For the first 11 subtribes Merriam gives a total of 46 villages. Of these, 30 can be identified with names furnished by Goddard, whereas 16 are in addition to Goddard's list. Goddard on the other hand gives 23 which were not secured by Merriam. Since both these workers operated carefully through informants and both were thoroughly conversant with the local dialects, we may accept the combined total of 69 villages, large and small, occurring within the territory of Goddard's first, or southernmost, 11 subtribes. The average is 6.27 villages per subtribe. For the Chiskokaiya and the Kaikichekaiya, Merriam mentions only one village each, that bearing the subtribal name. It is quite clear from his list that he did not push his field investigations into these groups. Hence we must fall back on Goddard's data, which include 18 villages in all. The average for the 13 subtribes therefore is 6.7 villages, and the total 87. All the villages have long since been totally deserted and Goddard could count only house pits. (Merriam made no counts of any kind.) He did this for only two groups, the Baskaiya and the Slakaiya. Here he found and mentions on pages 103 and 105 twenty sites containing house pits. In all there were 92 pits but for two localities he specifies a certain number plus "several" others. If we allow 4 to represent "several" in each of these, then, the total number of pits is 100 and the average per site or village is 5.0. Now since we are dealing here only with pits and not counts of houses remembered by informants, a reduction according to the Kroeber principle is justified for it is quite probable that all the houses once standing on the pits were not simultaneously occupied. When Kroeber has no other data, he recommends a reduction by one-sixth. I think that in this instance it would be proper to reduce by one-fifth, or 20 per cent. This would give an effective average of 4 houses per village. In the 13 communities covered by Goddard and by Merriam there were 87 villages, which at 4 houses per village would give a total of 348. No evidence is offered by either author to the effect that the remaining 5 subtribes differed in any essential way from the first 13. Hence we must ascribe to them 134 houses, making 482 in all.