( 127 ) 



evidently yonug, the feathers of the breast beiug still j)artly green, unci both these 

 have rosy red tips to the rectrices. The third ?, probably adnlt, although in moult, 

 has a few bine edges to the feathers of the interscapular region, and the tail 

 entirely black below, without any rosy tips. Of the six males four have rosy red 

 tips to the tail-feathers, and three of them have no blue on the interscapulium, one 

 some faint blue edges. One has no rosy tii)s and distinct bine edges on the inter- 

 scapulium, but one is deep blue from the lower hind-neck to the tail, though right in 

 the middle of the interscapulium the feathers have only blue borders. I thus come 

 to the conclusion that the development of blue is generally stronger in A. donalis 

 (lors'dis, and that .1. d. stilaeitsis is very seldom quite so blue above, but more fully 

 adult Sala birds should be studied. Wallace declares that the bill in the Sula 

 specimens is red, while it is blackish with a reddish orange spot near the base of 

 the culmen, bnt this statement is erroneous, as all our Sula skins have the bills 

 exactly like the A. dorsalis dorsalis, in which, however, the bUl is red in very 

 young birds. 



T. Tanygnathus muUeri (Miill. & Schleg.). 



Five specimens from iSnla Mangoli look like a distinct form at tirst sight, as 

 their mandibles are black. But it seems evident that this is due to the juice of 

 some fruit with which they have also soiled parts of their feathers and which has 

 coloured the mandibles entirely black. 



8. Trichoglossus flavoviridis Wall. 



A series from Sula Mangoli. " Iris bright orange, bare orbital space pale 

 orange ; feet pale dull lead-colour ; bill orange-red." 



9. Eudynamis facialis Wall. 

 In Froc. Zoo/. Sor. Lorn/. ISQ'^. p. 339, Wallace described as a new species 

 Ts. facialis, from a single specimen whicli differed from E. melanorhijnchus of 

 Celebes in its smaller size and different proportions, the wing being at least half an 

 inch shorter, the tail one and a half inch shorter, and also by its "chin and face, as 

 far as the eyes, being white." Some irregularity in the white facial markings, says 

 Wallace, " led my friend Dr. Sclater to suppose that it might be a case of incipient 

 albinism. This, however, would not affect the determination of the species, which 

 depends on its different proportions from its nearest ally." 



These observations of the great explorer are quite true. It is excusable that 

 Shelley, in Cat. B. Brit. Mns. XIX. p. 327, did not recognise Vi. /aa'a//,«, as only the 

 one specimen, Wallace's type, was known, but mention of the fact of its differing 

 should have been made. 



Doherty has now sent five adult mn.hs and one young female from Sula 

 Mangoli, and these specimens bear out fully the statements of Wallace. The wings 

 of the males measure 1S3 — 190 mm. (thoat^ oi J'J. melanorki//w/it!s 198 — 210) j the 

 tails 166 — 180 (those of E. melanorhynchns at least 200). Also the white facial 

 markings are not accidental, but due to locality, as three out of our five males show 

 them either indicated, or have the chin and forehead as far as tlie eyes white, like 

 the type, while in about twenty-live adult males of E. melaiiorli//iicki(s none has any 

 white at all. The iris of the adult male is described by Doherty as " orange-red ; 

 feet and bill black." The immature male had the " iris dull crimson ; feet and bill 

 brownish black." 



