Notes on the Red-billed Leiothrix in Hawaii HARVEY 1. FISHER and PAUL H. BALDWIN!

THE NAME JAPANESE HIL.L ROBIN often several small importations previous to this used in Hawaii for Leiothrix lutea (Scopoli) time (Table 1). These probably were made is an unfortunate misnomer, as is Pekin for cage purposes, but some of the long-time Nightingale, because the bird does not occur residents believe Leiothrix was established in in the wild in Japan nor does it occur as far the Hawaiian Islands before 1918 by escapees north as Shanghai, much less Pekin or Pei­ from cages. pmg. In Table 1 are summarized all importa­ Henshaw (1911: 63) recommended the tions of the species shown by official records. introduction of this species into Hawaii as It is worthy of note that the table does not early as 1911 and noted that it lives to some include the importations of 1918 and 1928 extent on insects and small fruits. Caum from San Francisco mentioned by Caum (1933: 38-39) states that the species was (lac. cit.). There may have been other un­ first introduced into the Territory of Hawaii recorded introductions. The table does show in 1918 by Mrs. Dora Isenberg, who liber­ a total of 41 importations, including some ated several birds on the island of Kauai. 1,037 individuals. These individuals have However, the records kept by the Territorial been released on all the larger islands of the Board of Agriculture and Forestry indicate southeastern end of the Hawaiian Chain; the islands are Kauai, Oahu, Molokai, Maui, and , Department of Zoology and Entomology, Uni­ versity of Hawaii, and Bernice P. Bishop Museum, Hawaii. Honolulu, respectively. From official records the only indication TABLE 1

PORTS OF EMBARKATION AND SUMMARY OF DATES AND NUMBERS OF Leiothrix IMPORTED INTO HAWAII

PORT OF EMBARKATION YEAR TOTAL Hongkong Sydney Shanghai Kobe Yokohama 1911 1 1 1913 2,6 3 11 1915 4 4 1916 5 2 7 1917 1,4,6,3 14 1918 2 2 1919 2, 12, 12 26 1922 1 1 1923 3 3 1924 2 2 1928 3,50 40,30,19 142 1929 6,184 190 1932 1 1 1933 7 8 1934 75 75 1935 28 28 1936 . 1,4, 1,80 200 100,37,36 459 1937 1 _.-_._-2 60 63 Totals 181 2 208 354 292 1,037

45 46 PACIFIC SCIENCE, January, 1947

we have of the original home of these intro­ Range and probably goes to about 4,000 feet duced birds is the name of the port from in the Waianae Mountains. In other·words, which they were shipped, and this tells its vertical distribution includes all elevations little. It was hoped that the subspecific deter­ available except those below 400 feet, and it mination of the Hawaiian birds might aid seems likely that the species might be found in revealing the native home of the intro­ below this level if all suitable cover condi­ duced stock. However, H. G. Deignan, of tions in deep valleys were to be investigated. the United States National Museum, has Although the altitudinal range is consider­ graciously examined 15 of our specimen.s able, the birds are most abundant above 600 from the Hawaiian Islands and has identified to 800 feet; this fact may be due to the lack them as Leiothrix lutea lutea. The type local­ of sufficient cover at lower elevations. . ity of this race is the mountains of Anhwei On Hawaii the bird has been found chiefly Province south of the Yangtze. Since lutea at elevations ranging from near sea level ranges over all of China south of the Yang­ (Puna district and windward Hawaii) to tze and east of the Kansu frontier, the origi­ 7,500 feet on Mauna Loa. We may assume nal home of the breeding stock now in the that the upward limit of their iivable range Hawaiian Islands cannot be further specified. on Hawaii is between 8,000 and 9,000 feet. However, there is no definite upper limit to DISTRIBUTION IN THE-HAWAIIAN ISLANDS the distribution in fall and early winter, Caum(1933: 39) noted that large flocks when the birds band together and wander all . were present on .Kauai and rightly assumed over the island. Flocks have frequently been that the birds. were breeding successfully on seen at more than 13,500 feet on Mauna Loa, that island. However, he was uncertain about but these flocks are soon reduced by deaths the status of the species on other islands, caused by exposure and starvation. simply saying that the birds were "reported" Temperature may be a factor in limiting to be breeding on Molokai, Maui, and Ha­ upward distribution; this is indicated by in­ waii and "believed" to be breeding on Oahu. creased mortality rates above 9,000 feet on Subsequent unpublished observations have Hawaii, where the mean temperature is 40° indicated that the birds are reproducing and to 50° F. with but little seasonal variation. are increasing the extent of their ranges· on However, it is not likely that temp€rature is Hawaii, Oahu, Molokai, Kauai, and Maui. the important factor below this altitude, and The bird is abundant, widespread, a'nd nest­ even above 9,000 feet, food, wind, and cover ing successfully on Hawaii. It is numerous conditions are probably the more important and nesting in the Koolau and Waianae limiting factors. mountain ranges on Oahu. In early June, 1945, adults were observed feeding nearly Throughout its range (vertical and hori­ fledged young in the mountain valleys at zontal) the species is limited to the more Mapulehu, Molokii, at an elevation of 500 moist, forested areas having a dense under­ feet. On Maui it is generally present in the story of plant growth. On Oahu this means moderately wet to wet forests of Haleakala it is restricted to the mountain valleys; sel­ and probably West Maui. dom is the bird observed on the higher ridges The vertical distribution on individual except during its movements from one valley islands is unrecorded except on Oahu, Molo­ to another. Despite the bird's proclivity foc kai, Maui, and Hawaii, where we have ob­ moist valleys, rainfall apparently is not the served the species. On Oahu the species d~termining factor except possibly in areas ranges from 400 to 3,000 feet in the Koolau having less than 40 inches a year, for yearly Red-billed Leiothrix in Hawaii - FISHER and BALDWIN 47

.averages of precipitation in its observed Here are scattered ohia lehua (Metrosideros range on Oahu vary from 30 to 200 inches. collina var. polymorpha) trees up to 12 and On Hawaii the species is rare or absent in 15 feet in height, interspersed with koa areas having less than 20 inches a year; (Acacia Koa) trees up to 30 feet high and this is true in the leeward districts: In the with kukui nut (Aleurites moluccana) trees. Kau district, where rainfall is 20 inches or These three kinds of trees form in many less in much of the lowland, Leiothrix is rare places a loose, open canopy. In certain areas below 3,000 feet and absent below 2,500 the kukui trees form a dense canopy. The feet. On Maui the annual rainfall in parts understory consists of ti plants (Cordyline of the' range of the species is 350 inches. terminalis) up to 8 feet in height, a few Presence or absence of surface water may Gardenia Remyi trees, and an occasional be a significant factor. The species has often white hibiscus (Hibiscus Arnottianus). The been observed bathing in water up to 2 inches lowest-growing vegetation is made up chiefly in depth. Further, the birds are most fre­ of palm foxtail (Setaria palmifolia) and oi quently observed within a few hundred yards shrubs (Stachytarpheta cayennensis). This of such water. We have on several occasions growth is in many places so dense that it is noted birds drinking rain water caught in difficult to penetrate. A further factor in basins formed by the large fallen leaves of creating an almost impenetrable thicket is the ti plant (Cordyline terminalis) and the the presence near the ground of many fallen kamani tree (Terminalia Catappa) . branches. Also, in a few places the inter­ twining branches of the hau tree (Hibiscus HABITAT IN THE HAWAIIAN ISLANDS tiliaceus) add to the maze (Fig. 1). A cover of dense vegetation near the ground is the major characteristic of the habi­ tat of Leiothrix. Without exception the birds are found in areas having such a gro,wth. It seems immaterial whether or not a canopy is formed above this undergrowth, as long as there is a thicket 10 to 15 feet in depth. We have never found this species feeding, nest­ ing, or even remaining long in the large plantings of imported ironwood (Casuarina) or exotic Eucalyptus; there is rarely any ground growth in such places. The birds are frequently found in guava (Psidium Gua- i.e java) thickets not having a high canopy. FIG. 1. Thicket formed by branches of the hau tree (Hibisctls tiliacellS) in Nuuanu Valley, Oahu, . On Oahu the species is most abundant in at 1,000 feet elevation. the dense understory of forested and par'tly forested areas on the floors and steep slopes On Hawaii this babbler is most abundant . of the valleys. Usually there is in this habitat in forests from 1,000 to 5,000 feet elevation a stream, a small pond, or some temporary and with 40, or more inches of rain each year. catch basins (fallen leaves, concavities in Forests with a variety oLhabitats, such as koa rocks) to hold rain water. On the extremely forest partly modified by cattle grazing and steep slopes of Manoa Valley, Oahu, at 1,100 with thimbleberries (Rubus rosaefolius) and feet where the rainfall is between 100 and other fruiting plants, harbor large numbers 150 inches annually is a "typical" habitat. of the species. The density reaches 80 to

.------..~----.----._------_..__...... _------48 PACIFIC SCIENCE, January, 1947

100 birds an acre at times at Kipuka Puaulu on Hawaii. However, some observers think (4,000 feet). Guava thickets at elevations the flocks represent groups migrating into of 1,000 to 2,000 feet are also tenanted in the lowlands for the winter. There is a good numbers. However, virgin tree fern movement at this time, but it is more of a forests favorable to the native Hawaiian general diffusion into peripheral areas not birds are in some areas without Leiothrix. A heavily occupied during the breeding season, forest of this type at the Twin Craters (3,800 and it is by no means limited to downward feet) in Hawaii National Park usually had. movements, for, as mentioned previ,ously, in not more than one individual per acre. Simi­ winter the birds appear conspicuously in the lar forests of the Upper Olaa Forest Reserve ohia lehua forest of Mauna Loa at high ele­ near a farming district exhibited a somewhat vations. greater density. By May at Kipuka Puaulu the bands have On Maui in the Kipahulu Valley the vir­ broken up, and most of the birds are segre­ gin, undisturbed ohia lehua forests (eleva­ gated by pairs. By late summer (July and tion 4,000 to 6,000 feet and rainfall 250 to August) small bands of 4 to 12 birds are 350 inches a year) had very few birds of this common. Consequently, it appears that flocks species. In koa forests in this same valley, are re-formed almost immediately after pa­ but between 2,000 and 4,000 feet elevation rental duties at the nest are completed. How­ and with a rainfall of 150 to 250 inches a ever, some of the birds remain in flocks year, the population density of this species throughout most of the breeding season was slightly greater, an'indication that the (March through June). bird was generally present, but uncommon. NESTING The birds are seldom seen more than 15 feet above the ground and then only when Perhaps nesting occurs at all elevations the ground cover is very. thick, or when the found in the spring and summer range of birds are in heavy tree foliage. Only occa­ Leiothrix, but on Oahu nests have been sionally can they be observed in the higher found only between 500 and 2,500 feet ele­ strata of the ohia and koa trees. Most fre­ vation. On Hawaii the highest elevation at quently the birds are heard in the low thickets which a nest was found was 6,100 feet at and not seen. If one is quiet it is possible to Kipuka Kulalio. Without exception on Oahu observe the bird flitting from shrub to shrub the nests have been less than 10 feet from next to the ground. These movements com­ the ground and located in the densest parts bine short fluttering flights and hops of a few of the understory; one nest was only 18 inches. inches above the ground. On Hawaii nests have been found from 3 to 7 feet from the FLOCKING ground. Choice of the nest site seems to We have seldom observed a flight of more depend not so much on the particular kind than approximately 50 feet, but others report of vegetation as on its density, for nests on that flights of 200 feet are not uncommon in Oahu have been found in staghorn fern the more open parts of the range of the (Dicranopteris iinearis), oi, palm foxtail, species. Longer flights are more frequently and guava, on the finer branches of low­ found during the winter, when the birds are growing hau trees, and attached to the stems banded together into traveling groups of 20 of ti leaves. On Hawaii, sites selected for or even 100 birds. Some of the flocks in nests included dense shrubs (aalii, Dodonaea favorable places seem to confine their move­ viscosa; pukeawe, Styphelia Tameiameiae) , ments to a limited area, as at Kipuka Puaulu the branches of trees (mamani, Sophora Red-billed Leiotprix in Hawaii - FISHER and BALDWIN 49

chrysophylla) , and fronds of tree ferns tied into the nest by small fibers (fibrovascu­ (Cibotium Chamissoi). lar bundles) from the leaves of the foxtail. Occupied nests have been found between However, the bulk of the nest was woven of March 3 and May 7 on Oahu, and from early longitudinally split leaves from which the March through June on Hawaii. It is likely, epidermis and mesophyll had disappeared, however, that further investigation on both leaving only 8 to 15 parallel fibers. The islands would reveal a longer breeding lining of the nest consisted of single fibers period. from the same plant; they were loosely laid On March 3, 1945, a nest with four eggs in to form a springy inside surface for the was found at 1,200 feet on the west side of nest. The only part of the nest not coming Manoa Valley, Oahu. A complete descrip­ from the leaf of this grass was three rootlets tion of this nest follows, because we have (the largest was 25 em. long and 1 mm. in been unable to find any published account of diameter) woven into the rim of the nest. the nest of this species. Thene,st was located The rootlets gave additional rigidity to the in a well-shaded spot beneath a koa tree and free edge. in a dense growth of the exotic palm foxtail. Because of certain differences in construc­ It was well concealed by the rank growth of tion and in materials used it seems desirable the foxtail and would have passed unnoticed· to describe another nest, found May 18, .~. had not the bird flushed when the observer 1940, at 6,100 feet at Kipuka Kulalio, stopped some 18 inches from the nest, which Mauna Loa, Ha:waii. It was suspended in an was 36 inches from the ground, aalii tree from the crotch of a forked twig Support for the nest was a fork near the at a height of 6~ feet from the ground and terminal twigs of an oi shrub (Stachytarpheta was secured. by grass stems bent over the cayennensis) ; the heaviest branch of the fork arms of the crotch and incorporated into the wall of the nest. The outermost part of the was only 2 mm. in diameter. The nest was nest. was a sling of grass woven vertically attached to the fork by only a few fibers from from the rim on one side, under the nest,to the leaves of the foxtail, which were wound the opposite rim. This was the sale support, around the twigs and woven into the rim of as there was no base of twigs. Inside this the nest. While not swinging as free, for sling, grass was woven in various directions example, as the nest of an oriole, the nest to make up the wall, which was about 3 em. was definitely pendulous. thick. Grass was worked in horizontally The measurements of the outside of the along the rim. There was a lining of loosely top of the ovoid nest were 9 em. by 11 em.; woven grass, strands of which ran diagonally corresponding dimensions of the inside were up the sides to bend down at the rim at 4.5 em. by 6 em. Over-all depth of the nest angles of about 80° and run again across the was 8.5 em., but the depth of the cavity was floor of the nest. The whol

GREATEST LEAST moved from the nest and could not be found DIAMETER DIAMETER DEPTH on the ground; apparently they were carried em. em. em. away by the parent. By 5: 35 P.M. the third Outside 11.5 9.0 9.0 Inside . 5.5 4.5 5.5 egg had broken open. The shell was broken in a ring just proximal to the largest diame­ EGGS ter. The crown of the embryo was pressed The egg is shiny, with an extremely pale­ against the cap of the shell. The hatching blue ground color and a ring of red-brown or bird flexed its neck and then extended it, purple splotches around the larger end. A shoving the shell cap forward and freeing few pale-brown splotches are found on the its head. After resting a few moments it blunt end and the sides of the egg, less often worked its legs back and forth twice, simul­ at the apex. Small brown specks and dark taneously kicking its body forward and the scrawlings may overlie the brown and purple shell backward. splotches, especially at the blunt end. Weights and measurements of eggs from DESCRIPTION AND DEVELOPMENT two clutches are given below. The weights OF THE YOUNG were taken 3 days before the eggs hatched. When the nestlings were 1 hour old they WEIGHT LENGTH DIAMETER appeared as follows: skin, rich reddish apri­ gm. mm. mm. cot, except where feather follicles caused gray Clutch 1 egg 1 . 2.4 20.9 16.5 color as in humeral tract, dorsal surface of egg 2 . 2.4 20.3 16.4 neck, upper part of wing, and mid-dorsal egg 3 . 3.0 21.3 16.5 region, and over eye; bill, reddish apricot, Clrttch 2 except upper mandible (gray between nares egg 1 . 20.4 15.4 and tip); egg tooth, whitish yellow; rictus, egg 2 . 20.3 15.5 egg 3 . 20.0. 15.4 light yellowish apricot; legs and feet, apricot; egg 4 . 20.9 15.6 claws, yellow. Down feathers were present The number of eggs observed in a clutch above eye, along occiput and mid-dorsal line varied between two and four, with an average from interscapular region to posterior edge of about three (two nests with two eggs, two of oil gland. with three eggs, and three with four eggs). When the young were 24 hours old the skin color was less reddish, and the gray HATCHING feather tracts were more pronounced and ex­ The earliest hatching observed on Hawaii tensive; plumules, absent in ventral tract; was on March 14, and the latest was on June eyes, still closed and gray; legs, yellow-apri­ 16. On Oahu these dates were March 10 and cot; skin, loose in tibiofibular region and on May 15. neck, tight over belly; belly, more prominent Cracks were. found in all three eggs of a and bulging than at hatching. clutch at 3:00 P.M. on May 22. The cracks When the birds were 7 days of age the appeared to have resulted· from activity in­ first primary feather was 15.5 mm. long and side the egg. In one egg the cracks were just still within its sheath; bill, orange at edges, distal to the largest diameter. A second egg gray on ridge and side; rictus, whitish; legs, had cracks in various places around its great­ salmon. est circumference, while the third had only When they were 11 days old the first one small hole immediately distal to the primary was 33.5 mm. long and two thirds greatest diameter. By'5 :00 P.M. two of the out of its sheath; central rectrices extended eggs had hatched. The shells had been re- 3 mm. and 1 mm. out of their sheaths; plu- Red-billed Leiothrix in Hawaii - FISHER and BALDWIN

mules, gone from back but present on crown matter varied from all animal to all plant and femoral tracts, sparse on crural tract; . food, but the proportions usually were be­ feathers forming fluffy covering over entire tween 40:60 and 60:40 by volume. Grit body. The primaries were lined with golden­ was typically present in the gizzard. yellow; breast, yellow-green; back, dark yel~ Observational data on Oahu indicate that low~green; ventral tracts, whitish yellow­ the birds also feed on the fruits of strawberry green; undercoverts of tail, yellow (still in guava (Psidium Cattleianum), overripe pa­ sheaths); secondaries, black and dark green. payas, petals of small flowers, and small new The bill' was dark salmon, but superficially buds of various plants. Leiothrix has also gray; rictus, whitish along edge; legs, yellow­ been seen to capture insects which flew near ish-tan; and claws, yellow-orange at tips. its position, but in feeding on insects it does The weights of the young are summarized: not fly out from a perch in typical flycatcher fashion. AGE SIBLING A SIBLING B SIBLING C BIRD MALARIA gm. gm. gm. I 'Blood smears from 11 specimens of Leio- 1 day 3.1 3.0 3.2 7 days. 13.0 13.9 13.5 thrix from Hawaii were examined by the 11 days. 15.0 15.2 U. S. Fish and Wildlife Service. Plasmodium When the birds were 1 hour old, their vaughani was reported from one smear made movements were sluggish; the head was from a bird taken near Kipuka Puaulu in lifted to receive food, and swallowing was Hawaii National Park. - accomplished. When 24 hours old the nest­ INTERACTION WITH OTHER ANIMALS lings could squirm around; they rested on Although at times Leiothrix seems to be a their abdomens. They threw back their heads noisy, aggressive bird, we found no antago­ and opened their mouths frequently. Beg­ nism toward other avian species except on ging behavior and ejection of feces could be one occasion. An observer was at a nest of brought about by touching the nestling. the species, and under the stress of excite­ When not active they lay curled up on their ment one of the adults drove a foraging sides with legs and toes flexed. At 7 days of Olive-green Creeper (Paroreomyza bairdi age they still were unable to crawl. At 11 mana) from the nest tree. days one sibling left the nest when the ob­ Except for the Hawaiian Hawk on the server tipped it to look inside. It escaped island of Hawaii, there is no evidence to into the bushes by moving along the ground indicate that other avian species molest Leio­ in short flights (3 feet). The nestlings could thrix in any way. Elsewhere in the Hawaiian perch well, and they shifted their footholds , Islands, the only possible predators on this to co1?pensate for movements of the nest. species are mammalian-the rat, the mon­ Low-pitched alarm notes were uttered when goose, wild dogs and cats, and feral hogs­ they were trying to escape capture. and these would exert pressure on the popu­ FOOD lation by destroying eggs and young. Animal matter found in the stomachs of REFERENCES 13 birds from Hawaii National Park came CAUM, EDWARD 1. The exotic birds of Hawaii. from various species of Hymenoptera, Dip­ Bernice P. Bishop Mus. Occas. Papers 10 (9). tera, Lepidoptera, and Mollusca. Plant ob­ 55 p .• 1933. jects included fruits of the thimbleberry HENSHAW, HENRY W. Report of the Committee on the Introduction of Birds into the Hawaiian (Rubus rosaefolius) and other fruits and Islands. Hawaii. Forester and Agr. 8: 61-64, stems. The proportion of animal to plant 1911.

.-._------, -, ___1-