21 [Vol. xxxvi. 



the same subject in the B. O. C. I am afraid I must 

 disappoint you very much, because I agree "ndth my 

 '' opponent '' in almost every case. It is true that twenty- 

 five years ago I was more enthusiastic about the services 

 ■which the study of eggs would render to classification. 

 I never went so far as to attempt an ornithological system 

 based on oological characters, but I thought that eggs 

 would in many difficult questions finally decide the system- 

 atic position o£ birds difficult to place. I do not now hold 

 this view, because there are so many cases where eggs^ 

 if I may say so, leap out of the line, i. e. where certain 

 genera or species have eggs that difi'er materially from those 

 of all their allies. One of the most striking cases, already 

 mentioned by Mr. Jourdain, is that of Podoces humilis, a 

 member of the family Corvidse, which lays a snow-white 

 egg. Another is that of Dramas — also mentioned before 

 by Mr. Jourdain — which certainly lays an un- Wader-like, 

 siugle^ large, white egg. Again, the eggs of Montifringilla 

 are white, a character only found in this genus among the 

 Pringillidee. Opisthocomus , the "^ Hoazin," lays a typically 

 Ralline &^s,, yet its anatomy is thus singular, that it has 

 been placed in a special order — but close to the Rails, and 

 in this case I believe that the Ralline e^^ is really an 

 ancient character of significance. I have no doubt that 

 eggs preserve ancestral characters very long, because they 

 are less apt to be modified by external influences : the 

 duration of their existence is extremely short, as compared 

 with that of the birds ; during this period they perform no 

 active functions and come only into passive contact with 

 their surroundings. 



But striking exceptions from the rule are not confined to 

 coloration. The shape of the eggs is generally characteristic 

 of larger groups, such as orders or families, but there are 

 some exceptions : we find, for instance, that the Chatham 

 Island Snipe lays eggs that are not in the least pyriform, 

 and the same is the case with the Woodcock, yet both 

 genera are closely related to that of the Snipes, Gallinago, 

 with its pyriform eggs. The number of eggs in clutches is, 

 in some cases, very constant. 



Humming-birds lay always two eggs, and I do not know 



