74 



ORTHOGENETIC EVOLUTION IN PIGEONS. 



marked edge — pale yellowish-brown crescents — on all the feathers; these, however, 

 are best marked on the wing. The rest of the middle part of the feather (exposed 

 part) is a uniform brown, darker in the male (pi. 21, Vol. II), lighter in the female. 



I have frequently noticed further that the young birds, in many of the wing- 

 coverts and scapulars, have the dark pigment more concentrated on the inner edge 

 of the light margin, giving an appearance that reminds of the dark crescents within 

 the pale margins in the young geopelias. 23 Proximally of this dark bar I usually see 

 an irregular but plain light transverse band — a kind of imperfect or incipient bar. 



Again, the vinous red of the adult seems to be an extension of the reddish edges 

 of the turtle-dove's feathers. Lifting the feathers, we see that their basal portions 

 are dark gray, the dark graj' frequently extending farther on one web than on the 

 other. The gray center has simply been reduced so that the uncovered part of the 

 feather is all vinous red. 



The juvenile feathers of this species give strong evidence of its turtle-dove an- 

 cestry (pi. 21, Vol. II): 



(1) The pale yellowish-brown edges of the feathers incline more to red on the 

 primaries. 



(2) The neck-spots are always present, sometimes quite strong; but they never 

 meet on the back of the neck. 



(3) The mid-ribs of the feathers (coverts of wing) arc distinctly dark — a final 

 stage in the loss of dark centers, as shown in hybrids of ring-doves and turtle-doves. 



(4) One feature is particularly significant. On the tertials (upper inner second- 

 aries) and on the third or fourth posterior rows of coverts we see that the lateral 

 edges as well as the tips are pale; but the lateral edges are decidedly more reddish 

 than the tips, quite closely following the plan seen in the Japanese turtle-dove young. 

 The color is here not so strong as in the turtle-dove, and it is not so well delimited 

 towards the central field of the feather. In fact, it extends inward rather diffusely 

 to or near the mid-rib on the lower web more than the upper (inner) web. There 

 is a subterminal concentration of dark pigment following the light edge, as if light 

 and dark alternated in the manner of transverse bars. 24 



The probable relationship of several genera of pigeons is indicated in the fol- 

 lowing diagrams: 25 



T. orientalis 



T. turtur 



St. humilis 



'St. douraca 



'Spil. tigrina; Spil. suratensis ^\ St. risoria 



'St. alba 

 Spil. chinensis 



Columbidae (e. g., C. palumbus) 



Derivation of species of TurturinaV' 9 and part of Family 

 lu Combidae. 



Turtle-dove type 



/ 1 



Nesopelia 



I 

 Zenaidura 



I 

 Zenaida 



' Chamaepelia 



Columbidae (e. </., Ectopistes migratorius) 



Melopelia 



Derivation of genera of Zenaidinse and part of Family 

 Columbidae. 



a This appearance is also strikingly simulated in the dark borders to the red chequers borne by certain homers 

 I have seen. 



u SHgmatopelia senegalensis, not described here, is figured in plate 27, Vol. II. — Ed. 



26 These diagrams have been constructed by the editor. The materials used in their preparation are as follows: 

 The statements already given in this chapter; data found elsewhere in this volume; data given in Vol. II — particularly 

 in Chapters XII, XI 11, XV, and XVII, and in the explanations of plates illustrating those chapters; data on behavior 

 in Vol. III. Ed. 



26 Professor Whitman was convinced that all doves having a "coo" of three syllables are derived from the Japanese 

 turtle. — Ed. 



