\ 
780 The American Naturalist. [September, 
sion I reached the 16th of August, 1891, on James Island, and 
I have published it in 1892.° 
If Geospiza was represented by three species, when all the 
islands were still in connection, each of these, after the segre- 
gation into different islands, developed its own races, which 
gradually became species. We always can recognize these 
three forms if they are present ona single island, and they 
never intergrade on the same island. We can distinguish 
three parallel series of Geospiza. First the large forms, repre- 
sented by Geospiza strenua Gould; second, the medium-sized 
forms represented by Geospiza fortis Gould, and, third, the 
small forms represented by Geospiza fuliginosa Gould. The 
same is true for Camarhynchus. The table shows these differ- 
ent series. 
Ishall now make a few remarks about the birds from 
Charles, Hood, Barrington, and South Albemarle, which were 
contained in a box which disappeared in Guayaquil. The loss 
is not quite so unfortunate as stated by Mr. Ridgway. He re- 
marks that it contained more than forty land birds from the 
southern part of Albemarle Island, but this statement, as will 
be seen from the list which I now give, is not correct. The 
only two species of birds which are lost are two specimens of 
Camarhynchus, and the new species of Nesomimus from Bar- 
rington Island, of which accidentally no alcoholic specimens 
were preserved. All the other species contained in the box 
are represented by alcoholic material. 
List of lost specimens from Charles, Hood, Barrington and 
South Albemarle, and the number of alcoholic specimens pre 
served : 
CHARLES ISLAND. 
2 Dendroica aureola (Gould), 5 in alcohol. 
6 Geospiza fortis and fuliginosa Gould, 33 in alcohol. 
26 Cactornis intermedia Ridgw., 8 in alcohol. 
11 Camarhynchus. 
2 Myiarchus magnirostris (Gould), 3 in alcohol. 
‘Baur, G. Ein Besuch der Galápagos Inseln. Biolog. Centralbl. Bd. XI, 
1892, p. 248-249. 
