i%o.] Chapman o?i the Winter Distribution of the Bobolink. A J 



duras we also have another record by the same author,* to 

 which I will refer later, and we find the species included by 

 Zeledonf in his list of the birds of Costa Rica. Swainson'sj 

 record from the highlands of Mexico, which probably refers to 

 the western race, D. o. allunucha, completes the Mexican and 

 Central American records with which I am familiar. 



Turning now to the second route, leading through the Wind- 

 ward Islands, we find that from Hayti, San Domingo, or Porto 

 Rico we have as yet no notice of the bird's appearance. Our 

 first record, therefore, is from the small island of Sombrero in the 

 Virgin group, where Lawrence§ records it, and in the Catalogue 

 of the British Museum || a specimen is cited from Nevis; both 

 records are without data, but there is a specimen in the American 

 Museum labelled, "Sombrero, Sep. 17, 1S62, Julien," on which 

 the first was probably based. From Guadeloupe and Martinique 

 the bird is given by L'Herminier,^" and this completes the 

 Windward Island records until we come to the last of the chain, 

 Grenada. From this island we have some very welcome and 

 valuable information by Wells,** who during three years' obser- 

 vations had met with the species on only one occasion, when he 

 captured one from a flock of five birds. Reference to this speci- 

 men in the American Museum collection shows it to have been 

 taken October 1, 1S85. 



Passing now to a consideration of the data from the third, or 

 last route mentioned, we find that great numbers of birds select 

 this more direct line of migration. After leaving the island of 

 Cuba the birds' first resting place would be the Caymans, distant 

 about one hundred and seventy-five miles. Mr. Cory writes me 

 that his record of their occurrence on. Little Caymanff is based on 

 two specimens, a male taken April 29, and a female taken April 

 30, and while it is probable that they also visit these islands in 

 the fall, we may presume that the greater numbers follow the 

 Cuban coast to at least Cape Cruz, or to a point directly north of 



* Ibis, 1866, p. 194. 

 •- + Proc. U. S. Nat. Mus., VIII, 1885, p. 107. 

 \ Phil. Mag., New Ser., I, p. 435. 

 \ Ann. N. Y. Lye, 1S64, p. 99. 

 || Cat. Brit. Mus., XI, p. 332. 



IT Cf. Lawrence, Proc. U. S. Nat. Mus., 1, 1878, p. 450. 

 ** Ibid., IX, 1886, [* 616. 

 ttAuk.VI, 1889, p. 31. 



