( 137 ) 



3. Megapodius duperreyi forsteni Gray. 



Differs from M. duperrei/i duperreyi iu its dark olive-brown and not reddish 

 legs. The upperside is also darker, the under tail-coverts are not so rufous, but 

 more deep brown, the abdomen less greyish. This form is confined to the Southern 

 Moluccas, Coram, Amboina, lUirn, and Haraku. 



We have the skins of si.\ adults from Burn, collected by Doherty, Dumas, and 

 Bruijn's hunters. 



4. Megapodius duperreyi affinis Meyer. 



Very similar to M. duperreyi forsteni, but smaller, and the abdomen generally 

 more washed with rufons, as a rule somewhat lighter, more like M. duperreyi duperre.i/i 

 in colour, but with dark legs, Wka forsteni! Doherty notes ; " Iris chestnut, feet 

 almost black, bill dull olive-brown, darker at base." 



We have before us 10 adults and one chick collected on Jobi Island by William 

 Doherty and Bruijn's hunters. 



It must be remarked that the name affinis was given to specimens from llubi, 

 but we have not seen any from New Guinea, and it is therefore quite possible that 

 those from New Guinea again differ from the Jobi Island ones, in which case the 

 latter would require a new name. 



Specimens from d'Urville Island or Tarawai, M. decoUatus Oust., 1878, should 

 also be re-examined. It seems to us somewhat hazardous to unite decoUatus 

 with the Jobi Island birds without actual comparison. According to Oustalct 

 himself (1881) his decoUatus does not differ from the Rubi examples, the types of 

 aj/inis ; but actual comparison with Jobi ones does not seem to have been effected. 

 From the locality one might come to think that decoUatus would be the same as 

 brunneirentris rather than affinis. 



5. Megapodius duperreyi brunneiventris Meyer. 



We are fortunate iu having one skin (sex not stated) from the original locality, 

 Stephansort, Astrolabe Bay, Kaiser Wilhelm's Land, collected by the late Dr. Erik 

 Nyman. This specimen is quite like our " affinis " from Jobi Island, except that 

 the abdomen and outer aspect of the wings are a shade more rufous, and the wing 

 measures about 1 cm. more than those of our Jobi examples, which we provisionally 

 called affinis. It is quite possible that this form, if it is well separable, may have 

 to be called decoUatus. 



6. Megapodius duperreyi eremita Hartl. 



Differs from the previous forms by its almost bare reddish forehead and perhaps 

 smaller skull. The feathers of the crown are rather short, not forming a distinct 

 crest. Mr. Grant has united (Cat. B. Brit. Mus. XXII. p. 453) the forms from 

 the Bismarck Archijielago and the Solomon Islands under the name of eremita. 

 The type of eremita came from the Echiquier Islands, west-north-west from the 

 Admiralty grciup, and not from " Bougainville Island," as stated by Mr. Grant. It 

 is true that Hartlaub says the Echiquier Islands were also called Bougainville 

 Islands, but that is not Bougainville Island ! The Echiquier bird is .>iaid to be the 

 same as those from tlie Admiralty group and the Bismarck Archipelago, where 



10 



