7i2 



THE AMERICAN NATURALIST. [Vol. XXXVII. 



to the antecedent vertebrae of the so-called "sacrum." The 

 first pair have small unciform processes, their ha-mapophyscs 

 articulating with these bones on the last dorsals in the usual 

 manner; the last pair, which are very delicate in structure, vary 

 exceedingly in length, and terminate in free extremities." Now 

 the just-quoted description does not tally with what I find in the 

 skeleton of C. alcyon from the U. S. National Museum, for in 

 the first place in the sternum of that individual there are font 



upon the left. The fault here, however, or this discrepancy 

 seems to be due to some difference in arrangement at the fore 

 end of the series, as the left costal process is longer, and has a 

 suspicious looking spine ancliyloscd to it, at about the point 



total number of ribs in these two specimens of C. alcyon may, 

 however, have been the same, while a difference only existed as 

 to the number that connected with the sternum by costal ribs. 

 This will not apply, though, to what we find in C. cabanisi, for 

 here, although there are four dorsal pairs of ribs, supplied by 

 the 15th, 16th, 17th and 18th vertebrae, only the three leading 

 pairs connect with the sternum by means of ha-mapophyses ; 



one pai^of "sacral ribs," and their haemapophyses are still 

 shorter than the pair on the last dorsal ribs, and these sacral 

 ribs are likewise without epipleural appendages, being the only 

 pair that do lack them in this species. The arrangement of the 

 ribs in birds cannot always be relied upon nor even that the 

 same species always have the same number ; but I cannot well 

 account for the difference in the number of the cervico-dorsal 

 vertebrae in these two kingfishers, the method of anchylosis 

 of the pelvic bones with the sacrum sometimes has something to 

 do with it, but apparently not here, for as near as I can count in 

 the pelves of these adult birds, there appear to be twelve ver- 

 tebrae in the pelvic sacrum of C. alcyon, and but eleven in C. 

 cabanisi. This evidently does not help, for to satisfactorily 

 account for the difference in question, the count should stand 

 the other way. 



Aside from the number of vertebrae in the pelvic sacrum, how- 



