[68] Progne, (subis var?) leucogaster. Progne leucogaster, Baird, Rev. Am. B. 1865, 280. (Southern Mexico to Carthagena.) Progne dominicensis and P. chalybea, Auch. (nec Gmel.).
From a careful examination of specimens of the above forms, the opinion that they are all local differentiations of one primitive type at once presents itself. The differences from the typical subis are not great, except in the white-bellied group (dominicensis and its allies), while an approach to the white belly of these is plainly to be seen in P. cryptoleuca; again, some specimens of dominicensis have the crissum mixed with blackish, while others have it wholly snowy-white. While the male of cryptoleuca is scarcely distinguishable, at first sight, from that of subis, the female is entirely different, but, on the other hand, scarcely to be distinguished from that of dominicensis and leucogaster. Adult males of the latter species are much like adult females of dominicensis, while Floridan (resident) specimens of subis approach very decidedly to the rather unique characters of elegans. It is therefore extremely probable that all are merely local modifications of one species.
[69] C. cyaneoviridis, Bryant; Baird, Rev. 303 (Bahamas).
[70] Vireosylvia calidris, Baird, Rev. Am. Birds, 1865, 329. (Motacilla calidris, L. Syst. Nat. 10th ed. 1758, 184.)
[71] V. calidris var. barbadense, Ridgway.
[72] V. olivacea var. chivi. Vireosylvia chivi, Baird, Rev. 327. (Sylvia chivi, Vieill. Nouv. Dict. XI, 1817, 174.)
[73] V. flavoviridis var. agilis. Vireosylvia agilis, Baird, Rev. 338. (Lanius agilis, Licht. Verz. Doubl., 1823, No. 526.)
[74] V. magister, Baird.
[75] V. gilva var. josephæ. Vireosylvia josephæ, Baird, Rev. 1865, 344 (Vireo josephæ, Sclater, P. Z. S. 1859, 137, pl. cliv). Comparing typical examples of this “species” with those of gilvus from North America, they appear very widely different indeed, so far as coloration is concerned, though nearly identical in form. But a specimen from an intermediate locality (54,262, Orizaba, Mexico, F. Sumichrast) combines so perfectly all the characters of the two, that it would be impossible to refer it to one or the other as distinct species. It therefore becomes necessary to assume that the V. josephæ is a permanently resident tropical race of a species of which V. gilvus is the northern representative; which theory is strengthened by the fact that of the latter there are no specimens found south of the United States, indicating that in winter it does not pass beyond their limit, or at least not far to the southward.
[76] The Jamaican bird is V. calidris, not barbatulus. In all probability, however, they do not differ in habits and notes.—R. R.