EVOLUTION OF THE COLORS OF BIRDS. 167 



stead of the repetition of parts we have here the repeti- 

 tion of color pattern. 



The two fused spots of the second feather (fig. 17) 

 are probably not homologous with the two bands of the 

 first feather, but are two additional spots. This may be 

 due to the fact that the terminal portion which contained 

 them has become aborted. At any rate, the character of 

 this first feather is so different from the second that an 

 abrupt transition of pattern might be anticipated. It 

 will be noticed that there are two other tendencies man- 

 ifested in this series — for the lowest spot to increase in 

 size and for the white bands to be crowded toward the 

 edge of the feather; both of wdiich tendencies take place 

 gradually and reguhirly. On the opposite wing the 

 markings are very similar, feather for feather. Merely 

 in slight details do they differ, as for instance in regard 

 to the fusion of spots, which is not so decidedly pro- 

 nounced in the second feather and more nearly ap- 

 proached in the third. Again there is a tenth white 

 bar, which is barely indicated in the second feather of 

 the wing figured but well marked on the- same feather of 

 the other wing. 



There is a more marked change between the last pri- 

 mary and the first secondary (fig. 26), The subterm- 

 inal spot which has been gradually enlarging suddenly 

 breaks out into a broad terminal band of bluish with a 

 white spot still left. A trace of black has also persisted 

 along the line of the shaft. As a sort of premonitory 

 symptom of the coming light band, a small spot of 

 white is noticed at the base of the last three primaries. 



The next important change occurs in what I take to 

 be the first of the tertiaries (fig. 34) where a decided 

 tendency toward the elimination of the black bars first 

 asserts itself. This rapidly advances until the simple 

 stage of figure 39 is reached. The same sort of 



