l8o Rtcent Literature. [April 



mopolitan, continental, North American, West Indian, etc., while those 

 of the second class are compared in detail with their nearest allies. His 

 conclusions are (i) that the Avifauna of the Bahamas is strongly West 

 Indian, and that this group of islands is entitled to rank as a fauna of the 

 Antillean region, the endemic species having been derived from West 

 Indian stock, with a slight intrusion from Florida ; (2) that Cuba has been 

 the source of the greater number of forms; (3) that while North American 

 species occur numerously as migrants, they have not assisted in forming 

 the resident avifauna; (4) that the avifauna is of comparatively recent 

 origin; (5) that forms having a common ancestry, and which now oc- 

 cupy widely separated areas, may have become so differentiated as to re- 

 semble more each other than they do the original stock {e.g., Geotklypis 

 beldingi oi Lower California as compared with the Bahama and Florida 

 forms, with nearly parallel cases in the genera Sfindalis, Vireo, and Cer- 

 thiola) ; (6) that certain Bahaman forms, occupying contiguous islands, 

 have become differentiated under practically the same climatic or physio- 

 graphic conditions — in other words, simply through isolation; (7) that 

 we may probably assume that some of these endemic forms owe their 

 origin primarily to features of individual variation, which through isola- 

 tion, and hence close interbreeding, have become permanent. Several of 

 these important generalizations we do not remember to have seen pre- 

 viously stated. The paper is thus of unusual interest. — J. A. A. 



Chapman on the Grackles of the Subgenus Quiscalus.* — This inter- 

 esting discussion of a highly perplexing subject — the relationships of 

 our Grackles — is the result of the study of a series of over eight hundred 

 skins, largely of breeding males, from many widely separated localities. 

 This series shows the breeding range of ceneus to extend from Texas and 

 Louisiana to Great Slave Lake, "and from the eastern slopes of the 

 Rockies to the western slopes of the Alleghanies, while from Massachu- 

 setts to Nova Scotia it reaches the Atlantic seaboard ; Quiscalus quisciila 

 aglcBus is typically represented from New Orleans to Charleston, and 

 southward to the extreme point of the Florida peninsula; and Quiscalus 

 quiscula breeds from the northern limit of the range of aglcBus northward 

 to the southern limit of the range of cenens in the lower Connecticut and 

 Hudson River Valleys." 



As a preliminary to the discussion of their relationships, a detailed de- 

 scription is given of the coloration of each form, especially o^ quiscula, of 

 which lack of space unfortunately permits us to quote but little. 



In cBneus, throughout its range, aside from a trifling seasonal, and con- 

 siderable sexual, difference in brilliancy, there is practically no variation 

 in the colors of the plumage, except of the head and neck which in both 



*A Preliminary Study of the Grackles of the Subgenus Quiscalus. By Frank M. 

 Chapman. Bulletin American Museum of Natural History, Vol. IV, No. i. Article 

 I, Feb. 25, 1892, pp, 1-20; map, 



