A REVIEW OF THE BIRDS OF THE ISLANDS OF SIBERUTAND SIPORA, MENTAWI GROUP ^ (SPOLIA MENTA-WIENSIA) By J. H. Riley Assista7it Curator, Division of Birds, United States National Museum Off the west coast of Sumatra lies a long chain of islands of vary-ing size, stretching from Simalur in the north to Engano in thesouth, a distance of over 700 miles. Dr. W. L. Abbott became inter-ested in this chain and in the autumn of 1901 visited Simalur, PuloBabi, Pulo Lasia, and the Banjak Islands, making collections on eachof them. The bird material of this collection was worked up by Dr.Charles W. Richmond, who published an account of it.- In theautumn of 1902, Doctor Abbott, accompanied by Mr. C. Boden Kloss,revisited Simalur, and then went to the Pagi and Batu Islands, andNias, the latter already known ornithologically from Modigliani'sexplorations; in the autumn of 1904 he visited Engano, and earlyin 1905 returned to Nias. Dr. H. C. Oberholser^ has published anaccount of the second Abbott collection from Simalur, but the col-lections from the other islands never have been reported upon as awhole, though Oberholser has described many new forms from them,in one paper* no less than 97, with scattered descriptions at subse-quent dates. In this early paper the descriptions are very brief andare unaccompanied by measurements or information concerning thenumber of specimens upon which his conclusions were based. Forthis reason it is utterly impossible for any subsequent worker to reachany definite conclusions with material from adjacent islands withoutconsulting topotypical material. ^ The present paper is one of a series under the general heading of " Spolia Menta-wiensia," dealing with collections in natural history made in 1924 in the Mentawi Islandsby Mr C. Boden Kloss and his assistants of the RaflBes Museum in Singapore. TheNational Museum is indebted to the friendly cooperation of Dr. W. L. Abbott for a shareIn the material secured during this worlt. ? Proc. U. S. Nat. Mus., vol. 26, 1903, pp. 485-524.3 Idem, vol. -55, 1919, pp. 473-498.* Smiths. Misc. Coll., vol. 60, no. 7, 1912, pp. 1-22.No. 2775.?Proceedings U. S. National Museum, Vol. 75, Art. 422482?29 1 1 2 PROCEEDINGS OF THE NATIONAL MUSEUM vol.75At the time Doctor Abbott was active in this region, he was unableto get permission from the Dutch authorities to visit Siberut andSipora, two islands of fair size lying north of the Pagi Islands, andwith them constituting the Mentawi Group.When, in the autumn of 1924, Mr. Kloss and a party from theRaffles Museum, Singapore, visited Siberut and Sipora to make ageneral natural history survey, Doctor Abbott contributed funds in ,aid of the enterprise, in return for which support the United States iNational Museum was to receive a set of the duplicates.After writing an account of the birds, in conjunction with Mr.F. N. Chasen,^ Mr. Kloss, the Director of the Raffles Museuni, for-warded the whole collection to Washington, with the exception of 'certain specimens which he took to London for further study, andrequested the writer to prepare a review of the collection after mak-ing comparisons with related material in Doctor Abbott's collections ifrom neighboring islands.A report upon the avifauna of the whole chain of islands off thewest coast of Sumatra would be welcome, but the writer does notwish to anticipate the results of Doctor Oberholser, who has beenengaged upon the birds of this general region for so many years.For this reason he has confined himself to the Mentawi Group alone,more particularly to Siberut and Sipora, and has mentioned relatedforms from other islands only when in his opinion this would leadto a better understanding of the Mentawi birds.Chasen and Kloss, in their paper (cited above), named 11 newsubspecies from Siberut and Sipora, three of which are not recog-nized in the present paper, but the writer ^ has described three addi-tional ones which are believed to be valid. Two of these are fromSiberut and Sipora, the other from Sipora and the Batu Islands.This arrangement leaves 10 forms confined to the two northern iislands, but in the Mentawi Group as a whole there are 32 peculiarforms.The islands off the west coast of Sumatra generally are closelyrelated in their avifauna to the mainland and Sumatra, as wouldnaturally be expected, with a few well-differentiated species andmany more or less well-marked races. As a rule, the forms fromSimalur seem to be more strongly characterized than those from theremainder of the chain, with the possible exception of Engano.Messrs. Chasen and Kloss are to be congratulated upon their suc-cessful trip and on their published ornithological results ; their paperhas left very little to be supplied here that is additional or new. s Ibis, 1926, pp. 269-306. ? Proc. Biol. Soc. Wasiiington, vol. 40, 1927. pp. 95, 96. ART. 4 BIEDS OF THE MENTAWI GROUP RILEY 6For an account of the physical features of Siberut and Sipora andother details the reader is referred to the " Ibis " paper mentionedabove. ANNOTATED LIST OF SPECIESTRERON CURVIROSTRA SMICRA OberholserTreion curmrontra smicra Oberholser, Smiths. Misc. Coll., vol. 60, no, 7,1912, p. 3 (Tana Bala Island, Batu Islands).Treron curvirostra Chasen and Kloss, Ibis, 1926, p. 273.Two males and one female, Sipora; six males and two females,Siberut.The Sipora and Siberut birds, compared with T. c. hypothapsinaof Engano, have the greens above and below much more yellowish,with the yellow on the greater wing coverts brighter.The type of Treron curvlrostra smicra of the Batu Islands is animmature male not in full plumage (the describer evidently had onlythe one specimen) ; it is brighter and more yellowish than the Enganobird, but not as yellowish below as the adult males from Sipora andSiberut, It approaches the female from Sipora and Siberut in colorof the underparts ; the head is washed with greenish, but is becominggray. It would be unsafe to separate the Sipora-Siberut birds untilwe know what the adult of typical smicra is like.Treron curvlrostra pega of Nias is quite distinct, very pale below,greenish glaucous, washed on the chest with light grape green witha buffy tinge, and without the yellowish tinge on fchest as in smicra;there are also other differences. The female is without the buffytinge on the chest, which is darker.Treron curvlrostra haliploa of Simalur resembles pega, but isdarker below and the chest is washed with grape green, without buffytinge. This form was evidently based on a single specimen.There thus appear to be four races of this species on the islands offthe west coast of Sumatra, as below : Treron curvirostra hypothapsina '^ from Engano.Treron curvirostra smicra,^ from Sipora, Siberut, and the BatuIslands.Treron curvirostra pega ^ from Nias.Treron curvlrostra haliploa ^ from Simalur.DENDROPHASSA VERNANS MESOCHLOA OberholserDendrophassa vemans mesochloa Oberholser, Smiths, Misc. Coll., vol. 60,no. 7, 1912, p. 2 (Nias Island).Treron vemans Chasen and Kloss, Ibis, 1926, p. 274.Two males, Sipora ; one male and one female, Siberut. ' Oberholser, Smiths. Misc. Colls., vol. 60, no. 7, 1912, p, 3. * Idem, p. 4. 4 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 75The specimens from Siberut and Sipora are intermediate betweenthe Nias form and that from the Pagi Islands, but on the whole nearerto that of Nias.Considering only those forms of the species described from theSumatran Avest coast islands, we have : Demlrophassa vemans mesochloa^ from Nias.Dendrophassa vemans folioftila^ from North Pagi.Dendrophassa vertmns iniza^ from Simalur.The males of D. v. miza are more greenish on the back and arelarger than the Nias bird. The wings of the males measure: 166,163, 160.5 mm., and of the female (type), 156 mm. The culmen(from cere) is also longer; in males, 12, 12.5, 11.5; in the female11 mm.The male of the Nias form is not so deep a green on the back and issmaller than the Simalur bird. Males measure, wing, 148.5, 157, 154,151, 151.5, 151, 154.5, 154 mm.; culmen (from cere), 10-11.5 mm.Females, wing, 152 (type), 148, 148; culmen (from cere), 10-10.5 mm.One male from Siberut measures, wing, 161 ; one female. 147.5 mm.Two males from Sipora have the wings 153 and 156 mm.A male from Tana Bala, Batu Islands, is here assigned to the Niasform, with wing 149 ; culmen from cere, 11 mm.Males from the Pagi Islands are slightly lighter below than thosefrom Nias, and this slight difference also holds in the females.There appears to be little or no difference in size, however. Thewings of males measure: North Pagi Island, 148, 151.5, 161, 162.5mm. ; culmen, from cere, 11-11.5 mm. Females from the same islandhave the wing 154.5, 155, 157.5, 155, 150 (type of polioptila) ;culmen, from cere, 10-11 mm. Males from South Pagi Island havethe wing 146, 149 ; culmen from cere, 10-11 mm.A male from Engano is like the Nias bird ; it measures, wing, 157 ; culmen, from cere, 11.5 mm.The color differences between Pagi and Nias birds are only aver-age and not constant ; some of the Pagi males can not be distinguishedfrom those of Nias. The differences between the females seem to bemore constant, but there are only three females in the Nias series,and they were taken later in the season.The single female from Siberut is greener below, with the yellowon the belly more restricted than in any specimen before me. andthe under tail coverts are very light, with a mere indication of chest-nut along the basal part of the shaft. The latter charaxjter is veryvariable and there are others in the series just as light or more so.It seems to me that no great violence would be done if polioptilawere united to mesochloa, in which case only two forms would " Obeiholser, Smiths. Misc. Coll., vol. 60, no. 7, 1912, p. 3. ART. 4 BIRDS OF THE MENTAWI GROUP RILEY Obe recognized from the islands off the west coast of Sumatra, asfollows : 1. Dendrophassa vernans mdza, confined, as far as known, to Sima-lur ; the males larger and more greenish below.2, Dendrophassa vernans mesochloa, extending from Nias to andincluding Engano; the males smaller and more j'ellowish below.Ij. v. polioptila becomes a synonym of this.MUSCADIVORES AENEUS VICINUS RileyMuscadivores aeneus vlcinus Riley, Proc. Biol. Soc. Washington, vol. 40,1927, p. 95 (Sipora Island).Ducula aenea Chasen and Ki-oss, Ibis, 1926, p. 274.One male, seven females, and one unsexed bird from Siberut; sixmales and two females from Sipora.The series submitted from the tw^o islands seem to be alike, asChasen and Kloss have remarked, and it only remains to comparethem with specimens from the other islands. Muscadivores aeneo-ihorax of Engano, on account of its differently colored under tailcoverts and other features, is so distinct that it need not be consid-ered further in this connection. Specimens from Simalur south tothe Pagi Islands are quite similar in color and only differ slightlyin degree or in size. Nias birds have very little vinaceous wash onthe breast, in certain lights not evident at all; it has been namedCarpophaga consohrina Salvadori,^? Birds from Pulo Babi and PuloLasia have been separated from consohy^lna on lai'ger size. Theyappear also to be slightly more vinaceous on the breast and hindneck and have been named Muscadivores consohnna habiensis Rich-mond.^' Simalur birds hav^e been separated on size from consohnna^as decidedly smaller, and the measurements show that they are some-what so. There is apparently little or no difference in color. Theyliave been named Muscadivores aeneus nvistu^ ^- Oberholser. TheSiberut-Sipora series is a puzzle. The breasts of these birds arewashed with deeper vinaceous; in one or more specimens the napeand hind neck is deep purplish vinaceous; in others the breast isvinaceous lilac, the head and nape gray like consohrina from Nias; inother specimens the vinaceous wash on the breast is not so pro-nounced. The specimen with the most pronounced deep purplishvinaceous nape and vinaceous lilac breast agrees with the descriptionof Carpophaga vandepolli Biittikofer.^^ except for some minor de-tails. If all the vinaceous-naped, vinaceous-breasted birds were simi-lar, one would be inclined to believe two species were involved, but ?? Ann. Mus. Civ. Geneva, ser. 2, vol. 4, 1887, p. 558." Proc. Biol. Soc. Washington, vol. 25, 1912, p. 103." Smiths. Misc. Coll., vol. 60, no. 7, 1912, p. 2." Notes Leyden Mus., vol. 17, 1896, p. 190. 6 PROCEEDINGS OF THE ISTATIONAL MUSEUM vor.. 75there is a perfect gradation from the darkest naped and breastedbird to the gray-naped and ahnost gray-breasted one. Regardless ofwhether the nape is vinaceoiis or gray, the series averages more vina-ceoiis on the breast and nape than consohrina from Nias. A seriesfrom Xortli and South Pagi contains none of the deep vinaceoiisnaped or breasted birds; it averages, however, more vinaceous onthe hind neck and breast than consohrina from Nias. Two malesfrom Tana Bala, Batu Islands, are like those from the Pagi Islands.I am inclined to think there are two phases of consohrina occurringoh Nias, corresponding to the vinaceous naped and breasted bird onSipora and Siberut, and that the Mentawi-Batu bird is a variableform distinct from consohrina.Summarizing, I will either have to unite all the birds from theislands off the west coast of Sumatra, except Engano, under onename, or recognize a number of forms. The differences between theforms is admittedly slight, but it seems better to recognize themrather than merge them into one variable race. This being the case,I think, considering the material before me, that the following formscan be made out : 1. Muscadivores aeneus mistus Oberholser (Simalur Island). Sixmales measure: Wing, 226-240 (229) mm.; five females, wing, 215-225 (222) mm.2. Muscadivores aeneus hahiensis Richmond (Pulo Babi and PuloLasia). Three males measure: Wing, 237-246 (243) mm.; twofemales, wing, 232-238 (235) mm.3. Muscadivores aeneus consohrinus (Salvadori) (Nias). Eightmales measure: Wing, 229-244 (236) mm.; two females, wing, 231-241 (236) mm. A male and a female from Pulo Tuanku, BanjakIslands, I am inclined to place here. The male has a wing 234 mm.,and the female 232 mm.4. Muscadivores aeneus mciwus Riley. (Batu and MentawiIslands). The wings of the males from the various islands repre-sented measure as follows : Tm'O, Tana Bala, Batu Islands, 227-232 (229.5 mm.).One, Siberut, 234 mm.Five, Sipora, 224-240 (231.5 mm.).Five, North Pagi, 215-245 (231 mm.).Six, South Pagi, 222-246 (22S mm.).The wings of the females : Seven, Siberut, 211-234 (226 mm.).Two, Sipora, 230-232 (231 mm.).Three, North Pagi, 220-228 (226 mm.).Four, South Pagi, 218-228 (222 mm.). ART. 4 BIRDS OF THE MENTAWI GROUP EILEY 7MYRISTICIVORA BICOLOR BICOLOR (Scopoli)Coluniba (iicolor) Scopoli, Del. Flor. Faun. Insub., vol. 2, 1786, p. 94,(New Guinea).Myristickyora bicolor Chasen and Kloss, Ibis, 1926, p. 275.Ten males and three females, Sipora.The United States National Miisemn also has it from Simalur,Pnlo Babi, and Engano.MACROPYGIA EMILIANA ELASSA OberholserMacropygia emiliana elassa Oberholser, Smiths. Misc. Coll., vol. 60, no. 7,1912, p. 2 (Sikakap Strait, North Pagi Id.).Macropygia phasianeUa Chasen and Kloss, Ibis, 1926, p. 275.Eiffht males and 7 females, Siberut; 10 males and 9 females,Sipora.The series secured on Siberut and Sipora is a very fine one, muchlarger than any series available from any of the neighboring islands.Comparing the females first, as the series of this sex available is morecomplete from the adjacent islands, it is to be remarked there is con-siderable variation both in measurements and color in specimensfrom the same locality. In some the tail is quite dark and the feathersbroad while in others the tail is lighter and rusty and the feathersnarrower. The specimens with rusty tails have the light bars onthe mantle yellowish buff, while in the dark-tailed birds these barsare rusty. All indications are that the light or rusty-tailed speci-mens are younger birds. Comparing birds of the same age fromSiberut with those from Sipora there appear to be no constant differ-ences in color and apparently none in size.The specimens from the Pagi Islands are similar in color to thosefrom Sipora. The series measured averages slightly smaller in thewing, however, but the measurements overlap and for the present,in my opinion, it is best to consider the Siberut, Sipora, and PagiIslands birds as belonging to the same form.Three females from Xias are more rusty on the back and tail, andthe black marking on the jugulum is barely indicated or entirelyabsent ; they average larger than Siberut females.One specimen, unsexed but almost certainly a female, from Sima-lur, the type of Macropygia emiliana hypopercna Oberholser ^* is adeeper brown below with the feathers of the wings more heavilymargined with lighter and rustier brown, and it is smaller than theNias bird. The males as far as they go bear out the above remarks. " Smiths. Misc. Coll., vol. 60, no. 7, p. 2. 8 PEOCEEDINGS OF THE NATIONAL MUSEUM VOL. 75From the above, in my opinion, we can recognize three forms atpresent from the islands off the west coast of Sumatra, as follows : (1) Maci'opygia eniiliana hypopercna Oberholser, confined to Sima-lur Island ; (2) MaxTOfygia emMiana modiglianii Salvador!, confined to NiasIsland; and(3) Macropygia emiliana elassa Oberholser, from Siberut, Sipora,North and South Pagi Islands. No specimens have been availablefrom Engano.The measurements (in millimeters) of the specimens from the dif-ferent islands are as follows : Eight males from SiberutTen males from SiporaTwo males from the Pagi IslandsOne male from NiasSix females, SiberutSeven females, SiporaFive females, Pagi IslandsThree females, NiasOne (female?), Simalur Wing5-191 -191 -184181 -184174. 5-187170 -176185 -187.174 168.177173170 Tail 166 -174153 -176161 -179152 -178173 Culmen9-1110-11.10-10.1110-119. 5-1210. 5-1110. 5-11.11.CHALCOPHAPS INDICA INDICA (Linnaeus)Columba indica Linnaeus, Syst. Nat., ed. 10, 1758, p. 164 {India orientali).Chalcophaps indica indica Chasen and Kloss, Ibis, 1926, p. 275.One female, Siberut Island; the United States National Museumalso has an immature female from North Pagi Island.RALLINA FASCIATA (Baffles)Rallus fasciatus Raffles, Trans. Linn. Soc. London, vol. 18, 1922, p. 328(Sumatra).Rallina fasciata Chasen and Kloss, Ibis, 1926, p. 276.One male, Sipora.The one male taken on Sipora does not seem to differ in size orcolor from specimens from other parts of the specific range. TheMuseum also contains a male from Engano Island; rather morehazel above and heavier barred below than usual, but it can bematched by a male from Trong, Lower Siam, so the differences aredoubtless individual.AMAURORNIS PHOENICURA CLEPTEA OberholserAmaurornis phoenicura cleptea Obebholsee, Smiths. Misc. Coll., vol. 60,No. 7, 1912, p. 2 (Mojeia River, Nias Island).Amaurornis phoenicura javanica Chasen and Kloss, Ibis, 1916, p. 276.One male, Siberut ; three adult males, one immature male, and fourfemales, Sipora. ART. 4 BIEDS OF THE MENTAWI GROUP RILEY 9The specimens from Siberut and Sipora, apparently do not differfrom birds from Nias and Simaliir. The pure white of the lowerabdomen and anal region given by the describer as one of the char-acters of this race does not hold, specimens with these parts as de-scribed and others having them tinged more or less strongly withIsabella color occurring in the above series. The race can be main-tained, however, as a sort of intermediate between chinensis andjavanica; smaller than the former and somewhat larger than thelatter.In my report upon a collection of birds from Celebes ^^ in com-menting upon this race I made a slip of the pen in saying A. p.cleptea is smaller than A. /;. javanica', it is actually larger. Sincethat paper was written the National Museum has received a numberof specimens of the Chinese race; all large birds.ARENARIA INTERPRES INTERPRES (Linnaeus)Tringa interpros Linnaeus, Syst. Nat., ed. 10, 1758, p. 148 (Islands ofGothland, Sweden).Arenaria interpres interpres Chasen and Kloss, Ibis, 1926, p. 276.One female, Sipora, October 24.PLUVIALIS DOMINICUS FULVUS (Gmclin)Charadrius fulvus Gmelin, Syst. Nat., vol. 1, pt. 2, 1789, p. 687 (Tahiti).Charadrius apricarius fulvus Chasen and Kloss, Ibis, 1926, p. 276.Two males and two females, Siberut, September 14r-20; one maleand one female, Sipora, October 27 and 29.CHARADRIUS LESCHENAULTII LessonCharadrhis leschenaultii Lesson, Diet. Sci. Nat. (Levrault), vol. 42, 1826,p. 36 (Pondichery, India).Charadrius leschenaulti Chasen and Kloss, Ibis, 1926, p. 276.One unsexed, Sipora, October 27.NUMENIUS PHAEOPUS PHAEOPUS (Linnaeus)Scolopax phaeopus Linnaeus, Syst. Nat., ed. 10, 1758. p. 146 (Sweden).Numcnius phaeopus Chasen and Kloss, Ibis, 1926, p. 276.One female, Siberut, September 12; one male, Sipora, Novem-ber 5.These two specimens have the white rumps of N. p. phaeopus, butseem to have longer bills; it is possible there are more than twoforms of the species.The culmen of the male measures 85 ; the female, 95 mm. "Proc. U. S. Nat. Mus., vol. 64, art. 16, 1024, p. 22. 10 PROCEEDINGS OF THE NATIONAL MUSEUM vol.75CAPELLA STENURA (Bonaparte)Scolopax stetiura Bonaparte Ann. Stor. Nat. Bologna, vol. 4, 1S30, p. 335(Sunda Islands).Capella stenura Chasejn and Kloss, Ibis, 1926, p. 277.One male, Sipora, October 23.GLAREOLA MALDIVARUM ForsterOlareola (Pratincola) maldivarum Forsteir, Faunula Indica, 1795, p. 11(Maldive Islands).Olareola maldivarum Ohasen and Kloss, Ibis, 1926, p. 277.One female, Sipora, October 26.HEMIGARZETTA EULOPHOTES (Swinhoe)Herodias eulopJiotes Swinhoe, Ibis, 1860, p. 64 (Amoy, China).Egretta eulophotes Chasen and Kloss, Ibis, 1926, p. 277.Chasen and Kloss record a male from Sipora, but the specimenwas not forwarded with the collection.DEMIGRETTA SACRA SACRA (Gmelin)Ardea sacra Gmelin, Syst. Nat., vol. 1, pt. 2, 1789, p. 640 (Tahiti).One unsexed, Sipora Island.The United States National Museum does not possess a sufficientnumber of specimens from the southern part of the range of thespecies to enable me to work out the races, if any, into which it canbe divided. The specimen submitted is quite dark, but it can bematched by other specimens from the northern part of the rangeand a male from the Paumotu Islands is equally dark. The Museumcontains a female from Simalur Island that is also dark and quitenear the Sipora bird.BUTORroES JAVANICUS SIPORA Chasen and KlossButorides striatus sipora Chasen and Kloss, Ibis, 1926, p. 277. (SiporaId.)Two adult females, Sipora ; one immature, Siberut.The above females resemble a male and female from Java verymuch. They are darker above and the white streak down the fore-neck is wider ; there appears to be little or no difference in size. Thetype of Butorides javanlcus actophilus, a female from North PagiIsland, is considerably lighter on the breast and sides of neck, haswider buffy margins to the wing-coverts and secondaries, and a con-siderably longer wing. The type of Butorides javanicurS icastopterusfrom Simalur Island is very close to the type of actophilus and aboutthe same size; it is darker on the breast. A specimen from Niasevidently belongs to icastopterus, if recognizable. Both B. j. actophi- ART. 4 BIRDS OF THE MENTAWI GROUP?RILEY 11lus and B. j. icastopterus are larger and lighter colored than B. j.javanicus. The immature of B. j. sipora is much darker than anyimmature with which I have been able to compare it. I have, how-ever, seen no immatures of B. j. javanicus.The wings of the various races mentioned above measure asfollows : A male from Java, 164 ; female, 166.Type of actophilus, North Pagi Id., 192.Two females, Sipora Id., 170, 172.Two males, Siraalur Id. (icastopterus), 191 (type), 180.One unsexed, Niasi Id., 177.The United States National Museum does not at present possess asufficient amount of material from the wide range of the species topermit me to work out the various forms into which it has been pro-posed to divide it. ? Some of the races have been separated on veryscanty material, indeed, but I prefer to leave the question in abeyance.SPILORNIS ELGINI SIPORA Chasen and KlossSpilornis elgini sipora Chasen aud Kloss, Ibis, 1926, p. 278 (Sipora Id.,W. Sumatra).One male and one female, Sipora.While the describers are probably correct in making this a formof /S. elgini and S. miniTnios a form of S. cheela, they are mistakenas regards S. ahhotti and S. klossi, since both are well-marked species./S. ahhotti is similar to S. e. sipora above, but the mantle has thefeathers edged narrowly with clay color and the feathers of the crestmore broadly tij^ped with black; below, S. ahhotti is quite different,ochraceous tawny (instead of mummy brown), the chest and upperbreast being distinctly barred with blackish, with the white spots onthe breast and belly smaller. S. ahhotti also is a much larger bird,having a wing in the male ranging from 328 to 360 mm.S. klossi Richmond of Great Nicobar Island is the most distinctlymarked species in the genus. Quite small and very pale. Swann ^?examined the series in the National Museum and gave a good con-densed description ; a fuller description was given by the describer.*^To make it a race of S. cheela is only to obscure its great distinctness.CUNCUMA LEUCGGASTER (Gmelin)Falco leucogaster Gmelin, Syst. Nat., vol. 1, 1788, p. 257 (New SouthWales).Cuncuma leucogaster Chasen and Kloss, Ibis, 1926, p. 279.One immature male, Sipora, October 26. " Syn. Accipitres, ed. 2, pt. 3, 1922, p. 137." Proc. U. S. Nat. Mus., vol. 25, 1902, p. 304. 12 PROCEEDIN^GS OF THE NATIONAL MUSEUM vol. 75PERNIS PTILORHYNCHUS PTILORHYNCHUS (Temminck)Faho ptilorhunchns Temminck, PI. Col., livr. 8, IMarch, 1821, pi. 44 (Javaaud Sumatra; later text to pi. 270, restricted to Java).Pernis aplvorus ptilorhynchus Chasen and Kloss, Ibis, 1926, p. 279.One male, Siberut, September 22.This specimen is in A'ery peculiar plumage. The head creamy-white; the center of the croAvn sayal brown, becoming darker on thenape, with a few very fine black shaft streaks; a broad loral streakto the eye bister mixed with white and with fine black shaft streaks ; a broacl postocular streak cinnamon; chin and throat light buff;breast cinnamon buft', deepening on the belly, lower flanks, andthighs to cinnamon, the breast with narrow sepia shaft streaks, thesestreaks becoming mere hair lines of a much lighter color on the re-mainer of the lower parts; the crest is rather short and black. Blan-ford ^^ describes similar specimens as the young. Like most hawksit probably goes through several stages before assuming the adultdress. OTUS BAKKAMOENA MENTAWI Chasen and KlossOtus bakkamoena mentatci Chasen and Kloss, Ibis, 1926, p. 279 (SiporaId., W. Sumatra).One adult male and one immature female, Siberut ; one adult maleand one immature female, Sipora.The above race has been compared by the original describers withO. b. lemplji of Java. A series of two males, four females, and oneimmature male (nearly adult) in the United States National Museumfrom Java is fairly uniform above (except some are slightly lighter,especially the immature) ; below there is quite a little variation, somehave the buff tone much deeper (clay color) grading to others withlittle buff at all, appearing grayish. The latter are probably youngerbirds, as the immature one is very light. The general effect aboveis grayish with a tawny-olive wash.Compared with the Javan bird, 0. h. mentawi is, as the describerssay, more deeply colored below and the other differences mentionedhy them hold good; but this form is also much darker above, theupper parts being washed with cinnamon brown.The two immatures are much lighter than the adults, one muchredder than the other.The type of Pisorhina umbra Richmond from Simalur Island ismuch smaller than O. h. mentaivi; tawny in color with the blackmarkings much reduced and almost lacking; the scapulars with alarge Avliite spot followed by a blackish one posteriorly on the outerweb; the belly with a few white crossbars. Wing, 143 mm. O. h. " Fauna Brit. India, vol. 3, 1895, p. 407. AKT. 4 BIRDS OP THE MENTAWI GROUP RILEY 13 riientami has the tarsus feathered almost to the base of the toes infront, ^Yhile O. umbra has the lower part of the tarsus bare for abouta quarter of its length. 0. wrribra is a very distinct species not veryclosely related to O. h. nnentakoi; on Engano, however, a larger formof O. unibra occurs, somewhat darker on the back and lighter on thelliroat. It has been named Otus umhra eiiganensis Riley."'Other localities represented by specimens of Otus hakkamoena inthe United States National Museum show great variation and thereare evidently a number of forms, but the material is not sufficient towork them out at present.PSITTINUS CYANURUS PONTIUS OberholserP.nttinus cymmrus pontius Oberholser, Smitlis. Misc. Coll , vol. 60, no. 7,1912, p. 5 (South Pagi Id. ) .?Chasen and Kloss, Ibis, 1926, p. 279.Three males and two females, Sipora ; three males, Siberut.Comparing specimens of the same age from Siberut and Siporawith those of South Pagi, there seems to be little or no difference.The majority of the South Pagi birds are more yellowish below, moreinclined to grayish on the mantle with crown not so deep a blue, butthis is due to age.Simalur Island is inhabited by the very distinct Psittinus abbottiRichmond.^*' LORICULUS GALGULUS GALGULUS (Linnaeus)Psittacus galgulus Linnaeus, Syst. Nat., ed. 10, 1758, p. 103 (India).Loriculus galgulus Chasen and Kloss, Ibis, 1926, p. 280.One male, one female, and one unsexed, Siberut; two males andthree females, Sipora.The specimens from Siberut and Sipora apparently do not differappreciably from birds from the mainland, Sumatra, and Borneo.Loriculus galgulus laniprochlorus Oberholser -^ from Nias is notdifferent enough to warrant recognition in my opinion. The type isan exceptionally small bird, but a young male from the same islandis as large as many males from Sumatra and Borneo. It apparentlydoes not differ in color. The type oi L. g. doUchoptei'us Oberholser ^^from Engano can not be matched by any specimen before me as tosize or color. It is larger than L. g. galgulus^ lighter on the back,more greenish rather than yellowish below ; and the forehead bluishglaucous. It measures: Wing, 88; tail, 34; culmen from cere. 11.5.A female from Sipora approaches it as to length of wing, but it isthe only specimen in quite an extensive series that does. The rac?was founded upon a single female; it apparently is a valid form. '? Proc. Biol. Soc. Washington, vol. 40, 1927, p. 93. =0 Idem, vol. 13, 1902, p. 188." Smitlis. Misc. Coll. vol. 60, no. 7, 1912, p. 5. 14 PROCEEDINGS OF THE NATIONAL MUSEUM vol.75EURYSTOMUS ORIENTALIS CALONYX SharpeEurystonms calonyx Shabpe, Proc. Zool. Soc. London, 1890, p. 551 (Nepal).Eurystamus orientalis calonyx Chasbn and Kloss, Ibis, 1926, p. 280.One male, Sipora, October 29.This specimen is quite dark, especially the head and upper back.It agrees more nearly with the form found in northeast China thanwith any of the other forms into which the species has been divided.It is evidently a bird of the year ; the maxilla is blackish, red only atthe base. The wing measurement given by Chasen and Kloss (281)is much too large; I make it, 195.RAMPHALCYON CAPENSIS ISOPTERA OberholserRamphalcyon capensis isoptera Obeeholsee, Proc. U. S. .Nat. Mus., vol 35,1909, p. 671 (Sikakap Strait, Pagi Islands).Rhamphalcyon capensis Chasen and Kloss, Ibis, 1926, p. 280.One adult male and two adult females, Siberut; six adult malesand one immature male, Sipora.The above series represents two phases, a dark-headed and a light-headed one, but I do not think they are different forms. Neitherphase seems to be confined to one island and the measurements showno difference. The series from the Pagi Islands show the twophases. The blue of the rump and back in the Siberut birds is alittle bit deeper than in those from Sipora, but the difference isvery slight. Taking the Siberut-Sipora series as a whole they agreerather closely with R. c. isoptera of the Pagi Islands, both in colorand size. The Siberut-Sipora series have a wing, 146.5-158.5; cul-men, 77.5-85 mm. The Pagi Island birds, wing, 148-159; culmen,77.5-87 mm.Ramfhalcyon capensis nesoeca Oberholser from Nias Island has alighter-colored head than isoptera.Ramphalcyon capensis sitnalurensis Richmond of Simalur is muchdarker below, with the mantle more brownish than in isoptera.Ramphalcyon capensis sodalis Richmond from Pulo Tuangku,Banjak Islands, is a larger, bluer-backed edition of simalurensis.So far as known, there are four more or less well-marked races ofthis kingfisher on the islands off the west coast of Sumatra fromSimalur south to the Pagi Islands, as given above. No form hasbeen described from Engano, the most isolated island of the group.ALCEDO ATTUIS BENGALENSIS GmelinAlcedo bengalensis Gmelin, Syst. Nat., vol. 1, pt. 1, 1788, p. 450 (Bengalen).Alcedo atthis bengalensis Chasen and Kloss, Ibis, 1926, p. 281 (Siberut).One female, Siberut, September 27.Probably a migrant. AET,4 BERDS OF THE MENTAWI GROUP?RILEY 15ALCEDO MENINTING PROXIMA RichmondAloedo menrnting proxima Richmond, Proc. Biol. Soc. Washington, vol.25, 1912, p. 104 (North Pagi Island).Alcedo meninting Chasen and Kloss, Ibis, 1926, p. 281.One adult male, Sipora.The above specimen is more greenish above and lighter below thanthe type of proxima; in the latter character it can be matched by afemale from North Pagi, but in the former there is no specimen inthe North Pagi series consisting of five specimens that matches it.Alcedo meninting callima and Alcedo meninting subviridis Ober-holser,^^ from the Batu Islands and Nias, respectively, agree in beingmuch deeper blue above than A. m. proxima. They are muchalike; the only difference I can detect is the apparently longer billof A. m. callima, but this presumed difference might disappear witha larger series. The series of A. m,. callima consists of three speci-mens from the Batu Islands; that of A. m,. subviHdis, three fromNias and one from Pulo Tuangku, Banjak Islands.CEYX RUrroORSUS RUFmORSUS StricklandCeyx rufidorsa Strickland, Proc. Zool. Soc. Loudon, 1846 (1847), p. 99(Malacca) ; cf. Hartert, Nov. Zool., vol. 9, 1902, p. 430.Ceyx rufidwsus rufidorsus Chasen and Kloss, Ibis, 1926, p. 281 (part).One male, Siberut; three males and one female, Sipora.The United States National Museum has a male and a femalefrom the Batu Islands (Tana Bala and Tana Masa).The above series seems to have more of the Chinese violet washabove when compared with the series in the Museum from variousislands in the South China Sea, Borneo, and Sumatra, though thefemale from Sipora is as red on the mantle as Bornean birds. Onemale from west Sumatra has as much of the purple wash as Siporabirds. The presence or intensity of the purple wash must be moreor less individual. There appears to be no difference in size.CEYX DILLWYNNI SharpeCeyx dillvyynni Shabpe, Proc. Zool. Soc. London, 1868, pp. 591, 593 (Labuan).Ceyx 7-ufidorsus rufidorsus Chasen and Kloss, Ibis, 1926, p. 281 (part).An unsexed adult specimen from Sipora, November 26, apparentlybelongs to this species; it is quite different from C. r. j'ufldorsuswith which it was taken on the same day. Below it is white, withan ochraceous orange band across the breast, the flanks of the samecolor, instead of the light cadmium of C. r. imfidorsus; the mantle isblack, the feathers rather broadly tipped with ochraceous orange. It 22 Smiths. Misc. Coll., vol. 60, no. 7, 1912, p. 7. 16 PROCEEDIlSrGS OF THE NATIONAL MUSEUM vol. 75may be not true G. dUlwynni at all, but it agrees fairly well withspecimeixs from Johore and East Sumatra, except it has more blackin the mantle and the scapulars are scarcely washed with blue atall. The tail is black, the inner webs of the feathers ochraceousorange; the latter may be an indication that the specimen is notfully adult. ENTOMOTHERA COROMANDA PAGANA OberholserEntomothera coromanda pagana 0BB2iH0LSEK, Proc. U. S. Nat. Mus., vol.48, 1915, p. 648 (North Pagi Id., W. Sumatra).Halycon coromanda minor Chasen and Kloss, Ibis, 1926, p. 281.One adult and one subadult female, Siberut; one adult male,Sipora.These seem to be identical with North Pagi specimens. E. c.neophora Oberholsei' from East Sumatra seems to be practicallyidentical in color with E. c. pagana, but the latter has a longerwing. E. c. minor from Borneo is darker and smaller than E. c.neophora. All these races have been founded on rather scantymaterial, but as they show some slight differences are at least worthyof provisional recognition.SAUROPATIS CHLORIS CHLOROPTERA OberholserSauropatis chloris chloroptera Oberholsek, Proc. U. S. Nat. Mus., vol. 55,1919, p. 379 (Sibabo Bay, Simalur Id.)Halcyon chloris chloroptera Chasen and Kloss, Ibis, 1926, p. 282.One female, Siberut Island.The single female submitted agrees with a series from SimalurIsland. Sauropatis chloris amphiryta Oberholser of Nias hardlydiffers in size or color from the Simalur bird and is hardly worthyof recognition. Specimens from Engano Island are much smallerthan S. c. chloroptera and represent a well-marked race, SauropatischloHs azela Oberholser.-^ We thus have two races on the islandsoff the west coast of Sumatra : (1) Sauropatis chloris cMoroptera from Simalur south to thePagi Islands; and(2) Sauropatis chloris azela, confined to Engano.If the Nias bird were recognized, it would make a break in therange of S. c. chloroptera which the slight differences assigned toamiphiryta by its describer and shown by the specimens does notwarrant. HALCYON CONCRETA (Temminck)Dacelo concreta Temminck, PI. Col., livr. 58, Maj-, 1825, pi. 346 (Sumatra).Halcyon concreta Chasen and Kloss, Ibis, 1926, p. 282.One immature male, Siberut. ? Proc. U. S. Nat. Mus., vol. 55, 1919, p. 377. AKT. 4 BIRDS OF THE MEISTTAWI GROUP RILEY 17There are no immature males in the National Museum with whichto compare it. It does not agree with Sharpe's description ^* of thisplumage, but resembles the female; the back is darker, duller green,and lacks the ochraceous spots; the feathers of the chest and breastare edged with blackish. There are a few dark-blue feathers of theadult plumage appearing on the back.HYDROCISSA CONVEXA CONVEXA (Temminck)Buceros convexus Temminck, PI. Col., livr. 89, February, 1832, pi. 530 (Javaand Sumatra).Anthracoccros coronatus convexus Chasen and Kloss, Ibis, 1926, p. 282.Two males and four females, Siberut and two males, Sipora.The National Museum contains four males from Nias, one male(marked female) and three females (unsexed) from North PagiIsland, two males. South Pagi Island, one male, Pulo Mansalar, W.Sumatra, and a good series from the mainland, the Tambelans, andBorneo, but only one male from Sumatra and one female from Java.AH the above apparently belong to one form and have a wingfrom 260 to 305 ; average 14 specimens 288 mm.The bird from Tana Bala, Batu Islands, has been separated asHydrocissa conveosa harussensis Oberholser ^^ and was based on twomales with a longer wing measurement. The specimens have wingsof the same length, 315 mm., but whether this difference Avould bemaintained in a larger series is problematical. It is undeniablygreater than in any male I have been able to measure from anyother part of the range of the species, but not as great as the varia-tion within Hydrocissa c. convexa.MEROPS PHILIPPINUS JAVANICUS HorsfieldMerops javanicus Horsfield, Trans. Linn. Soc. London, vol. 13, 1821, p.171 (Java).Mei'ops superciliosus javanicus Chasen and Kloss, Ibis, 1926, p. 283.Three males, three females, and one unsexed, Siberut, September16-October 2.The above series are in worn and faded condition; the backs andlower parts much mixed with blue. Merops philippinus philippinusof the Philippine Islands has the back and breast washed with buffy ; in the worn state, the buffy wash is accentuated. Specimens fromother parts of the range of the species have the breast more wavshedwith greenish, with little or no buffy suffusion. Siberut birds belongto the Javan race. The United States National Museum also hasspecimens of this race from Simalur Island, Batu Islands (TanaBala), and South Pagi Island. 2* Cat. Birds. Brit. Mus., vol. 17, 1892, p. 286.?Journ. Washington Acad. Scl., vol. 14, 1924, p. 300.22482?29 2 18 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 75I must confess that I can see no practical utility in making racesof forms distant geographically because of resemblances in plumage.It is in some cases positively misleading and obscures fundamentaldifferences. Two forms of the same genus that are widely separatedand differ quite markedly, even though there are certain resemblances,may have had quite different origins or have come from a commonancestor now extinct." For all practical purposes they are now spe-cies, with their own set of forms or not as the case may be. Species,at least in tire case of land birds, should have more or less of acontinuous distribution. Birds of the same genus widely separatedgeographically and easily distinguished had best be treated as species,even though they may resemble some distant form; the resemblanceis covered by the genus.COLLOCALIA VESTITA AEROPHILA OberholserCollocalia fuciphaga aerophila Oberholsek, Proc. U. S. Nat. Mus., vol. 42,1912, p. 16 (Siaba Bay, Nias Island). ? Chasen and Kloss, Ibis, 1926,p. 283.One male and two females, Sipora.These have been compared with the type of aerophila. They havea slight greenish gloss on the wings not seen in the type, but I candetect no other differences, and as the Nias bird was described froma single specimen it may not represent the form typically. A singlefemale in the Museum from Simalur Island has the wing a duller,almost glossless black and the sides of the face lighter than aerophila.It has been referred to C vestita vestifa by Oberholser,^*' but thisdisposition of it I am inclined to doubt is the correct one. I have beenunable to find the male specimen from the same island mentionedin the paper cited.COLLOCALIA LINCHI OBERHOLSERI StresemannCollocalia Unchi oierholseri Stbesemann, Nov. Zool., vol. 19, 1912, p. 348(North Pagi Island).?Chasen and Kloss, Ibis, 1926, p. 283.Two females, Sipora.These two specimens have shorter wings than any in the typicalseries from North Pagi Island, consisting of two males, one female,and two unsexed birds. The two males measure: Wing, 104, 106.5;tail, 43, 43. The female : Wing, 107.5 ; tail, 41. The two unsexed : Wing, 105, 106; tail, 42.5, 42.5. The two females, Sipora: Wing, 95,103 ; tail, 41, 40. There appear to be no constant differences in color.With a larger series from Sipora the supposed difference in thelength of wing would probably disappear. *> Proc. U. S. Nat. Mus., vol. 42, 1912, p. 16. ART. 4 BIRDS OF THE MENTAWI GROUP RILEY 19Collocalia Unchi linchi from Java, of which the Museum now hasa good series, is so very different in color from the races credited tothis species that it is very doubtful if they should continue to beregarded as forms of it. The Javan series stand out from the otherraces of the species in having the upper parts washed with dark ivygreen ; all the other races of the species, at least so far as representedin the United States National Museum, have the upper parts washedwith dull blue-green black. These colors are hard to match or define,but perfectly obvious to the eye. The differences are so pronouncedthat it would be better, from a practical standpoint, to recognizeC. I. linchi of Java as a distinct species without subspecies.HEMIPROCNE COMATA COMATA (Temminck)Cijpselus comatus Temminck, PI. Col., livr. 45, April, 1824, pi. 268 (Su-matra). ? Chasen and Kloss, Ibis, 1926, p. 283.Five males and four females, Siberut.Apparently not different from Sumatran specimens. The NationalMuseum also has it from Nias, Tana Bala (Batu Islands), and NorthPagi Island. HIEROCOCCYX FUGAX (Horsfield)Cueithis fugax Horsfield, Trans. Linn. See. London, vol. 13, 1821, p. 178( Java ) . Hierococcyx fugax Chasen and Kloss, Ibis, 1926, p. 283.Two immature birds, Siberut. These are, as Chasen and Klossremark, " too young to deal with subspecifically."CACOMANTIS MERULINUS THRENODES Cabanis and HeineCacomantis threnodes Cabanis and Heine, Mus. Hein., vol. 4, 1862, p. 19(Malacca).Cacomantis merulmus threnodes Chasen and Kloss, Ibis, 1926, p. 284.Four males and two females, Siberut.Only two of the above are fully adult. The United States NationalMuseum contains only one specimen from Malacca; it is deeper incolor on the breast and belly than the two adult males from Siberut,but Chasen and Kloss -" have compared Federated Malay States speci-mens with those from Siberut and found them not separable.Oberholser ^^ diagnosed the Nias Island form as Cacomantis Tneru-linus subpallldus, assigning as characters smaller size with head andlower parts paler. The character of smaller size does not hold, butit is paler than the specimens from Siberut. Whether the differencesare great enough to warrant recognition I am unable to dicide with-out examining additional material. It was founded on two males, "Ibis, 1926, p. 284. =8 Smiths. Misc. Coll., vol. 60, no. 7, 1912, p. 4. 20 PROCEEDINGS OF THE NATIONAL. MUSEUM vol.75the type not quite adult. They measure (the type first) : Wing,98-102; tail, 95-105; culmen, 19.5-19 mm. The two Siberut malesmeasure: Wing, 10^100; tail, 98.5-97; culmen, 18.5-18.5 mm.CHALCOCOCCYX XANTHORHYNCHUS (Horsfield)Cuculus xanthorhyncJius Hoksfield, Trans. Linn. Soc. London, vol. 13, 1821,p. 179 (Java).Chalcococcyx xanthorhynchus Chasen and Kloss, Ibis, 1926, p. 284.One male, Sipora.The United States National Museum contains only five specimensof this species. Three from Sumatra, one from Palawan, and onefrom Mindanao, Philippines, The three Sumatran specimens aresomewhat alike; all have a bronzy wash mixed with the violet ofthe upper parts. The Sipora specimen lacks the bronzy wash onthe upper parts and the violet comes near prune purple (blackishpurple in the Sumatran specimens) ; the bill is about the same sizeas the Sumatran birds. In color the Palawan bird is like that ofSipora, but the bill is much smaller. The specimen from Mindanaois somewhat intermediate between that of Sumatra and Palawan,but the bill is about the same size as in the Sumatran specimens.When sufficient material is available it Avill probably be found thatthe species breaks up into a number of more or less well-defined races.CENTROPUS SINENSIS BUBUTUS HorsfieldCentropus J}ul)utns Hoksfield, Trans. Linn. Soc. London, vol. 13, 1S21, p.ISO (Java).Centropus sinensis iuiutus Chasen and Kloss, Ibis, 1926, p. 284.One male, Siberut.The National Museum possesses only a male and female of thecoucal from Java; the female much more purplish on the hind-neckthan the male. The Siberut specimen resembles the Javan male,except the iridescence on the hind-neck is more bluish and the tailis more bronze green ; they are about the same size.Throe adult males and two unsexed adults from Nias Island aremore purplish on the hind-neck, but do not differ in size. The colorof the hind-neck seems to vary somewhat, probably due to age.UROCOCCYX AENEICAUDA (J. and E. Verreaux)Phoenicophaciis aeneicaudus J. and E. Verreaux, Rev. et Mag. Zool., 1855,p. 375 (Ceylon, error; type locality designated by Chasen and Kloss, Ibis,1926, p. 285, Sipora Island).Phoenicophaes curvirostris aeneicaudus Chasen and Kloss, Ibis, 1926,p. 2&1.Three males, six females, and one unsexed, Siberut; six males, sixfemales, and one unsexed, Sipora. ART. 4 BIRDS OF THE MENTAWI GROUP RILEY 21The specimens from the two islands do not seem to differ in anyway. The National Museum collection contains two males, onefemale, and one unsexed example from North Pagi Island and onemale and two unsexed birds from South Pagi Island. These ap-parently do not differ in color from the Siberut-Sipora series, butthe males apparently have a slightly longer yving. The nine malesfrom Siberut-Sipora have a Aving, 161-173.5 (average 166.5) ; threemales from the Pagi Islands, 175-180.5 (average 177.5) mm. Thisdifference, how^ever, might disappear with a larger series from thelatter locality.As Chason and Kloss have remarked -" aene'icmida has a differentlyshaped nostril from erijthrognathus (type of the genus UrococcyxShelle}') , and horneensis has another style different from either. Theyare all apparently closely related and it may be that the shape of thenostril has no generic significance in these birds.CALYPTOMENA VIRIDIS SIEERU Chasen and KlossCaJyptomcna viridis siberu Chasen and Kloss, Ibis, 1926, p. 285 (SiberutIsland).Two adult males and five immature males, Siberut.This appears to be a very good race. It is much darker, less yel-lowish green than C v. continentis; it also appears to be somewhatlarger. One immature male from North Pagi and three femalesfrom South Pagi in the National Museum probably belong to theSiberut form ; they have longer wings than birds from the mainland.I have been unable to compare C. v. sihej'u with C. v. viridis. ofSumatra, as this form is not represented at present in the NationalMuseum. PITTA MOLUCCENSIS MOLUCCENSIS (P. L. S. Mnller)Tiirdus moluccensis P. L. S. Mullee, Xatursyst. Suppl., 1776, p. 144(Moluccas, error; Tenasserim).Pitta hrachijura cyatwptera Chasen and Kloss, Ibis, 1926, p. 285.One unsexed bird, Sipora, Nov. 3.This specimen differs in the color of the rump from any I havebeen able to examine from Sumatra and the mainland. In thelatter the rump is deep dull violaceous blue, while in the Sipora birdit is Venetian blue. It does not differ in size from Sumatran speci-mens, and the color of the rump may be aberrant, hence I hesitateto name it.Pitta moluccensis lepta Oberholser ^? was founded upon two malesfrom Siaba Bay, Nias, and one female from Pulo Tuanku, BanjakIslands. The only characters, given are smaller size, especially the 2sibis, 1926, p. 285.50 Smiths. Misc. Coll., vol. 60, no. 7, 1912, p. 8. 22 PROCEEDINGS OF THE NATIONAL. MUSEUM vol 75bill. Five males and two females from east Sumatra measure:Wing, 119-130.5 (125) ; culmen, 26-29 (28) mm. The two malesfrom Nias: Wing, 118-119; culmen, 25-26 mm. The female fromPulo Tuanku: Wing, 117.5; culmen, 25 mm. There are apparentlyno constant color differences.The type of P. m. lepta ]ias blackish central strips on the feathersof the mantle and some of the middle wing coverts have a deep greencentral stripe. The Pulo Tuanku specimen i^ similar and I havefound a few mainland specimens in like condition; possibly an agecharacter. HIRUNDO RUSTICA GUTTURALIS ScopoliHirundo gutUiralis Scopoli, Del. F!or. et Faun. Insubr., vol. 2, 1786, p. 93(New Guinea).Hirundo rustica fftitturalis Chasen and Kloss, Ibis, 1926, p. 286.One female, Siberut, September 17.ARIZELOMYIA LATIROSTRIS LATIROSTRIS (Raffles)Muscicapa latirostris Raffles, Trans. Linn. Soc. London, vol. 13, 1821, p.312 (Sumatra).Alseonax latirostris Ckasen and Kloss, Ibis, 1926, p. 286.One male (so marked on label), Siberut.This is a very brown-backed bird, the chest with dusky streaks.The upper parts are browner than in any specimen with which I havebeen able to compare it. It is a bird of the year in fresh autumnalplumage which may account for the browner, more ferruginousplumage. MUSCITREA GRISOLA VANDEPOLLI (Finsch)Pachycephala vandepolU Finsch, Notes Leyden Mus., vol. 20, 1899, p. 224(Pulo Tello, Batu Islands).Muscitrea grisola nesiotis Oberholseb, Smiths. Misc. Coll., vol. 60, no. 7, 1912,p. 11 (Sibabo Bay, Simalur Island).Muscitrea grisola grisola Chasen and Kloss, Ibis, 1926, p. 287.One male and one female, Siberut,Muscitrea grisola 7iesiotis was founded upon a male and female andwhile they are browner on the upper parts and the head is duller,browner gray than in Muscitrea grisola grisola from the mainlandon an average, yet there are specimens of the latter that match themin this respect. The Simalur specimens have a larger bill, however.The description of Pachycephala vandepolU Finsch ^^ seems to fit itvery well, though the measurements given for the culmen are some-what smaller than those I obtain for the Simalur male, but thisapparent difference may be due to employing a different method ofmeasurement. ?i Notes Leyden Museum, vol. 20, 1899, p. 224. AET. 4 BIRDS OF THE MENTAWI GROUP RILEY 23The two Siberut specimens a^ee with those from Simalur in size,but are of a lighter, less rufescent brown on the back; they are inmore worn, less fresh plumage, however.There is a young female, in the United States National Museum(179947), from South Pagi Island which has the top of the headbrownish olive instead of deep mouse gray; back lighter and morerusty brown; secondaries outwardly edged with tawny; lower mandi-ble horn color, except at tip. These differences are due to age inmy opinion.Five males of M. g. grisola, measure: Wing, 82-87 (83.9) ; culmen, 13-15(13.9) mm.Type of M. g. nesiotis (Simalur) : Wing, 87; culmen, 15 mm.The male from Siberut : Wing, 85 ; culmen, 17.Two females of M. g. grisola: Wing, 77-83 ; culmen 13.5-14.One female from Simalur : Wing, 86 ; culmen, 15.One female from Siberut : Wing, 83 ; culmen, 15.One immature female from South Pagi : Wing, 79 ; culmen, 15.HYPOTHYMIS AZUREA LEUCOPHILA OberholserEypothymis azurea leucophila Obeeholser, Proc. U. S. Nat. Mus., vol. 39,1911, p. 607 (North Pagi Island, W. Sumatra).Eypothymis aznrea sipora Chasen and Kloss, Ibis, 1926, p. 287 (SiporaIsland, W. Sumatra).Five males and one female, Siberut; five males and three females,Sipora.The material submitted seems to be identical with the small seriesin the United States National Museum from the Pagi Islands onwhich the name leucophila was founded.On the west Sumatra islands the following species and forms arefound : Hypothymis aihotti Richmond, Pulo Babi and Pulo Lasia.nypothymis azurea consoirina Richmond, Simalur Island.Hypothymis azurea amelis Oberholser, Nias Island.Eypothymis azurea isocara Oberholser, Banjak Islands.Eypothymis aznrea ponera Oberholser, Batu Islands.Eypothymis azurea leucophila Oberholser, Mentawi Islands.Eypothymis azurea rlchmondi Oberholser, Engano Island.The Mentawi form is the only one in the above list with a whitebelly, and it is more nearly related to the subspecies occurring onSumatra. CULICICAPA CEYLONENSIS AMPHIALA OberholserCulicicapa ceylonensis amphiala Oberholser, Smiths. Misc. Coll., vol. QOrno. 7, 1912, p. 12 (North Pagi Island, W. Sumatra).Culicicapa ceylonensis percnocara Chasen and Kloss, Ibis, 1926, p. 288.Three majes, two females, and one unsexed skin, Siberut.The series agrees with two males (one the type) from North PagiIsland. 24 PEOCEEDIXGS OF THE NATIONAL MUSEUM Si "aoATwo specimens from Simalur Island {Culicicaqja ceylonensis perc-nocara Oberholser) and two from Nias Island {Gulicicapa ceylon-ensis pellonota Oberholser) described in the same paper with thePagi bird beIon-34 (32)."' Batu Islandsand Mentawi Islands (Siberut, Sipora, North and South Pagi).(5) Gracula javu)ia cni Salvadori. Somewhat smaller thanbatuensw the bill pronouncedly so; culmen, 25-30 (28).^*LAMPROCORAX PANAYENSIS PACHISTORHINUS OberholserLamprocorax chalyheus pachistorMnus Obebholsek, Smiths. Misc. Coll.,vol. 60, no. 7, 1912, p. 17 (South Pagi Island).Aplonis panai/ensis altirostris Chasen and Kloss, Ibis, 1926, p. 296.Seven males and three females, Siberut ; two males and fourfemales, Sipora.The United States National Museum possesses seven adult malesand three adult females from South Pagi Island. The birds fromSiberut, Sipora, and South Pagi Island apparently belong to oneform, but it is not the same as that from Nias Island {altirostris).The Mentawi Island subspecies has a larger, heavier bill and isslightly less glossy.Tlie specimens of this species in the United States National Mu-seum from the islands off the west coast of Sumatra fall naturallyinto three recognizable forms, to which is to be added a form de-scribed from Pulo Pinie not here represented. They are as follows:(1) Lamprocorax panayensis altirostris.Calornis altirostris Salvadori, Ann. Mus. Civ. Stor. Nat. Gen., ser. 2,vol. 4, 1887, p. 553, pi. 9, fig. 1 (Nias).Lamprocorax chalyheus rliacilnorhynchus Obekholsbir, Smiths. Misc. Coll.,vol. 60, no. 7, 1912, p. 17 (Simalur).Lamprocorax panayensis nesodramus Oberholsee, Jouru. Washington Acad.Sci., vol. 16, 1926, p. 516 (Pulo Babi).The Museum has specimens from Nias, Simalur, and Pulo Babi. s' Twenty-nine specimens.^ Eight specimens.22482?29 3 34 PROCEEDIXGS OF THE NATIONAL MUSEUM vol. 75This form is slightly smaller with a smaller bill than that fromthe Mentawi Islands. The birds from Nias, Simalur, and Piilo Babimeasure as follows:Nias. Three adult males : Wing, 96-103 (99) ; tail, 62-62.5 (63.8) ; culmen, 19. Four adult females: Wing, 97-104 (100) ; tail, 56-65.5(60); culmen, 17-18 (17.6).Simalur. One adult male (type of rhadinorhynchus) : Wing,101.5; tail, 60; culmen, 17.5. One adult female: Wing, 102.5; tail,62 : culmen, 17.5.PuLO Babi. Two adult females: Wing, 102-104; tail, 62.5-64;culmen, 17-18.(2) Lmiiprocorax panayensis pacMstorhinus Oberholser.Somewhat larger than altirosfrls, but less glossy and with a largerbill. Apparently confined to the Mentawi Islands.Fourteen males measure: Wing, 102-113 (106.3): tail, 62-69.5(65) ; culmen, 18.5-21.5 (19.8) ; and ten females, wing. 100-105(102.4) ; tail, 61-67.5 (63) ; culmen, 17.5-20.5 (19.3).(3) Lamprocorax panayensis enganensis.Calornis enganensis Salvadori, Ann. Mus. Civ. Storia Nat. Gen., ser. 2,vol. 12, 1892, p. 137 (Engano).Similar to pacMstorhinus^ but with a longer wing and tail; thebill smaller and weaker.Five adult males measure: Wing, 112-114 (112.7); tail, 70-77(72.9) ; culmen, 18-19 (18.3) ; and three adult females, wing, 105.5-114 (109.5) ; tail, 67-74.5 (70.5) ; culmen, 18.5-19 (18.7).(4) Lamprocorax panayensis leptorrhynchus.Aplonis panayensis leptorrhynchus Stresemann, Nov. Zool., vol. 20, 1913,p. 377 (Pulo Pini).Of this form the United States National Museum collection con-tains no specimens, but from the measurements given by the de-scriber it is as large or larger than enganensis. No length of culmenis given, unfortunately. It must be a recognizable form, however, asthe measurements would indicate a larger bird than altirostns orpachistorhiiws, and it could not very well be enganenms^ as the .Mentawi form would come in between.MOTACILLA CINEREA CASPICA (S. G. Gmelin)Parus caspicus S. G. Gmelin, Reise d. Russlaiid, vol. 3, 1774. p. 104. pi. 20,fig. 2 (Enzeli or En.seli, Caspian Sea).Motacilla cinerea caspica Chasen and Kloss, Ibis, 1926, p. 296.One male, one female, and one unsexed bird, Sipora, October 12-21. AKT. 4 BIEDS OF THE MEISTTAWI GROUP RILEY 35BUDYTES FLAVUS SIMILLIMUS (Hartcrt)Motacilla flava smdllima Habteet, Vugel Pal. Fauna, vol. 1, pt. 3, 1905,p. 289 (Kuiuchatka to Molucca.s, etc.) ; Nov. Zool., vol. 26, 1919, p. 167(type from Sulu Island). ? Chasen and Kloss, Ibis, 1926, p. 296.One female, Sibenit, September 20; one female, Sipora, October 11.The above two specimens are Youn<^ birds: they agree fairly wellwith northern Celebes specimens in the same stage of plumage.DENDRONANTHUS INDICUS (Gmelin)Motacilla indica Gmelin, Syst. Nat., vol. 1, pt. 2, 1789, p. 962 (India).Dendronanthus iudicus Chasen and Kloss, Ibis, 1926, p. 297.One male and two females, Siberut, -September 12-15 ; two malesand five females* Sipora, October 22-November 2.The United States National Museum possesses specimens from ?Simalur Island, October 25-November 30 ; Nias Island, February 20-March 25 ; Tana Bala, Batu Islands, February 5 and 6.CHALCOSTETHA CALCOSTETHA PAGICOLA OberholserChalcostctha calcostetha pagicola Oberholser, Smith.s. Mis. Coll., vol. 60. .10. 7. 1912, p. 17 (North Pagi Island).ChaJcostetha calcostetha calcostetha Chasen and Kloss, Ibis, 1926, p. 297(Sipora).Chalcostetha calcostetha siberu Chasen and Kloss, Ibis, 1926, p. 297(Siberut).Five males and four females, Sipora ; one male and two females,Siberut.The tAvo females from Siberut are paler than those from Sipora.but a female from North Pagi Island exactly matches them. Thereis no difference in size. I am inclined to think the deeper color ofthe Sipora females is due to the fresher condition of the plumage.They were collected in middle or late October, while the Siberutfemales were collected in middle or late September. One female in.the Sipora series, collected October 21, is somewhat worn and is .nlmost, if not quite, as pale as Siberut females. Specimens fromNias and the Batu Islands do not appear to differ from those fromthe Pagi Islands. If any form is to be named, it would be thatfrom Sipora; but this would divide the habitat of pagicola, so I donot believe it advisable to provide a name for it on the material athand.Specimens from Simalur Island are larger than fo-gicola^ especiallythe bills. It has been named Chalcostetha calcostetha heliomarptaOberholser.^**From the islands off the west coast of Sumatra there are tworecognizable forms, judging from the material before me, as follows : ^Journ. Washington Acad. Sci., vol. 13, 1923, p. 229. 36 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 75(1) Chalcostetha calcostetha heliomarpta Oberholser. SimalurIslaivl.(2) Chalcostetha calcostetha par/icola Oberholser. Nias Island;Piilo Pinie, Batu Islands; Mentawi Islands (Siberut, Sipom, Northand South Pagi Islands).AETHOPYGA SIPARAJA PHOTINA OberholserAethopyga siparaja photina OBEitHOL;!;Ei5, Siuiths. Misc. Coll., vol. 60, no. 7,1912, p. 18 (North Pagi Island).Aethopyga siparaja siparaja Chasen and Kloss, Ibis, 1926, p. 297 (SiporaIsland).Aethopyga siparaja siheru Chasen and Kloss, Ibis, 1926, p. 298 (SiberutIsland). *Fifteen males and 14 females, Sipora; 4 males, Siberut.This race is darker on the average than A. s. siparaja of Sumatra.Three out of the four males from Siberut have the feathers of therump washed with red, giving an orange effect to this region, but ^ohas one of the males from Sipora and the type of photina \q identical.I do not see, under the circumstances, any other alternative than tomerge siheru with photina.After studying and measuring the specimens of this species fromSumatra and the various islands off the west coa^t of Sumatra, Iwould reduce the recognizable forms from this region to three, asfollows : (1) Aethopyga siparaja siparaja (Raffles).Certhia siparaja Raffles, Trans. Linn. Soc. London, vol. 13, 1820, p. 299(West Sumatra).Aethopyrja siparaja tinoptila Oberholser, Smiths. Misc. Coll., vol. 60, no.7, 1912, p. 17 (Pulo Siumat, near Simalur Island).Aethopyga siparaja meJanetra Oberholser, Smiths. Misc. Coll., vol. 60, no. 7,1912, p. IS (Pulo Lasia).Aethopyga siparaja heliophiletica Oberholser, Journ. Washington Acad.Sci., vol. 13, 1923, p. 231 (Pulo Bangkaru, Banjak Islands).Range.?Sumatra, the Malay Peninsula,*" Simalur Island, PuloLasia, and the Banjak Islands.(2) Aethopj/gasiparajaniasensis'H.AiiTERT,OrmthMonsitsh., vol. 6,1898, p. 92 (Nias Island).A series of seven males are lighter red above than A. s. siparaja.Range.?Confined to Nias Island, so far as known.(3) Aethopyga siparaja photina Obekholser, as given above.Range.?Mentawi Islands (Siberut, Sipora. North Pagi, and SouthPagi ) . '"The range of this form on the mainland is uncertain and the bird there may provenot to be this form at all. ART. 4 BIRDS OF THE MENTAWI GROUP RILEY 37Allowing for the inequality of the various series, there does notappear to be any appreciable difference in size among the races.LEPTOCOMA BRASILIANA HYPOLAMPIS (Oberholaer)Cinnyris brasiliana hypolampia Obehholseb, Smiths. Misc. Coll., vol. 60, no. 7,1912, p. 19 (South Pagi Island).Leptocoma drasiliana brasiliana Chasbn and Kloss, Ibis, 1926, p. 298.Four males and one female, Sipora; one male and three females,Siberut.There does not seem to be any constant difference between speci-mens from the two above islands or from the Pagi Islands. Twomales from Java and two males from Singapore have noticeablysmaller bills. I am prepared to recognize three forms from theislands off the west coast of Sumatra as follows : (1) LeptocoTna hrasiliana rwecyTwrKyncha (Oberholser).Similar to L. h. hrasiliana but bill much larger. Culmen 16-15.Simalur Island.(2) Leptocoma hrasiliana oenopa (Oberholser).Similar to L. h. mecynorhyncha but bill smaller, though largerthan in L. b. hrasiliana, culmen, 14-15.5.Nias Island.(3) Leptocoma hrasiliana hypolampis (Oberholser).Top of the head more coppery, less golden, and the rump withless violet wash than in L. h. oenopa; there is no difference in size.A poorly marked form hardly worthy of recognition. MentawiIslands (Siberut, Sipora, North and South Pagi).ANTHREPTES MALACENSIS NASAEUS OberholserAnthreptes nialaceusis nasaeus Obbbholseb, Smiths, Misc. Coll., vol. 60, no.7, 1912, p. 20 (Sikakap Strait, Pagi Ids.).Anthreptes malacensis malacensis Chasen and Kloss, Ibis, 1926, p. 299.Nine adult males, four immature males, and four females, Siberut;one adult male, one immature male, and one female, Sipora.The above material agrees in size and color with the small typicalseries from the Pagi Islands. The Mentawi Islands form on theaverage has a longer and heavier bill, and the iridescent purpleedges to the feathers of the mantle are more pronounced than inAnthreptes malacensis malacensis. 38 PEOCEEDIIirGS OF THE NATIONAL MUSEUM vol. 75According to the material in the United States National Museum,three forms can be recognized from the islands off the west coast ofSumatra, as follows:(1) Anthreptes malacensis pelloptilus Oberholser.Bill larger and heavier than 7nalacensis\ the iridescent purple ofthe mantle more pronounced. Three males have the culmen, 18.5-19(18.8). Simalur Island.(2) Anthreptes malacensis pollostus Oberholser.Hardly different from fnalacensis; appears to have less iridescentpurple on the mantle. Not a well-marked form. Culmen, 16-17.5(16.8), Nias Island.(3) Anthi'eptes malacensis nasaeus Oberholser.More iridescent purple on the mantle than in inalacensis orpollostus \ bill slightly larger and heavier. Culmen, 17-18.5 (17.7).Mentawi Islands (Siberut, Sipora, North and South Pagi).ARACHNOTHERA LONGIROSTRA LONGIROSTRA (Latham)Certhia Iwigirostra Latham, Ind. Orn., vol. 1, 1790, p. 299 (Bengal, Sylhet).Arachnothera longirostra exochra OsEaiHOLSER, Smiths. Misc. Coll., vol. 60,no. 7, 1912, p. 19 (South Pagi Island).Arachnothera longirostra hypochra Obebholseb, Smiths. Misc. Coll., vol. 60,no. 7, 1912, p. 19 (North Pagi Island).Arachnothera longirostra Chasen and Kloss, Ibis, 1926, p. 299.Five males, four females, and one unsexed bird, Siberut ; ten males,three females, and one without sex, Sipora.There seems to be little or no difference in size between the seriesfrom Siberut and Sipora and that from North and South PagiIslands, and the difference in color, if any, is negligible. A smjillseries from the mainland, north of Singapore, and two males fromeast Sumatra seem to agree in size and color with the series from theMentawi Islands. A few specimens from near . Singapore averagesmaller and represent a different form. It has been named Arachno-thera I. heliocrita Oberholser.*^Three specimens from Pulo Bankaru, Banjak Islands, have longerbills, especially the type. It has been named Arachnothera longiros-tra zarhina Oberholser.'*^Two specimens from Nias Island are paler than any I have ex-amined and the bills are longer than the average, but not quite as " Journ. Washington Acad. Sci., vol. 13, 1923, p. 227? Smiths. Misc. Coll., vol. 60, no. 7. 1912, p. 19. AKT. 4 BIRDS OF THE MENTAWI GROUP?EILEY 39long: as the type of zarhina. It has been named Arachnothera I.niasensis van Oort.^^There is quite a sexual difference in size, the females having consid-erably shorter wings and bills, and it is probable that abnormallysmall birds marked as males may be wrongly sexed.Seven males (including type of antelia) from Tenasserim and Trong,measure: Wing, 67.5-70 (68.6) ; eulmen, 37-41 (38.1).Two males, (including type of melancluma) from east Sumatra measure:Wing, 70-71.5 (70.7) ; eulmen, 39-39.5 (39.2). ical chrysogenys.ArachnotJiera chrysogenys astilpim Oberholser *^ and Arachiiotherarhi'nsogeriys intensifava Stuart Baker*'' were both described fromTenasserim and are probabl}'^ synonymous; the former name shouldbe used for this form, being the older.DICAEUM CRUENTATUM BATUENSE RichmondDlcaeum sumatraniim hafuense Richmond, Proc. Biol. Soc. Washington,vol. 25, 1912, p. 104 (Pnlo Pinie. Batu Islands).Dlcaeum cruentatiwi l>attie}ise Chasen and Kr.oss, Ibis, 1926, p. 300.Five males and one female, Sipora.The above series agrees with the type of 'batv?nse. The UnitedStates National Museum possesses an immature male of apparentlythe same form, from South Pagi Island. The chin and throat areneutral gray, with a few scattered white feathers coming in on thechin; the upper parts dark grayish olive, the crown, mantle, andrump with a few black scarlet tipped feathers appearing; other-wise it is like the adult. Already the white band down the center ofthe lower parts has progressed further forward than in D. c.suniatranum.An adult male, collected by Dr. W. L. Abbott on Nias Island(180,07-1), apparently of D. c. sumatramirm, has a somewhat largerbill than D. c. hatuense and the throat, chin, and foreneck are neutralgray. In hatuense there is a rather broad band of ivory whiteextending to and including th^ chin.The five males measure: Wing, 44.5-49 (45.8) ; culmen, 9-10 (9.8). Thetype of batuense measures : Wing, 45.5 ; culmen, 10. The Nias specimenof suniatranum: Wing. 48.5 ; culmen, 10.5.DICAEUM TRIGONOSTIGMUM PAGENSE OberholserDlcaeum triyonostlgmum pagcnse Oberholser, Journ. Washington Acad.Sci.. vol. 16, 1926, p. 520 (South Pagi Id.).Dlcaeum triyouostigma Chasen and Klo.ss, Ibis, 1926, p. 300.Seven adult males, two immature males, and three females, Siberut;seven adult males and eight females, Sipora. *6 Journ. Washington Acad. Sci., vol. 13, 1923, p. 227.^"Bull. Brit. Oin. Club, vol. 46, 1925, p. 14. AltT. 4 BIEDS OF THE MENTAWI GROUP PJLEY 41The above series and specimens from South Pagi Island appear tobe identical.The Nias Island form has been named D. t. lyprum by Oberholser.*^AVith it I would place a male from Pulo Pinie, Batu Islands, whichappears to have a somewhat lighter throat.Simaliir Island specimens are larger than those from Nias or theMentawi Islands and the female is quite different from that of anyother form examined. It is darker, and grayer above with very littleolive green wash and the rump is only crossed by a narrow band ofyellowish citrine; below it is grayer, the breast with only a narrowband of barium yellow down the center. It has been named Dicaeimiiririonostigtna antioprochmi Oberholser.*" Dicaeuni trigonostigmamelcmthe Oberholser from Pulo Lasia was evidently founded upon asingle male. It agrees with Simalur Island specimens in size; incolor it is slightly darker above and on the throat. The differencesare so very slight that in my opinion it should be merged with theSimalur form.Summarizing, I am prepared to recognize three subspecies from thewest Sumatra islands as follows : . t(1) Dicaeuni trigonostigmum pagense Oberbolser. Mentawi Islands(Siberut, Sipora, South Pagi).(2) Dicaeum trigonostigmum lyprum Oberholser. Nias Island and PuloPinie, Batu Islands.(3) Dicaeum trigonostigmum antioproctuni Oberholser. Simalur Islandand Pulo Lasia.Outside of the Simalur Island form, which is characterized like somany of the races from this island by superior size, the other twosubspecies are not strikingly different in color or size from D. t.tr'igonostiginmih of the mainland, and if it were not for their insularhabitat would hardly be worthy of recognition. The only differencesbetween the mainland form and those from the islands being thesomewhat darker upper parts and the more yellow, less orange washon the rump of the latter.Check list of Mentawi birds Family TRERONIDAE1. Treron curvirostra smicra Oberholser2. Dendrophassa vernans mesochloa Ober-holser3. Muscadivores aeneus vicinus Riley4. Myristicivora bicolor bicolor (Scopoli) *' Smiths. Misc. Coll., vol. 60, no. 7, 1912, p. 21. Siberut XXX Sipora NorthPagi South PagiIslands X 42 PROCEEDINGS OF THE NATIONAL MUSEUMCheck list of Mentawi birds?Continued ai:t. 4 BIRDS OF THE MENTAWI GROUP RILEYCheck list of Mentawi birds?Continued 43 44 PROCEEDIli^GS OF THE NATIONALi MUSEUMCheck list of Mentawi birds?Continued VOL 75 Siberut Sipora NorthPagi SouthPagiFamily PITTIDAE55. Pitta mohtccensis moluccensis (P. L. S.Miiller)Family HIRUNDINIDAE56. Hirundo rustica gutturalis Scopoli.Family MUSCICAPIDAE57. Arizelomyia latirostris latirostris (Raffles)58. Muscitrea grisola vandepolli (Finsch)59. Hypothymis azurea leucophila Ober-holser60. Culicicapa ceylonensis amphiala Ober-holserFamily CAMPEPHAGIDAE61. Graucalus sumatrensis crissalis Salva-dori62. Volvocivora culminata culminata (A.Hay) Family IRENIDAE63 . Irena puella crinigera SharpeFamily PYCNONOTIDAE64. Brachypodius atriceps chrysophorus(Oberholser)65. Microtarsus melanoleucos proximus Riley.66. Pycnonotus plumosus poTphyreus Ober-holserFamily TURDIDAE67. Copsychus saularis pagiensis Richmond.68. Copsychus saularis subspecies?69. Kittacincla melanura SalvadoriFamily SYLVIIDAE70. Orthotomus sepium ochrommatus Ober-holser71. Orthotomus septum concinnus RileyFamily LANIIDAE72. Lanius tigrinus DrapiezFamily CORVIDAE73. Corvus enca enca (Horsfield).. XXXXX XXX XXX XXX X X X X XXXx' XXX XXx' X X X X ART. 4 BIEDS OF THE MENTAWI GROUP RILEY 45 74.75.76. 77.78.79.80. 81. 82.83. 84.85.86. 87.88.89.90.91.92.93. 94.95. Family DICRURIDAEDicruropsis borneensis viridinitens Sal-vadoriDicrunis leucogenis periophthalmicus(Salvadori)Dicrurus leucogenis siberu Chasen andKlossFamily ORIOLIDAEOriolus chinensis sipora Chasen andKlossOriolus chinensis siberu Chasen andKlossOriolus chinensis richmondi Oberholser..Oriolus xanthonoius mentnwi Chasen andKlossFamily STURNIDAEAgropsar siurninus (Pallas)Family GRACULIDAE.Gracula javana batnensis Finsch.Lamprocorax panayensis pachistorhinusOberholserFamily MOTACILLIDAEMotacilla cinerea caspica (S. G. Gmelin).Budytes flavus simillimus (Hartert)Dendronanthus indicus (Gmelin)Family NECTARINIIDAEChalcostetha calcostetha pagicola Ober-holserAethopyga siparaja photina Oberholser..Cyrtostoinus ornatus subspecies*^ ?Leptocoma brasiliana hypolampis (Ober-holser)Anthreptes malacensis nasaeus Ober-holserArachnothera longirostra longirostra (La-tham)Arachnothera chrysogenys isopega Ober-holserFamily DICAEIDAEDicaeum cruentatum batuense RichmondDicaeum trigonostigmum pagense Ober-holser Siberut XX XX XXX X Sipora X XXXX XXXXXX NorthPagi X South X XXXX XX PagiIslands X " Chasen and Kloss, Ibis, 1926, p. 305, report this for the Pagi Islands under the name Leptocoma jugular isornata. It can hardly be ornata (type locality Java), however. Doctor Abbott did not obtain it in thePagi Islands o