Kroeber's list shows 108 towns plus 10 mentioned by Curtis as being in Karok territory on the Salmon River. The first 84 villages were covered by Ned, who gave house counts for 61 of them. Using wherever possible the houses actually seen, not merely heard of, by Ned we get a total of 248. This is a little smaller than Kroeber's total for the same sites of 254. In this group of 84 villages 9 have counts from Curtis but not from Ned, with a total of 24 houses. By Kroeber's own showing Curtis' counts are as reliable in the aggregate as those of Ned. Sites 85 to 108 are derived only from Mrs. Jacops who did not give counts. Kroeber proposes (pp. 34-35) to reduce these to 15 settlements and assign an average value of 4 houses per village. This seems entirely reasonable, and gives us 60 houses. We may now add the 10 villages on the Salmon River cited from Curtis by Kroeber and, to be conservative, assign an average count of 3 houses each. The total of all Karok houses then becomes 362. At the customary 7.5 persons per house the population of the Karok is 2,715. or with sufficient accuracy, 2,700. _KAROK ... 2,700_ THE HUPA There are four sources of consequence for the Hupa population. The first is the discussion to be found on pages 128 to 132 in the Handbook by Kroeber, which includes a census furnished to the government by the Yurok in 1851. The second is a monograph published by Goddard (1903). The third is a particularly exhaustive village list compiled by Merriam. The fourth is a map drawn by Gibbs in 1852, photostatic copies of which are to be found in the Merriam collection. The towns of the Hupa fall naturally into two subdivisions, the first comprising those in Hupa Valley proper and the second those above the valley which extended along the main Trinity River and its South Fork. The first included 12 villages which are mentioned by name by Goddard and are shown on his map. For most of them he indicates houses by dots and solid squares which can easily be counted. Kroeber lists on page 129 the same 12 towns and all but one of them appear on Merriam's list. (These are numbered 1-12 in table 5. p. 100, herein.) Village no. 2, Dakis-hankut, is omitted by Merriam but is shown with houses by Goddard. Village no. 8, Totltsasding, is stated by Kroeber to have been "unoccupied in 1850." Goddard however merely says that it had been deserted for a long time. On the other hand it had been sufficiently well known to the Yurok for them to have a name for it, and Merriam does not question its existence. These two villages may therefore be retained in the list.