48 ORTHOGENETIC EVOLUTION IN PIGEONS. 



In the fourth feather we find at first glance no striking change in the general 

 form and extent of the two spots. It is only by close inspection of the minute 

 features that we discover that the trend of cumulative changes is still in the same 

 direction, and that as the band fails more and more of its highest development, 

 its individuality becomes less and less distinct from that of the bar. Both spots 

 are a little more strongly and evenly pigmented than in the third feather. The 

 proximal borders are less band-like, but fall only in fine degrees below the condi- 

 tions seen in the third feather. 



In the upper spot there is still a suggestion of the band in the slight preponder- 

 ance of pigment in the proximal part of the triangle. There is also a very slight 

 incurving of the upper side of the spot, corresponding to the deeper curve seen at 

 the same place in the third feather. 



In the lower spot the condition of the proximal border is particularly instructive. 

 Although the spot as a whole is a little more strongly pigmented than in the third 

 feather, the border takes a step in just the opposite direction; that is, it sinks 

 plainly below its height in the third feather. The two characters are no longer 

 in equipoise; the scales turn in favor of the spot. 



Let us compare closely, in order to see with what nicety the transitional degrees 

 are marked. As shown in the figure, the loss of pigment is so nearly even for the 

 entire proximal border that it still looks as if it were the same thing before described, 

 except as weakened in quantity of pigment. As before, the pigment is densest at 

 the lower end, becoming gradually thinner towards the shaft, until in the upper 

 third its individuality, if it have any here, is quite beyond recognition. The con- 

 ditions in the two spots have changed so very little that we have no difficulty in 

 seeing just where the pigment should lie if present; but I find that no border is 

 here marked off; band and bar are one in this third. 



The bend of the border around the lower angle of the spot and its pointed 

 extension are still quite distinct, though not so sharp in outline as before. The 

 distance of the bend from the tip of the feather is here 1.7 mm. greater than in the 

 third feather. 



In the feathers thus far examined, the inclination of the border and the distance 

 of the bend from the tip of the feather run as shown in the accompanying table. 

 From these measurements it will be seen that the distance from the tip is closely 

 correlated with the inclination of the proximal side of 

 the spot, every 1 mm. in one column corresponding to 

 about 2° in the other. As we descend the series, the 

 band gradually loses its transverse position and sinks 

 out of sight in the proximal border of the enlarging 

 spot. As we ascend, relations are reversed; the spot is 

 reduced in length, its proximal border becomes less and less inclined and more 

 and more band-like. In one direction the band merges in, in the other it emerges 

 from, the bar. Evolution is here linear, and hence it can have but one direction. 



» The figures for inclination given here do not agree, except in the sense of an orderly increase, with those given 

 in connection with plate 17. — Ed. 



