values computed from the statements of Johnston, Wessels, and Savage are too high, that derived from Ryer may be somewhat too low. An intermediate figure of 70 inhabitants per village for the valley and lower foothills would perhaps come as close as we can get to the truth. This, with 49 villages, gives 3,430, somewhat more than the 2,633 cited as the average of the general estimates.

Inhabiting the higher foothills and extending to the upper limit of habitation from the San Joaquin to the Kaweah rivers were the Western Mono. This tribe lived just above the Yokuts and at points was in very close association with them. As a whole the Western Mono constitute a racial and ecological unit and as such it is probably preferable to consider them as a single population entity than to segregate them by rivers, as has been done for the Miwok and the Yokuts. It will be necessary, therefore, to digress for this purpose and subsequently return to the discussion.

The classic ethnographic work on the tribe, and the only work which contains any numerical data, is that of Gifford (1932) on the North Fork division of the Mono. This is supplemented by Merriam's manuscript entitled "Monache Tribes, Bands, and Villages." Gifford gives the names (text and map) of 67 North Fork villages, or, as he prefers to call them, hamlets. These were quite unlike either those of the Miwok or of the valley Yokuts, being very much smaller and subject to an extraordinary turnover in inhabitants. Gifford makes it very clear that each family was accustomed to move every few years from one settlement to another and that sites were being continually occupied and deserted. The 67 names are therefore no criterion for population. For the time of the American occupation Gifford estimates the number of persons in the group or subtribe as approximately 300, which, divided directly by 67, would give the absurd average of 4 persons per hamlet. However, a more detailed analysis is possible.

Of Gifford's 67 names, 2 may be deducted as being only camps, leaving 65 which at some period were permanently occupied. In his Appendix A (pp. 57-61) he lists the sites, together with the number of houses in each and the number of males and females inhabiting them. From these data may be computed the total number of families and the mean number of persons per family. There were 227 families in all. However, 36 of these are listed two or more times by virtue of moves made from one hamlet to another, which were remembered by Gifford's informants. This would leave 191 families for the subtribe, provided Gifford recorded all the moves. But Gifford clearly implies that he did not, since his informants could not remember them all. Hence the number of families must be further corrected. In Appendix A, 15 out of a total of 65 hamlets were concerned in the moves recorded. These 15 hamlets were inhabited at different times by 61 families but many of these, owing to frequent change of residence, are repetitions. Actually there was a total of 24 different families rotating among the 15 villages. Now if in the other 50 hamlets the same process was going on, although Gifford was not able to record the moves, it is legitimate to apply the same ratio as is in fact found for the 15 hamlets. The crude total of 227 families must therefore be reduced to 89. From Gifford's complete list it can be determined that there were on the average 4.93 persons per family. This gives a population of 439 for the period remembered by the informants.

On general grounds it is to be expected that the conditions reported by Gifford's informants were not entirely aboriginal. This is also indicated by the value of 4.93 persons per family, which is somewhat too low for a stable prehistoric population. Moreover, Gifford himself states that there were formerly 44 more houses than there were in the time referred to by the informants (figures given individually for the hamlets in App. A). About 1850 there were 227 houses, and if 44 are added, the aboriginal number would have been 271. Each house may be assumed to have held one family but the houses were probably occupied in rotation. The crude estimate of 271 houses or families, each containing (according to aboriginal standards) a possible 6 persons, would mean a total of 1,626 for the subtribe. If, however, we apply the correction factor for family moves we must reduce this estimate to 640, a far more reasonable figure. For the North Fork Mono, therefore, we may accept as the best estimate obtainable a population of 440 for the period near 1850 and of 640 for precontact time.

The other subtribes of the Mono provide no data comparable with those available for the North Fork group. Some method of extrapolation is thus called for.

The village method is very unsatisfactory. Kroeber says substantially nothing on this score and Merriam, although he lists 19 villages for the North Fork Mono, gives no more than one or two or, at the most, half-a-dozen names for each of the other groups. Tribal distinctions are also very confusing. Kroeber in the Handbook mentions 6 Mono subtribes: North Fork group, Posgisa, Holkoma, Wobonuch, Waksachi, and Balwisha. Merriam subdivides to a much greater extent. His grouping may be expressed essentially as follows: