560 Annals of the Carnegie Museum. 



Measurements. — Male (eighteen specimens) : wing 77-83 (average, 

 80); tail, 53-63 (57.5); exposed culmen, 10. 5-12 (il.l); tarsus, 15-17 

 (16). Female: wing, 77-85 (80); tail, 51-62 (56); exposed culmen, 

 n-12.5 (11.3); tarsus, 15-17. 5 (I57). 



Range. — Island of Porto Rico, and Lesser Antilles south to Santa 

 Lucia. 



Remarks. — That two forms of the Ground Dove inhabit the Lesser 

 Antilles is perfectly apparent upon even a casual comparison of a 

 series from the more southern islands of the group, Grenada, Car- 

 riacou, Bequia, etc., with another from Antigua and Barbuda, the 

 general differences being well marked. Unfortunately from Mar- 

 tinique, the type locality of Chamapelia trochila Bonaparte, no adult 

 males are available, but two females examined seem referable to the 

 northern form. Moreover, as previously noted, while birds from 

 Santa Lucia are somewhat intermediate, a series from Dominica is 

 quite obviously closer in general coloration to the Antigua and Barbuda 

 birds. In view of these considerations it seems quite safe to adopt the 

 name trochila, based on the bird of the intermediate island of Martin- 

 ique, for the northern form. It is simply one of the many cases in 

 which, paradoxically speaking, the bird from the type locality is not 

 typical ! 



The character of the extent of the rufous chestnut on the outer 

 webs of the remiges, to which Mr. Riley has called attention, is one 

 upon which no dependence can be placed so far as C. p. passerina is 

 concerned, as previously shown, and this is true also of the other 

 forms so far considered. It is somewhat surprising, therefore, to find 

 that in this particular case it is a fairly constant diagnostic character 

 in both sexes, serving to separate the two Lesser Antillean forms 

 from each other. 



This form varies to some extent geographically, but scarcely 

 -enough to justify further rsubdivision. Birds from Dominica, for 

 example, show less vinaceous tinge on the wing-coverts than do those 

 from farther north, the approach of course being to C. p. antillaram. 

 On the other hand, St. Thomas examples are rather more brightly 

 colored in this respect. After careful comparison of a good series 

 from Porto Rico I believe they should also be referred to this form, 

 although the males differ in being rather more "solid" vinaceous below 

 — due to the slightly darker tips of the throat and breast feathers — 

 while the wing-coverts average more decidedly vinaceous, and the 



