64 Recent Literature. [ ZOE 
each of the three phases of guzscu/a. The variations of each form 
are then carefully followed throughout their breeding range, and 
the general conclusions as to relationship stated. The two most 
important conclusions are that—“ In the Alleghanies of Pennsylva- 
nia, in the Hudson Valley from Sing Sing to Troy, in eastern Long 
Island, in Connecticut, and in Massachusetts as far north as Cam- 
bridge, guzscula and @neus completely intergrade’’; and that— 
“This intergradation is in every instance accomplished through 
phase No. 3 of guiscula.’’ 
Mr. Chapman then argues very reasonably that guzscula is a dis- 
tinct species, and not a race of @neus. If this be not the case, he 
asks why enews should remain so perfectly constant over an im- 
mense area and then change into three different forms. It is, at 
least, impossible to see any environmental influence which could 
have produced such a modification as this, and the matter accord- 
ingly becomes inexplicable upon any theory except hybridity. 
Although Mr. Chapman has established by his careful investiga- 
tion at least the great probability that hybridization is the rule 
among the grackles, he is hardly justified in extending this to other 
species. Thus he says: ‘‘ Nor do I see any good reason why we 
should refuse to admit hybridization as a factor in the evolution of 
what we term species. * * * Difference in habit under what 
must necessarily be similar conditions will ever be an effectual bar- 
rier against the indiscriminate mixing of even closely -allied birds. 
But when two species whose natural economy, song, nidification, 
etc., are the same, and which agree in structural details and differ 
only in coloration, inhabit contiguous regions, is it unnatural that 
they should at first occasionally, and in the end regularly, inter- 
breed? The evidence in proof of such intergradation is gradually 
accumulating, and in the future I think we shall be forced to recog- 
nize hybridization, not only as a means which unites known forms, 
but which also gives rise to new ones.’’ 
The writer has apparently overlooked, in the above passage, the 
possibility of physiological selection interposing a barrier to hybrid- 
ization, even when the two species appear to be structurally iden- 
tical. If the theory of physiological selection is to have any valid- 
ity whatsoever, it is necessary to assume that such cases of habitual 
hybridization as are occasionally recorded, are exceptional and ab- 
normal. To be sure, it may be objected that this is arguing from 
