208 BULLETIN OF THE 



tion being as great between different specimens of tlie same specie3 

 as between different species of the same genus, and even of differ- 

 ent genera. The variation in the length of the toes is often due to 

 an increase or a decrease in the length of the nail, but by no means 

 rarely to variations in the length of the phalanges themselves. As 

 already stated, and as appears from the table, toes of less than the 

 average length accompany tarsi of the average or of more than the 

 average length, and toes of more than the average length accompany 

 tarsi of medium or less than the medium length. In compiling the above 

 table the specimens mentioned have been selected in each case from 

 a series of only twenty specimens of the species to which they respec- 

 tively belong, and represent the longest and shortest tarsus, middle toe, 

 outer toe, and hind toe met with in each series, and also the greatest and 

 least amount of difference in these several elements. They are all 

 taken from Tables H to Q (see pp. 210-219), which serve to show 

 the usual range of variation, in respect to size and proportions, in ten 

 species.* 



Individual Variation in other Parts. — In addition to the instances 

 already mentioned, individual variation of a similar character and equal 

 extent occurs in the relative size of other parts. The length of the bill, 

 for instance, is often compared to the length of the head, or to that of 

 the tarsus in specific diagnoses. Table G 1 (see next page) serves to 

 show the individual variation in respect to the proportion of length to 

 alar extent ordinarily met with in specimens of the same species. 



To show more fully, however, the exact nature and extent of what 

 may be considered as purely individual variation, tables of detailed 

 measurements of about twenty specimens of each of a number of 

 species are herewith appended (Tables II to Q). Care has been 

 taken to not only select specimens of the same sex, collected at the 

 same locality, and as nearly as possible at the same season, but also 

 such species as find their northern limit so near the locality at 

 which they were taken as to obviate. the complication of individual 

 witli geographical variation, which would result if the range of the 

 spcci.s extended far to the northward of tin 1 , locality in question. 

 In general, the specimens are all from Eastern Massachusetts, and 



* Icterus Baltimore, Dolichonyx oryzivorus, Pipilo erythrophthalmus, Sialia sialis, 

 ,; co tes carolinensis, Pyranga rubra, Geothlypis trichas, Harporhynchus rufus, 

 Tyraunus carolinensis, Hedymeles ludoviciana. 



