THE TUKTLE-DOVE PATTERN IN THE PHYLOGENY OF PIGEONS. 



95 



Spots or marginal streaks arc recapitulated in the tertials, scapulars, and coverts 

 of the juvenal wings (pi. 41, A and B, and text-figs. 20, 21). A comparison of these 

 streaks in the tertials of humeralis and the mourning-dove (Zenaidura) is made 

 possible by figures 3 to 5 of plate 41. 



The homology of the wing-chequers of the juvenal geopelias with the chequers 

 in Columba livia has been pointed out in the preceding chapter. In all of the geo- 

 pelias 41 these chequers are completely lost with the first feathers. 42 



In the ontogeny of the geopelias we see the different species all taking departure 

 from a common color-pattern, consisting of longitudinal spots or chequers, dis- 



TEXT-FIOUBE 21. Wing of juvenal 

 Geopelia humerali.i, age 10 weeks. 

 Natural size. Hayoshi del., Aug. 

 1899. 



The first row of coverts retains 

 only 4 feathers seen in the previous 

 (30-day) stage, and these are bo.ing 

 covered by new black-tipped feathers. 

 Only 3 old feathers remain in the 

 scapulars. 



TEXT-FKIURE 22. Wing of 

 adult Geopelia humeralis. 

 Natural size. Hayashi del., 

 June 1899. Drawn for 

 comparison with wings of 

 the young. 



tributed uniformly to all the feathers of the wing. We see this common and earlier 

 pattern retained for only a few weeks, the later pattern developing in two widely 

 diverging directions and leading to two types of generic rank. 



THE PHASING. 



The relation of the color-pattern of two of the genera of this subfamily lias 

 been already described Ocyphaps in ( 'hapters II and III (pis. 8, 14, 15, 16; Phtipx 

 in Chapter II (pi. G). (A short treatment of three other genera- ('<il<ij>cli<i, (End, 

 Tympanistria of this group is extracted from a short paper without illustrations 

 published 13 in 1907. ED.) 



41 Ceii/H liu or N//V/<i/Wm ciinmtii is described and figured in Chapter IX. ED. 



41 It has been observed thai I lie feathers of the oblique streak do not make their appearance at exactly the same 

 age in the several species of (iid/ulid. The order is as follows: (1) (}. tranqtritla: streak out at 17 days and a 

 second row with black tips begins; (2) (!. strinta: streak just begins at 18 days; (3) (1. ritncatii: two rows well 

 started at lit days; (4) (/. hiinirrnlix: begins about 19 to 21 days; at 28 days one row and three more black-edged 

 feathers (some with very narrow light edge'. 



a Avicultural Magazine, London (Journal of the Avicultural Society), n.s., vol. 5, No. 6, April 1907. ED. 



