MUSEUM OF COMPARATIVE ZOOLOGY. 369 



quatus, in which the color of the hill is unusually light, and the bill itself' 

 unusually produced. 



182.t Podiceps cornutus Latham. Horned Grebe. 

 "Not uncommon on the St. John's." — Boardman. 



183.t Podilymbus podiceps Lawrence. Carolina Grebe. 

 Abundant on the St. John's. 



Resume of the preceding Tables of Measurements, with suvplemental 



Remarks. 



The following tables present a brief summary of the measurements 

 given in Part IV. In the first table is given the average dimensions of 

 thirty-two species, based on specimens collected, in each case, essentially 

 from the same locality, and generally based on twenty or more speci- 

 mens, the number varying in the different species from thirteen to 

 sixty-five specimens. In all cases where the average sexual differ- 

 ence in size is appreciable, the dimensions are given for each sex. 

 In most cases very nearly all the specimens are from Eastern Massa- 

 chusetts, a few being from different localities in Southern Maine, and 

 a few from Northern Illinois. In a few species all the specimens cited 

 are from Eastern Florida ; in a few other species part of the specimens 

 are from Southern New England and a part from Eastern Florida ; 

 but in these cases a separate average is made of those from each of 

 the two localities. The number of the specimens on which the average 

 is based is given in each instance. 



The second table shows the range of individual variation in size in 

 the same species, based also on the same specimens. 



The third table shows the amount of geographical variation in size in 

 specimens of the same species from northern and southern localities, 

 these localities being generally Southern New England (Eastern Massa- 

 chusetts in the main) and East Florida. Only seven species are cited, 

 but I have traced about the same ratio of difference in a score or more 

 of others, of which the measurements have not yet been published. 

 Although the number of specimens compared from the two localities 

 has in many of these cases been comparatively small, enough have 

 been examined to show the general constancy of the variation in all 

 the species which breed at both these localities. 



It should be added that the specimens on which the generalizations 



vol. ir. 24 



