588 Messrs. G. M. Mathews and T. Iredale on some 



On p. 103, with regard to P. tenuirostris brevicaudus, this 

 point of view was again empliasized. 



At that time ]Mathews was in receipt of the first con- 

 signment from Owston; but the second amply confirmed 

 his conclnsion, for it contained a form of Pvffirais griseus 

 breeding on the Pescadores Islands — a new record for 

 the east Pacific Ocean, and a new breeding record for the 

 north Pacific Ocean, because previously the only breeding 

 places known of this species were in the south of New 

 Zealand. 



On the same group, the Pescadores, P. leiicomelas and 

 P. cuneatus were procured ; while P. cuneatus was proved to 

 breed on tiie Bonin group, and another form of P. griseus 

 was received from the Kuril Islands. 



Consequently it must now be admitted that all the species 

 recorded from the north Pacific Ocean brted there, and 

 thus a field for investigation is indicated, the fruits of which 

 cannot be anticipated. 



The next most important discovery is the receipt from the 

 Bonin Islands of a new species of Puffinus, intermediate 

 between P. opisthomelas Cones { = auricular is Townsend) 

 from the coast of California and P. newelli Henshaw from the 

 Hawaiian group. It is a distinct species, but these are its 

 nearest allies, and as P. newelli Henshaw is almost extinct, 

 it is a most interesting addition to the group of the true 

 Pujjinus. These discoveries suggest that a careful search of 

 every rocky islet in the north Pacific may bring to light just 

 as unexpected forms, and that no dogmatic conclusions 

 regarding the distribution of Petrels can yet be attempted. 



We would emphasize how coloration may be a generic 

 character, or even supergeneric, in connection with Puffinus 

 and Pterodroma. 



These two genera were long distinguished by their appear- 

 ance, there being little structural difference to grasp ; yet 

 when once one of the species of the former genus was 

 examined and compared with one of those of the latter, 

 all the species could be easily separated into these two 

 groups, though variation in coloration, form, and size was 



