568 PTEEYLOGBAPHY OF GOAT-SUCKEES AND OWLS—CLAEK. vol.xvii. 



Fundamental plan and variation in detail. — In regard to the funda- 

 mental plan of tlie pterjiosis, it needs only a glance at tlie figures to 

 show us that while dorsally the two groups are very similar, there 

 is ventrally at least one important difference. On the head the 

 feathering is much more dense and uniform in Striges than in Capri- 

 mulgi, but there are nevertheless signs in the former group of a 

 tendency towards a less uniform covering, as witness the longitudinal 

 rows of Asio and Mccjascops. Furthermore, in Ghordciles, there is 

 a much more uniform feathering of the infra-mandibular region than 

 in any owl except x^ossibly SiJeotyto. There is no constant differ- 

 ence between the two groups in the upper and lower cervical tracts 

 which are always narrow and clearly forked. The dorsal tract is prac- 

 tically the same in both, varying indeed in the different genera as to 

 its extent and its union with the forks of the upper cervical. The 

 strong humerals with an evident parapterum are also common to the 

 two groups and the peculiarly i)laced femoral tract is likeM'ise charac- 

 teristic of both, which is the more remarkable as the tibia is always 

 much more heavily feathered in Striges. As a rule, too, the owls have 

 the feet much more fully clothed, but as much of the tarsus is bare in 

 Gymnoglaux as in Antrostomus so that this slight difference is by no 

 means constant. On the ventral surface the two groups agree in sev- 

 eral minor details such as a strong hypopterum and hook, connecting 

 with the sternal tract, and the numerous scattered feathers on the 

 upper outer corner of the breast and on the shoulder, including a 

 connection between the lower cervical and the humeral tracts. But it 

 is on the breast that we find the first real difference in .the fundamental 

 plan, and this demands a careful examination. In the Caprimulgi, the 

 lower cervical after finikin g continues on each side as a single tract, 

 forming on the breast a broad and strong sternal, and on the belly, 

 after narrowing abruptly, the much weaker ventral tract. In the Stri- 

 ges on the other hand, the lower cervical, after forking, forms on each 

 side of the breast the well-marked sternal tracts, but does not continue 

 down on the belly to form the veutrals. These are, on the contrary, in 

 the typical Strigine i^tei-ylosis, entirely separate from the other tracts 

 at least as far up as the furcula and owing to their weakness at that 

 point their union with the sternals is often very indistinct. It will be 

 at once seen that this difference is really important, but when we con- 

 sider the condition of these tracts in Strix, we find an arrangement 

 that is really intermediate between the two groups, and this gives us 

 a hint as to how the Oaprimulgine form may have been derived from 

 such an arrangement as occurs at present in the owls. In Strix, as has 

 already been pointed out,' not only do tlie ventral and sternal tracts 

 fuse clearly near the furcula, but owing to the slight outward curve of 

 the former and a more abrupt inward curve of the latter, the two 

 unite at the posterior extremity of the sternals so as to form in reality 

 one broad tract on the breast, containing a longitudinal apterium. It 



