1903.] SYRINX OF THE ACCIPITRES, 163 



the tongue is covered by several rows of backwardly directed spines, 

 in addition to the row of larger spines along the posterior edge of 

 that organ. On the other hand, in one or more species of the 

 genera Milvus, Circus, LeucojJteroiis, Spizaetus, A7'chibuteo, Dryo- 

 triorchis, Morphnics, Asturina, Gypohierax (text-fig. 20, p. 162), 

 and Uruhitinga, the tongue has only the posterior row of spines. 

 Vultur is peculiar in possessing, in addition to the posteiior row of 

 spines, a lateral row on each side along the thick edge of the 

 tongue. There are traces of them in Aquila, Nisaetus, and 

 Gypaetus, and perhaps in some other forms. 



Conclusions, 



The principal conclusion of the foregoing enumeration of facts 

 concerns the classification of the Accipitres. It is perfectly plain, 

 as I think, that the structure of the syrinx permits of, or indeed 

 necessitates, the division of the Accipitres into two families or 

 subfamilies, which may be termed Falconidfe and Buteonidse (or 

 Falconinse and Buteoninse). It is important to notice that in 

 insisting upon such a division, I am in complete agreement with 

 Dr. Suschkin* and Mr. Pycraftf, both of whom have lately 

 attempted a classification of the group on Osteological grounds. 

 This satisfactory result shows the value of the form of the syrinx 

 in this group, as has been shown with the syrinx of other groups, 

 such as the Cuckoos. 



As to further subdivisions of these two main divisions, the facts 

 at my disposal do not allow of any expression of opinion. Nor 

 do I venture to lay too much stress upon the apparent agreement 

 in the structure of the tongue with the facts derived from a study 

 of the syrinx and bones, since the material at my disposal was not 

 very abundant. 



The next most important point which I have been able to 

 ascertain is the development of a bronchial syrinx in Dryoiriorchis. 

 This makes it more plain than ever that it is dangerous to compare 

 birds belonging to difierent groups which happen to possess the 

 bronchial form of the syrinx ; for while it is possible on other 

 grounds that the Goatsuckers and Owls, both of which contain 

 genera which have bronchial syringes, are allied, it is not probable 

 that they, the Cuckoos, and the Hawks ai-e all nearly akin. It 

 seems to me to be a form of syrinx which has been acquired 

 several times and independently. 



* Zoo]. Anz. vols. xxii. & xxiii. 1899 & 1900. 

 t P. Z. S. 1902, vol. i. p. 315. 



\v 



