294 A HISTORY OF BIRDS 



to whom I am indebted for the foregoing summary, taken from 

 his most valuable work, Variation in Animals and Plants, re- 

 marks : " The observations really show more than this, however. 

 It is true that, as a rule, the most extreme individuals in either 

 direction are eliminated, but if the distribution of the various 

 characters be plotted out . . . it will be seen that in the case 

 of some of the other characters, as in that of length, the elimi- 

 nating process acts much more on the extreme individuals in 

 one direction than in those in the other . . , the eliminated 

 birds were, on an average, heavier. This conclusion has already 

 been obtained by the simple process of taking averages ; but 

 the curves show in addition that it is the very heavy birds 

 which were more especially eliminated. Thus of fourteen birds 

 of 27-3 grms. and upwards obtained, only three survived. Simi- 

 larly, also in respect of the femur measurements it was found 

 that of the nineteen birds obtained with a known length of 6'8s 

 inch or less, only seven survived, whilst twelve were eliminated. 



" The next generation of birds collected in the storm-swept 

 area would accordingly be shorter in length, weigh less, have 

 longer legs, a longer sternum and a greater brain capacity 

 than the former generation, supposing, of course, that the 

 variations existing in these characters were partly of blasto- 

 genic and not wholly of somatogenic origin ; and this could 

 scarcely fail to be the case." 



Some birds vary more, in regard to certain parts, than 

 others ; and this is especially the case with organs that are 

 of great importance in procuring food. The Curlew, for ex- 

 ample, shows a striking range of variability in regard to the 

 length of the beak, ranging from 4'6 to 7 inches. The Common 

 Heron {Ardea cinera) affords a no less striking instance in the 

 variation of the length of the tarso-metatarsus, though no 

 measurements illustrating the extremes of length appear to 

 have been recorded. But the Limicoline birds all appear to 

 be variable in the matter of size, and this is especially notice- 

 able in such species as the Dunlin (Tringa alpind). 



Some valuable and striking facts have been brought to 

 light with regard to this matter of variation by Mr. R. M. 

 Barrington, who in his monumental work on The Migration 

 of Birds at Irish Light Stations made a large collection of 

 facts which have not received the attention they deserve. 



