300 ANATOMY AND AFFINITIES OF CCEREBIDJE— LUCAS. vol.xvii. 



of skeletal diagnosis, and this test applied to the passeres reduces them 

 to a family or two, as has been done by Huxley and Fiirbinger. 



It would almost seem that, aside from purely negative results, the 

 skeleton can be relied upon to show but two things, very general and 

 very close affinities, for the variation of parts is so infinite that between 

 any 10 given birds we may find every intermediate stage and establish 

 relationships in all directions. 



Then, too, characters which would be of nuich importance among 

 mammals appear, from their instability, to be of but little value in biids. 

 An example of this is found in the condition of the j)resacral vertebrae. 

 In a large number of Passeres there are 4 presacrals, the third and 

 fourth being fused and having a common transverse process; in others 

 there are 5 presacrals, the fourth and fifth being fused. Such char- 

 acters as these would seem to be of some importance, and yet Himatione 

 parva has the third and fourth presacrals fused, while H. sanguinea has 

 the fourth and fifth united. And these birds are undeniably closely 

 related. 



The same thing occurs again and again in other closely related spe- 

 cies, such, for example, as Merula migratoria and Turdus musicus, while 

 the instability of the character is well shown by the fact that it is by 

 no means uncommon to find sacra in which, on one side, the third and 

 fourth vertebrae are fused and on the other the fourth and fifth. 



The degree of value to be assigned the pterylosis is yet unsettled, and 

 this can only be done by accumulating and comparing the facts in the 

 case. It would be a great service if some one with ample time and unlim- 

 ited patience would plot the pterylosis, or even the configuration of the 

 dorsal tract, in as many small birds as could be obtained, for it would 

 then be possible to ascertain what correlation, if any, there is between 

 tract pattern and other characters. 



Between the continuous dorsal tract of a thrush and the inverted Y 

 of a swallow there is a great difference, and this difference should have 

 some definite meaning, exactly what meaning, is to my mind, not yet 

 evident. 



All the birds examined during the preparation of this paper have an 

 uninterrupted dorsal tract whose shape appears to be specifically subject 

 to great variation, but these variations are so slight and so innumera- 

 ble that, except for general purposes, the pattern ai)pears to be of little 

 service. 



The convolutions of the intestine are in very much the same case 

 as the pterylosis for, Judging by Dr. Gadow's figures and my own lim- 

 ited number of dissections, they are subject to great specific variation. 

 There is certainly a decided difference between the alimentary canal 

 (including the stomach) of birds so nearly alike as Ccereba cyanea and 

 C. eoeruleaj and the genera of tanagers vary widely. 



The indications are, as might not unnaturally have been expected, 

 that such parts as the tongue and alimentary canal are subject to great 



