388 THE AMERICAN NATURALIST. (VOL. XXXII. 
fore, it necessarily follows that either Dr.’ Baur’s statement of 
the size of the bird which he saw but did not obtain is very in- 
correct, or else that my doubt as to its being Æ. egretta was very 
well founded. It would be interesting to know by what process 
Dr. Baur was able, under the circumstances, to positively identify 
the species. 
In the “ Additions to the List of Birds given by Ridgway for the 
Different Islands” (pages 782-84), I have found it difficult to find 
out exactly what Dr. Baur means to show; but in my attempt to do - 
so have made one important discovery, which is that the species 
named, which are really additional to the lists given in my paper for 
the separate islands, were certainly not among the specimens which 
Dr. Baur sent me for examination, and therefore I cannot be respon- 
sible for the omissions. Many of the species which he names do 
occur in my lists, however, but in the case of most of these, owing to 
the circumstance that no Baur-Adams specimens were known to me, the 
“x” was not entered in the column for that collection. The impor- 
tance of the portion of the collection which was not sent to me may 
be realized from the fact that, according to Dr. Baur’s paper, his 
collection contained specimens of Camarhynchus pallidus (“ Cactornis 
pallida”) from Duncan, Chatham, and Jervis Islands, while he sent 
me only two specimens, one from Jervis, the other from James Island. 
Neither did I see a specimen of Mesomimus macdonaldi from Gardner 
Island ; had I been able to do so, it is hardly necessary for me to 
say that the mistake respecting the identification of this bird to 
which Dr. Baur refers (see footnote on page 783) would not have 
occurred. 
The remaining point upon which Dr. Baur’s criticisms bear is the 
first one mentioned by him, and the one to which he devotes most 
space; but I prefer to consider it last and most briefly, since it is 
chiefly a matter of opinion, while the others are questions of fact. 
What are genera and what are not is, in many cases, very difficult to 
determine. To Dr. Baur Cactornis and Geospiza seem to be distinct, 
and to have them so would better fit his theory of distribution. To 
me they are not distinct, because it is impossible to draw any line 
between them.’ It is, of course, disappointing to find sometimes 
that facts do not entirely support our theories ; but it seems to me 
1 I would here call attention to Dr. Baur’s erroneous quotation of my remarks 
on page 778, where, in the eighth line from the bottom, the following should be 
inserted after the first word: “I am still of the opinion that not a single character 
can be found 
