schafer, E., & w: H. Phelps. 1954. Las aves del Parque

Nacional "Henri Pittier" (Rancho Grande) y susfunciones ecologicas. Bol. Soc. Venez. Ciénc. Nat.

16: 3-167.

"4