1896.] **' [Cushiug. 



nection, the number of correspondences between the Arawak and the 

 Timucua and between the Timucua and Maskoki, being, for example, 

 about equal, and quite as readily explicable in both cases on the score 

 of acculturation or borrowing, as on that of descent. It is for this reason 

 that I have regarded archieologic evidence on this question of connec- 

 tions, as equal to, and in some waj's superior to, linguistic evidence ; and 

 a combination of tlie two kinds of testimony as superior to either. Wlien, 

 for instance, we find that the same word in both Carib and Timucua sig- 

 nifiesnotonly "Fish-pond" butalso " Vegetable garden," and when we 

 consider this in connection with the evidence I discovered on all the 

 ancient keys, of the actual filling in offish-ponds or enclosures to form 

 of them vegetable gardens, it seems to me we have quite strong indica- 

 tion of a wide-spread practice, commonly derived, by all these peoples. 



If the linguistic evidence relative to connections either toward the 

 north or toward the south, of the ancient key dwellers, is thus far so 

 scant as to be inconclusive, this is to a certain extent also the case with 

 the evidence afforded by the human remains we collected. In justice 

 to Dr. Putnam I must state here that the series of skulls in my collec- 

 tions, examined by him, were not the key-dweller skulls. They were 

 skulls derived from the Anclote region, and like those he mentions as 

 previously collected by Dr. Wyman and Mr. Clarence Moore, were 

 exhumed from sand mounds. The true key-dweller skulls found by us 

 in the muck beds at Marco and in the bone pit on Sanybel Island, num- 

 ber only thirteen, but they are pronounced to be, by Dr. Harrison 

 Allen, who is studying them preparatory to full publication, uniformly 

 distinct from those of more northerly and easterly parts of Florida. In 

 the first place, the occipital foramime of these remarkable skulls are 

 abnormally large and remain open in even the most mature of them, — a 

 characteristic seen in only one cranium of our northern series. In the 

 second place, a curious feature of all these key-dweller skulls is that in 

 no case is the occiput flattened. Finally, they are found to be more 

 nearly of the Antillean type, judged, it is true, by only one or two 

 specimens of the latter examined by Dr. Allen, than of the northern 

 Indian type. 



In connection with this, it is significant that the skulls of two dogs, in 

 our collections from the muck, were commented upon by the late Prof. 

 Edward D. Cope, as apparently, almost certainly, skulls of the species 

 of dog common in Incan times to the Peruvian and Bolivian Highlands. 



Likewise in justice to Dr. Putnam, I must again state here that while 

 there icas pottery not only on the terraces, but also in the muck depos- 

 its, of the keys, even of the southernmost keys I examined ; still, the 

 specimens I exhibited before the Society — three in number — so closely 

 resembled the wooden objects of the same general kind, also exhibited 

 and in greater number, that they may well have been mistaken for 

 vessels of wood unless particularly dwelt upon. It is a curious fact 

 that of all the pottery discovered by us actually in the muck deposit of 



