Vol. xl. ] 66 
that the blue-necked ones were young and the yellow- 
cheeked, yellow-naped ones were the fully adult birds. This 
was very natural, as they apparently had No genuine young 
birds which have a green occiput and hind-neck. The ease, 
however, of the two Birds-of-Paradise P. apoda noveguinee 
and P. raggiana which they themselves, in this same article, 
show to hybridise on the Fly River, should have made them 
hesitate, although perhaps they were not to be blamed, 
seeing that the facts concerning the occurrence of the yellow- 
cheeked yellow-naped form west of the Fly River were not 
then known and there were too few specimens from east of 
the ly River in collections to afford any proof. When 
Mr. Grant described his Cyclopsitta godmani in 1911, and 
Mr. Hartert and I, by an oversight, redescribed the same 
bird in 1913 as Opopsitta blythi meeki, we entirely overlooked 
these references of Salvadori and D’Albertis and also the 
plate in Gould’s ‘ Birds of New Guinea,’ or else we could not 
have failed to realise that the yellow-cheeked, yellow-naped 
bird which Salvadori and D’Albertis determined as the adult 
of their cervicalis was our new bird. Also Mr. Grant ought 
from the description to have seen that Van Oort’s 2 desma- 
resti cervicalis was his godmant. 
We have now aseries of cervicalis from various parts east 
of the Fly River, and at least 14 godmani from 8.W. Dutch 
New Guinea, and the facts are as follows:—Hast of the 
Fly River there are only found adult birds with blue occiput 
and hind-neck, true cervicalis Salv. & D’Alb., and the young 
of this has a green occiput and hind-neck ; while West of the 
Fly River there occur only birds with orange-red occiput and 
yellow hind-neck, which are godmani Grant and of which the 
young have the occiput duller orange and the hind-neck green. 
On the Fly River, as we have seen, all sorts of inter- 
gradations occur and are evidently the result of the two 
forms interbreeding where they meet. 
Now, however, arises the question of the status of godmant 
and cerviealis. If we treat them as two subspecies the 
question at once presents itself: if there are two subspecies, 
are they not also subspecies of desmarestz ? Opposed to 
