b PROCEEDINGS OF THE NATIONAL MUSEUM vol.77 



Family TRERONIDAE 



FRUIT PIGEONS 

 2. DENDROPHASSA VERNANS PURPUREA (Gmelin) 



Columha purpurea Gmelin, Sys. Nat., vol. 1, pt. 2, 1789, p. 784 (Java). 



Two males, Pulo Maratua (Semut), May 10. 



These seem to agree with specimens from eastern Borneo, which 

 in turn do not seem to differ materially from males from Java. 



3. HAEMATAENA MELANOCEPHALA BANGUEYENSIS (Meyer) 

 Ptilopus hanguei/ensis Meyer, Journ. fiir Oni., 1891, p. 70 (Eanguey). 



One male, Pulo Pandjang, May 2. 



This specimen agrees fairly Avell with three males of H. m. 

 hangueyensis from Mindanao ; the crissum appears to be a little 

 darker yellow and the longer under tail-coverts are a trifle deeper 

 red, that is all ; the wing a little longer. 



The United States National Museum does not contain a specimen 

 from Banguey, but does possess a male from Cagayan Sulu; an 

 island in the Sulu Sea to the eastward. This specimen has been 

 miade the type of Haemataena melanocephala enantia^ but the char- 

 acters upon which it is founded are very slight. 



H. m. mslanospila from Celebes is smaller and the throat and 

 crissum a deeper yellow than FI. tn. hangueyensis. The Pandjang 

 specimen agrees with Celebes specimens in the color of the crissum, 

 but with those from Mindanao in the color of the throat, and in size 

 is even slightly larger. 



The present form does not seem to have been recorded so far south 

 before. 



4. MUSCADIVORES PICKERINGII PICKERINGII (Cassin) 



Carpophaga picker ingii Cassin, Proc. Acad. Nat. Sci. Philadelphia, vol. 7, 1855, 

 p. 228 (Mangsi, Sooloo Ids.). 



Four males and two females, Pulo Sangalaki, May 26 and 28 ; one 

 male, Pulo Maratua, August 28; one female, Pulo Bakungan, May 

 17; six males and four females, Pulo Bilang Bilangan, May 30- 

 June 3. 



There does not appear to be any difference between the specimens 

 from the various islands listed above. Compared with the type of 

 pickeHngii and a female from Cagayan Sulu the above series seems 

 to average a little lighter on the mantle and back, but the series of 

 picheHngii is so small it would be unsafe to separate them. As far 

 as the material before me goes, fickenngii is divisible into three 

 forms as follows : 



s Oberholser, Jouru. Washington Acad. Sci., vol. 14, 1924, p. 296. 



