482 Mr. H. Seebohm on the 



possible that it would be as consistent to associate the Stone 

 Curlews with the Gulls as the Lapwings with the Avocets. 

 These seven genera further agree with the Laridae, and differ 

 from the Charadriidse^ in having no lateral occipital fon- 

 tanelles when adult. Huxley separates the Limicolae (under 

 the name of the Charadriomorphae) from the Gavise, which 

 he associates with the Tubinares, the Colymbi, and the Po- 

 dicipes (under the name of Cecomorphae) . Sclater separates 

 the Limicolse^ the Gavise, and the Tubinares from each other, 

 but he removes the Alcidas from the Gaviae and associates 

 them with the Colymbi and the Podicipes (under the name of 

 Pygopodes) . If the articulation of the dorsal vertebrae be an 

 important character, the Alcidae and the Gaviae cannot be 

 separated far from each other, nor can the former be asso- 

 ciated with the Colymbi or the Podicipes. 



As thus restricted, the Gavio-Limicolae possess the follow- 

 ing characters : — 



1. The posterior ends of the dorsal vertebrae, as seen in 

 lateral section, are not convex. 



2. None of the dorsal vertebrae above the ankylosed sacral 

 vertebrae are ankylosed. 



3. The first digit of the manus is present. 



4. The pelvis is not laterally compressed. 



5. The coracoid is furnished with subclavicular processes. 

 The combination of the 1st and 3rd of these characters is 



diagnostic of the group. 



In the bifurcation of their nasals the Limicolae are all 

 schizorhinal ; and of the Gaviae the Laridae and Alcidae are 

 all schizorhinal ; but the intermediate genera vary in this 

 respect. Dromas is typically schizorhinal, Chionis and Gla- 

 reola are almost so ; the angle is blunter in Thinocorus, and 

 well rounded in Cursorius, whilst Pluvianus and (Edicnemus 

 are typically holorhinal. 



Although this character, founded on the modification of 

 the bifurcation of the nasals, appears completely to break 

 down in the Gaviae, its importauce in other groups must not 

 be undervalued. 



It is impossible to determine the relative importance of 

 osteological characters. In the foregoing key I have given 



