NEOTROPICAL EEGIOX.] BIRDS and St Bartholomew, we find that this line divides the chain into two groups of distinct character that lying to the southward and eastward, almost identical with the " Windward Islands " of some geographers, 1 which, except ing Antigua and Barbadoes, are almost entirely volcanic, while no direct trace of recent volcanic action is known in the group lying to the northward and westward. These three divisions, however natural in appearance, can hardly be affirmed to form as many zoological provinces, owing to the absolute dearth of information respecting many of them, and the insufficient amount which has been received of the remainder. Taking the whole of the Antillean Subregion, the Bahamas, Cuba, Jamaica, St Croix, St Thomas, Som brero, St Bartholomew, Guadeloupe, Dominica, Martinique, and St Lucia are the only islands of the ornithology of which we have anything more than what may be called a casual account ; and hence, though valuable observations respecting some of the rest have been placed on record by travelling naturalists, any attempt to separate the Sub- region into proper provinces would necessarily be of the crudest kind, and here cannot be made, though doubtless such provinces will eventually be defined with precision. er- Still enough is known of the Birds of this Subregion to "i~ enable us to draw some conclusions, though certain of them seem especially likely to be overset by further in vestigation. It is inhabited by some 41 families, 1 of which (Todidce) is not found elsewhere, 2 more (Coerebidce and Cotingidce) are confined to the Neotropical Region, 8 (Mniotiltidce, Vireonidce, Tanagriclce, Icier idee, Tyrannidce, Trochilidce, Gathartidce, and Aramidae) are common to that and the Nearctic Region, but are peculiar to the New World, and 30 are of much more general distribution, but 2 of these (Trogonidce and Fregatidce) are not found in the Nearctic Region, and, as the above numbers show, there is no family common to that Region and the Antillean Sub- region without also occurring in other parts of the Neo tropical Region. On the whole, therefore, the affinity of the Subregion to the Neotropical rather than to the Nearctic Region is fully made out. About 1 40 genera are found in the Antilles, of which 30 are peculiar to it, being a considerably larger proportion than is elsewhere found in the Neotropical Region. Of these 30 genera, which are all Land-birds, 4 belong to Turdidve, 1 to each of Mniotil- tidoe, Vireonidce, Ampelidce, and Coerebidce, 2 to Tana- gridoe and Emberizidce respectively, 1 to each of Icteridce and Tyrannidce, 5 to Trochilidce, 1 to Caprimulgidce, 2 to Picidce, 1 to Todidce, 2 to Trogonidce, Cuculidce, and Strigidoe respectively, and 1 to Columbidce? Be sides these, 21 more, which it is perhaps unnecessary to particularize, do not exceed the limits of the Neotropical region, while 5 others, belonging respectively to the families Cypselidae, Fregatidce, Anatidce, Columbidce, and Rallidw, occur both here and elsewhere in that Region without reaching the Nearctic except as stragglers. Peris- soglossa, belonging to the family Mniotiltidce, is the only genus common to the Nearctic Region and the Antillean Subregion without occurring elsewhere in the Neotropical Region, all the remaining Antillean genera inhabiting both North and South America. The distribution of some of 1 In the language of others, the " Windward Islands " begin" with Martinique and so continue to Trinidad, while the remainder of the Lesser Antilles, lying eastward of Porto Rico, are called the " Leeward Islands." Other authorities name all the smaller islands so far as St Thomas "Windward Islands," and those further westward the " Leeward." Creoles appear to apply these terms relatively to their own habitation, just as the dweller on the bank of a river speaks of " up stream " and " down stream " with sole reference to the position he occupies on the margin ; and it might be wished, though that were vain, that the use of terms so little definite should be abandoned. 8 This genus (Starncenas) has been said, however, to occur in the Florida Cays, but the statement seems doubtful. the peculiar genera merits a little attention, for 19 out of the 30 are confined to a single island, or nearly-connected group of islands. Thus, Cuba has 6 believed to be re stricted to its soil ; Jamaica, 7 ; Hispaniola, 2 ; and the so-called " Windward Islands," probably 4 ; while none are known to be absolutely limited to the Bahamas, to Porto Rico, or the Virgin Islands. It is possible, indeed, that Hispaniola and Porto Rico, if as well explored as Cuba and Jamaica have been, might tell a very different story. Pursuing the subject further, and entering, so far as Resident space will allow, upon a consideration of details, we find a <l mi - that there are of Land-birds about 200 resident species, and o railt nearly 90 which are not resident, but migratory. These ^ last belong to about 55 genera, of which some 40 have no resident insular representatives, while those migrants per taining to genera which possess permanent residents are nearly all as much Neotropical as Nearctic in character. Most of these, so far as is known, visit Cuba only, where over 80 are recorded as occurring, while but 30 reach Jamaica. The number, however, in the island last named, and in others, would most likely be not inconsiderably increased did competent observers but exist, though the fact that a well-known species like the Humming-bird of eastern North America (Trochilus colubris) does not seek its winter-abode in any of the islands except the Bahamas and Cuba, tends to lessen the force of such a supposition, and points to our actual knowledge being not very far wrong. Among the more interesting of peculiar forms is one genus of Turdidce (Mimocichlci), which is represented by 4 distinct species, found in the Bahamas, Cuba, Jamaica, and Hispaniola respectively. Another genus of the same family (Margarops) has a species ranging from Hispaniola to the Virgin Islands, and a second species inhabiting Martinique and St Lucia only, these two islands possess ing at the same time a third and peculiar genus (liham- phocinclus), containing but a single species, common to both, while they also have another genus (Cinclocerthia), a distinct species of which inhabits either island, though a third species of the same is found in Guadaloupe and Nevis. The genus Certhiola, belonging to the Coei-ebidce, is remarkable in that it occurs in nearly every island ex cept Cuba ; but what seems to be still more extraordinary is that the species found in the Bahamas (G. bahamensis), and there alone in the whole Subregion, also occurs in Cozumel, an island off the coast of Yucatan, though Cuba intervenes. The distribution of Spindalis, one of the Tanagridce, resembles that of Mimocichla, above noticed, only that here Porto Rico also has its distinct representa tive species. Of the peculiar genera of Humming-birds, Eulampis has 2 species one ranging from St Lucia t< Nevis, but the second extending northward to St Thomas. AitJmrus, a very remarkable form, is restricted to Jamaica, while Mellisuga is common to that island and Hispaniola each of these genera consisting of but one species only, and the last is the smallest known bird. Orthorhynckus seems to have 3 species, one extending from St Thomas to Dominica, a second (perhaps barely separable) common to Martinique and St Lucia, and a third to St Vincent and Barbadoes ; and, finally, Sporadinus has one species in Cuba and at least one of the Bahamas, a second species in Hispaniola, and a third in Porto Rico. Of genera of Humming-birds, which are not confined to the Antilles, Lampornis, a widely-ranging genus, has two species peculiar to Jamaica and Porto Rico respectively ; while a third extends from Hispaniola to St Thomas. Calypte, which has two species in Mexico, has a third peculiar to Cuba, while Doricha is represented in two of the Bahamas (New Providence and Inagua) by as many distinct species; the other three described species of the form inhabiting
Central America, and none, so far as known, occurring ia