THE TURTLE-DOVE PATTERN IN THE PHYLOGENY OF PIGEONS. 91 



still further upward it is shorter, thinner, and more worn; and, finally, just a bare 

 remnant is left. 



On the sides of the neck the black edge is wholly absent, or reduced to a mere 

 dotted line that looks as if it were dropping off. The appearance is such as to 

 suggest, at first sight, that the feather is in process of abrasion, but this appearance 

 is probably due to imperfect or arrested development. 



On the sides of the neck, in the region of the turtle-dove spots, the feathers show 

 the clearest black and white bars and the nearest approach to pure white apical 

 bars; but this apical white has been gained through the loss of the Juvenal pale bar 

 and the first subterminal black bar. 



This shortening of the neck-feathers reminds of the fact that the differentiation 

 of n'ngr-feathers and spo<-feathers is invariably accomplished by cutting off terminal 

 parts of feathers. Ring-feathers are shorter than other neck-feathers, and look as 

 if cut off with shears. When a feather is differentiated only on one web, the other 

 web is seen to be much the longer. 



In this same normal male (Geopelia striata) we find the following facts concern- 

 ing the number of bars in the individual feathers (pi. 37, figs. 1 to 7) : 



(1) A feather from the front edge of the barred part of the neck, at about mid- 

 height, shows four black bars (not counting the apical black, which is almost absent ; 

 it is barely suggested to the naked eye, but evident with magnifying glass). 



(2) A feather from the middle of the left side of the neck shows the apical black 

 bar quite plainly, though it is tapering and very thin. Besides this bar there are 

 two other black bars, wider than those in No. 1 above, and then a third black bar 

 that is very weak (pale gray). Another feather from the same region would show 

 about the same, except that the third black bar is stronger. 



(3) The feathers just under the ear have the apical bar very thin, but visible 

 to the naked eye; then follow a white bar, a plain black, a white, and finally a weak 

 but evident dark bar. 



(4) Following from the back up along the mid-hind neck, 39 we find, to begin 

 with (on back), feathers which have only the apical black bar; then others with 

 1 extra black, then 2 extra black, then 3 extra black; still further along, 2, and at 

 the upper limit, only 1 extra black bar. 



The front feathers are lighter clearer white and black and the back of the 

 neck is darker, the white bar being replaced in the hind-neck with pale brown or 

 fulvous. This is as in Geopelia tranquilla. The rudimentary apical black bars of 

 the neck are most reduced at the middle height of the front edge of the neck; these 

 become stronger as we go from here to the mid-back of the neck. This agrees 

 closely with G. tranquilla. In G. striata I find about one more black bar to the feather 

 than in corresponding feathers of G. tranquilla. 



In Geopelia tranquilla the bars run all around the neck (pi. 39, fig. A). Those 

 on the front half of the neck are of lighter color clearer white and clearer black 

 than those on the back half; this of course harmonizes with lighter under parts and 

 darker upper parts of the body as a whole. 



Leaving the rudimentary apical black bar out of account it is so reduced in 

 front and on the sides as to be obsolescent the rule is that we see two black bars 



' The fate of the terminal black bar in this series of feathers was noted above. 



