369 REVIEWS. 



iirements demonstrate what he claims — that many or most species, 

 vary in total stature, in the length of different members, and in 

 the relative proportions of different parts, from twelve to eighteen 

 per cent, of the mean dimensions. This, it should be remembered, 

 is independent of geographical differentiation ; and it will, we trust, 

 be enough to put us on our guard, against too ready acceptance of 

 slight discrepancies in size as an element in our diagnoses of spe- 

 cies. To cite but a single case in point : — twenty-seven speci- 

 mens of Pants atricapillus, taken in the same locality and at nearly 

 the same time, differ over an inch in length ; that is, they grade 

 between the extremes of P. septentrionalis and P. Garolinensis. 



" Climatic difierentiatiou," like individual variation, is shown to 

 occur under three principal phases : — in total size, in length and 

 stoutness of bill, and in color. The first of these is well known 

 and need not detain us. The second consists in the curious fact 

 above mentioned, of development of the bill in inverse raXio to the 

 size of the bird, with decrease in latitude. Climatic variation in 

 color, Mr. Allen holds, may be in respect of both latitudinal and 

 longitudinal conditions. In the matter of latitude, he shows a 

 gradual increase in intensity of color to the southward, probably 

 dependent upon the greater energy of the sunlight ; and in this way 

 disposes of several West Indian forms, commonl}^ reputed as spe- 

 cies, pointing out that they are not more different from their Flori- 

 dan analogues, than these are from New England examples. As 

 to longitudinal variation, Mr. Allen brings prominently forward 

 the fact that "the general tendency from the East westward, is 

 to darker or deeper colors in specimens of the same species" (p. 

 237), and supports this by numerous unquestionable cases, many of 

 which will readily occur to the reader. The very notable excep- 

 tions afforded by the bleached specimens of the Colorado desert, 

 furnish the occasion for what we regard as decidedly the most im- 

 portant point in the whole discussion, and one that we do not re- 

 member to have seen in print before. Under head of " Causes of 

 Climatic Variation" (p. 239), Mr. Allen takes humidity of the at- 

 mospjhere, as determined by the mean annual rain-fall, to be the 

 real cause of this intensity of color. In examining Dr. Foster's 

 " The Mississippi Valley," some time since, we were struck with 

 the determinations there made of the hygrometric influences result- 

 ing in the production of forest, prairie and desert, according to the 

 mean annual water-supply, and, at the same time, perceived the 



