1917.] Chapman, Distribution of Bird-life in Colombia. 503 



Char, suhsp. — Similar to Phceoprogne tapera tapera (Gmel.), but underparts 

 without the median line of spots, guttate or broad central markings, which reach from 

 breast to belly in that species; the pectoral band generally more pronounced. 



Found by us only in the Tropical Zone of the Magdalena Valley and in 

 the Caribbean coast region. It was not secured in the Cauca Valley, nor 

 on the Pacific coast, but Mr. Cherrie and I secured specimens at Duran, 

 near Guayaquil, during a brief stop there in May, 1916. 



The receipt of additional material since this bird was described, places 

 its status in doubt. At the time of description I had seen only one bird 

 from north of the Amazon which showed the median pectoral spots char- 

 acteristic of P. 4. tapera; now, however, we have two mofe collected by Mrs. 

 Kerr on the Sinu River, in northern Colombia, which are apparently typi- 

 cal of the Brazilian race, that is, have a median row of larger, well-defined 

 fuscous spots from the breast to the abdomen. The occurrence of these 

 specimens at the northern limit of the range ascribed to the form which I 

 have described as immaculata, might be thought to invalidate the claims 

 to recognition of that bird. The fact remains, however, that evidently 

 without regard to age, or season, we have two quite unlike types of Phceo- 

 progne tapera, one of which is the characteristic form in the southern part of 

 the range of the species, the other characteristic of the northern part. Thus, 

 every one of twenty specimens from southern Brazil has the median line of 

 spots strongly developed; while sixteen of nineteen specimens from Vene- 

 zuela, Colombia and western Ecuador, are without these spots. 



I confess I cannot explain this unusual state of affairs. The character 

 by which these forms are differentiated appears to be of specific, rather 

 than of subspecific value. There is no intergradation between the two 

 types such as we find in representative geographic races. The two north 

 Colombia specimens are as typical of the spotted-breasted race as are birds 

 from southern Brazil. Possibly they may, independently of environment, 

 spontaneously exhibit a character which has been fixed in all the more south- 

 ern individuals of the species, and which in time may be found in all the 

 northern individuals as well. The case is as interesting as it is exceptional, 

 and it is greatly to be hoped that material with which to explain it may soon 

 be forthcoming. Meanwhile, it seems desirable to recognize, at least pro- 

 visionally, the race for which I have proposed the name immaculata, and as 

 an aid to the elucidation of the problem it presents, I add a list with data 

 of all our specimens of both forms. 



I. Specimens with the underparts spotted (P. t. tapera). 



Argentina: Embarcacion, 1 9 . 



Brazil: Chapada, Matto Grosso, Jan. 1, 9 ; Feb. 2, unsexed; Sept. 1, cf , 



