404 
Mr. D. G. Elliot on the Trochilidse. 
the specimens given above; and there is also a difference in 
the length of wing between specimens from the Rio Napo. 
It is therefore very evident that size, upon which the sup¬ 
posed species was given a distinctive rank, is not to be re¬ 
lied on any more than is a particular locality; and therefore 
it is impossible to keep the two separate, as there is not a 
single other character to distinguish them apart. And I the 
more readily relegate P. iolata to the condition of a synonym, 
as I do not understand wherein the study of ornithology is 
advanced by the creation of species so closely related that 
their distinctness cannot be discerned save by the assistance 
of a pair of compasses. 
Petasophora thalassina. 
Trochilus thalassinus , Swains. Phil. Mag. 1827, p. 441. 
Petasophora thalassina , Gould, Mon. Troch. vol. iv. pi. 227; 
id. Intr. Troch. (8vo ed.) p. 125. sp. 252. 
Hah . Mexico, Guatemala. 
A clearly marked species, most abundant in Mexico, but 
extending its range into Guatemala. Resembling somewhat 
in coloration P. anais , it is nevertheless a very much smaller 
bird, and cannot be confounded with that species. 
-|-Petasophora cyanotis. 
Trochilus cyanotis, Bourc. Rev. Zool. 1843, p. 101. 
Petasophora cabanidis, Heine, Journ. fur Orn. 1863, 
p. 182. 
Petasophora cyanotis , Gould, Mon. Troch. vol. iv. pi. 228; 
id. Intr. Troch. (8vo ed.) p. 125. sp. 253. 
Petasophora cabanisii , Lawr. Ann. Lye. Nat. Hist. N. Y. 
vol. ix. (1868) p. 126. 
Hah. Venezuela, Costa Rica, Veragua, Columbia, Ecuador, 
Peru. 
A species with a very wide range, and apparently gene¬ 
rally distributed in the different countries mentioned above. 
The birds from Costa Rica, for which Heine proposed the 
name of cabanidis (l>c.), are, in my opinion, not specifically 
distinct; and I have therefore placed this name as a syn¬ 
onym. 
