As already stated, the "best expert authority" of the Bureau of Ethnology reads Onekagoncka as signifying, "At the junction of the waters," and Osseruefion, in any of its forms, as signifying "At the beaver-dam." Possibly the names might be read differently by a less expert authority, but Oneka certainly means "Water," and Ossera means "Beaverdam." Add the reading by the late Horatio Hale of Oghracke^ "At the beaver-dam," and the locative chain is complete at the mouth of Aurie's Creek (Oghracke). Trihally, the names referred to one and the same castle, as has been noted, and the evidence seems to be clear that the location was the same. There is no evidence whatever that any other than one and the same place was occupied by the "first castle" between the years 1635 and 1667. It is not strictly correct to say that "castles were frequently removed." Villages that were not palisaded may have been frequently changed to new sites, but the evidence is that palisaded towns remained in one place for a number of years unless the tribe occupying was driven out by an enemy or by continued unhealthfulness. as the known history of all the old castles shows ; nor were they ever removed to any considerable distance from their original sites. Van Curler's description of the castle has been quoted. He did not say that it was palisaded, but he did call it a "fort," which means the same thing. Rev. Megapolcnsis wrote, in 1644: "These [the Tortoise tribe] have built a fort of palisades and call their castle Assarue." It was not an old castle when Van Curler visited it in 1635, or when Father Jogues was a prisoner in it in 1642, but