1S91] Recent Li/crafurc. 



91 



ficial ; the characters which were then tliought to he of threat importance 

 now seem of little worth. Of the three Sections which I have adopted 

 from Mr. Wallace, the one treated of in the present volume seems to be 

 the most unnatural. The Starlings are divorced from the CorvidjE, to 

 which they are undoubtedly allied; the Artamida; may be a Sturnine 

 family, but of that I am not yet assured; the Alaudida; find themselves 

 separated from the Motacillida;, and the PloceId:e from the Fringillida; 

 and Icterid;e." 



The volume is of course an invaluable hand-book of the groups treated. 

 The method of treatment is similar to that of the other volumes of the 

 series by the same author. 'Subspecies' and 'races' are freely recognized, 

 but in a way to avoid 'trinomials' with, in some cases, the singular result 

 of a form described, say in iSoo, being ranked as a subspecies of another 

 form described half a century or more later — an anachronism in nomen- 

 clature not by any means new in this series of volumes. 



In treating the genus Otocorls (or ' Otocorys' as Mr. Sharpe naturally 

 prefers to write it) he says of the American forms: "Anything more 

 puzzling than these races of Horned Larks it has never been my lot to 

 describe. The differences between O. alpestris and O. rnbca arc as well 

 marked as could be wished, but between these two extreme forms are in- 

 terposed a number of races which seem absolutely to connect them [!], 

 ami both of these American authors [Messrs. Ilenshaw and Ridgway] 

 atlmit that these connecting links actually exist [ !]. To write all of the 

 races under the heading O. alpestris would be to obscure the existence of 

 several highly interesting geographical forins, and I have therefore 

 thought it best to recognize the races determined by Mr. Henshaw and 

 confirmed by Mr. Ridgway, with certain notes of my own upon the series 

 now lying before me." (These include the specimens in the Henshaw col- 

 lection.) Mr. Dwight's paper on the same group was published too late for 

 consideration in the body of the volume, but it is mentioned in the 'Adden- 

 da,' and the three additional races proposed by Mr. Dwight after an exam- 

 ination of more than six times the material (2012 specimens) studied by Mr. 

 Henshaw, are accepted, but not ^'■more Americano,'^ under trinomials, but 

 of course after the stereotyped method of Mr. Sharpe. 



After stating that he quite agrees with Mr. Henshaw "that the large 

 Horned Lark of Europe cannot be separated from that of North Amer- 

 ica," his study of the British Museum series of American Horned I^arks 

 appears to have led him to separate the American forms into two specific 

 groups, as follows: i O. leticol<Enia , with 'subsp.' a. pruticola, and (3. 

 arenicola \ 2, O. alpestris, with 'subsp.' a. ckiysoltvma, (3. fcrcgriiia 

 (U. S. Colombia), -y- I'l'f^ea, 8. giraudi, €. strigata. On just what 

 grounds this division is made Mr. Sharpe fails to state; to American eyes 

 they are certainly inscrutalile : for a more arbitrary arrangeinent would be 

 hard to devise. Geographically his subspecies of alpestris are sepa- 

 rated from alpestris proper by tiie intervention of his whole leucoliema 

 group, his first subspecies of alpesiris {alpestris is limited in the breed- 

 ing season to Labrador and the Hudson Bay region) hexng ch>'ysol(eina. 



