sidered a national term they applied to themselves, and if they may be supposed traceable to Africa, this name may be connected with that of the people still known on that Continent as Karabas, the Efik tribe on the old Calabar and cross rivers. I suspect the name of the river Calabar itself is connected with theirs, the letters l and r being so commonly convertible. The Efik tribe with their ramifications are widely extended over the neighbouring districts, with the Mokos inland and the Ibas on the Delta of the Niger to Yoruba, Igarra, Apa and Itu. "Human sacrifices, slavery and cruel superstitions are said fearfully to prevail among them" and "very many of the tribes are yet there addicted to the practice of Cannibalism," Clarke's Specimens of Dialects, p. 86. In Cuba they are known as Carabalis, and are looked on by the other negro slaves with dislike on account of this propensity alleged against them. It is with the language also of this same people that in my former Paper I compared the Carib words given by Rochefort and Le Breton, showing a connection from that almost certain criterion of the relationship that existed between them, and confirmed by the national appellations and characters. It may prove true that the languages of Africa and that of the Caribs, as given by Le Breton, may have been very dissimilar. But the Vocabularies show the words of primary necessary use to have been almost identical, and this is evidence of a commixture at least, which is all that I contended for. Such words will cling tenaciously to the memory of the adult savage, even though he may have forgotten the verbal conformations of his mother tongue among people speaking a different language. Nay such verbal conformations I have no hesitation in saying may in certain circumstances be adopted as to alter materially the original language, and the Caribs in the West Indies driven or drifted over from Africa might thus have spoken a language in 1650 modified accordingly from what they brought over 200 years previously. Let us always bear in mind that in the present state of Ethnology we are only collectors for our successors, who from our labors may verify as facts what we can only put forward as conjectures. With this consideration as our guide I propose to give as another Appendix to this paper a condensed translation of the Carib Grammar of Le Breton compiled 200 years since, which is now so rare, as to command an exorbitant price, to be compared by those who feel an interest in the question with the modern languages of Africa, and also with that of the modern Caribs about to be published by this Society from the labors of my va-
