454 PROCEEDINGS OF UNITED STATES NATIONAL MUSEUM. 



The genus Ckhlhermima Bp. lias especially been regarded as an in- 

 termediate link between Tbruslies and Mocking-birds. In 1854 Bona- 

 parte divided tbe genus and made C.fuscata the type of tbe genus Cich- 

 lalopia, wbicb name as untenable bas been changed by Mr. Sclater into 

 Margarops. Unfortunately, however, this separation later has been 

 given up,* because tbe restricted genus Cichlherminia (type herminieri 

 Lafr.) unquestionably belongs to the true Thrushes, group MeruJea;, 

 while on the other hand the genus Margarops (including fuscata, den- 

 sirostris, and montana) as undoubtedly belongs to the Miminw. (See 

 ligs. on pages 457 and 470.) By separating and placing these genera in 

 this manner, the limits between Turdidw and Mimincv become very tren- 

 chant, and the arrangement of the families more natural. 



Later investigations have confirmed the doubts expressed by Profes- 

 sor Baird (Rev. p. 410) about the validity of the family Saxicolidw. Dr. 

 Cones in his "Birds of the Colorado Valley" (1878), p. 70, still retains 

 the term, but at the same time he very frankly confesses : " Eecognitiou 

 of the family tSaxicoUdcc is purely a conventional matter, in which most 

 ornithologists tacitly agree to follow each other upon no better ground 

 than that of precedent." Mr. Seebohm {I. c.) includes the genera Saxi- 

 cola and Sialia among the Tiirdimc. In fact, the SaxicoUdw are so 

 closely allied to the Thrusbes that they only can claim recognition as 

 a special group within the same subfamily. Moreover, I have distin- 

 guished as a separate group the Sialiecc, which have formerly been 

 united with the Saxicolea-, but whicb certainly differ more from the 

 Chats than from the Thrushes. The fact that I have found it necessary 

 to unite witb the Bluebirds a species which hitherto has been regarded 

 as a Turdus,\ shows where their true relations are to be found; the 

 shape of their legs, tail, bill, their habits, and coloration prohibit their 

 position within the same group as the Chats, showing the necessity of 

 establishing for them a separate group, coequal to tbe groups occupied 

 by the Thrusbes and the Chats. Besides, the grouj) Saxicolcw, which only 

 embraces one American species, viz, Saxicola oenanthe, by removing the 

 Bluebirds becomes more natural and homogeneous, including, as I now 

 believe, Saxicola, PrafincoJa, RuticiUa, etc. I cannot agree with tbe 

 authors of the Catalogue of Birds in the British Museum, who keep the 

 genus Fratincola within the family Muscicdpida' (vol. iv, p. 178), altbough 

 it, on the other hand, may be regarded as a well-defined genus in con- 

 tradistinction to the statement of Mr. R. Collett. (Chr. Vid. Selsk. Forh., 

 1881, No. 10, p. 3.) 



It will be seen that the following arrangement differs from that of 

 most systematists in separating Turdus merula and its allies as a group, 

 Meruh(v, distinct from and coequal to the Turdew and Zusciniece, and in 



*St;e Sclater aud Salv., Nomeuel. Nr. Neotrop, p. 2, aud Sharpc, Cat. Birds Brit. 

 Mns., VI, p. 32G. 



tBy Mr. Seebohm, however, iuchuled together with other heterogeneous elements 

 among the genus Geockhla, subgeneric group HcqierodchUv {op. cit., p. 151). 



