WATER BIRDS. 



65 



mutual relationships of the water birds. Their position and sequence has consequently 

 been completely changed, and the aspect of that part of the system is quite upset. 

 Ornithologists of the old school will have some difficulty in locating themselves. 



Instead of the two old orders, Natatores and Grallatores, or three with the addition 

 of the Herodiones, the modern researches have revealed four centres of relationship, 

 which may be styled Pluvialiformes, Anseriformes, Ciconiiformes, and Pelecani- 

 formes, the three latter forming the desmognathous, the first-mentioned group the 

 schizognathous series. The latter series more especially comprises the birds which 

 usually constitute the orders Pygopodes, Longipennes and Limicolae, the two first men- 

 tioned being here merged into Cecomorphae, according to Professor Huxley's view. 

 The arrangement may not be regarded as final, however, for there are reasons to sus- 

 pect that it will be necessary, ere long, to divide the schizognathous swimmers into 

 three orders, Eretmopodes for the first 

 two superfamilies of the present arrange- 

 ment, Tubinares for the superfamily 

 Procellaroidese, and Pluviales for the 

 rest of the members of the two orders. 



The fact is, that not only are the 

 gulls very nearly allied to the auks, but 

 their affinities with the Grallae, through 

 the plovers, are unmistakable. On the 

 other hand, the grebes seem to be only 

 distantly related to the other 'Pygo- 

 podes,' and the puffins and albatrosses 

 similarly so to the 'Longipennes' or 

 gulls. But there is one feature of more 

 novel systems which we are not willing 

 to adopt, viz., the position of the plovers 

 at the base of the series as the most 

 generalized forms. That in these birds 

 some anatomical features of quite a gen- 

 eralized character remain during the 

 whole lifetime, while in the gulls they 

 are only present in the embryonic state, 

 is no argument in favor of the view of 

 the latter being only a degraded branch 

 of the former, since we find ample proof 

 all through the class that one or more 

 character! of the ancestral stock may 

 survive in a highly specialized group, while they may be lost comparatively early in 

 another, which, on the whole, has departed only slightly from the common ancestor. 

 Our knowledge of these birds is as yet particularly deficient concerning the degree of 

 affinity between the different forms, and it will therefore be found that in the follow- 

 ing we have given the generally adopted sequence the benefit of the doubt. 



As stated above, the members of this order are strictly schizognathous ; another 

 cranial feature is that they are schizorhinal. Their palmate feet will serve to distin- 

 guish them externally, a feature which, among the Grallae, is only found in the avocets, 

 which are not easily confounded with the birds here in question on account of their 

 VOL. iv. 5 



FIG. 29. Skeleton of Colymbus cristatus. 



