The earliest serious effort to estimate the aboriginal population of California was made by Powers (1877, pp. 415-416), who arrived at a figure of 750,000 persons for the entire state. This effort was followed in 1905 by a more sophisticated attempt on the part of C. Hart Merriam, whose figure for the state was 260,000 persons. Merriam's figures were based on an estimate of the population of the mission strip, from Spanish data, and a gross extrapolation from that to the remainder of the state.

The first attempt at population estimates in detail and with the use of a variety of data was made by Kroeber (1925). The figure he got for the whole state was 133,000 persons, and he still used that figure, although with some reservations, as late as 1939 (see Kroeber, 1939, pp. 178-179).

The problem has recently been reopened by S. F. Cook. In 1943 he published an evaluation of Kroeber's estimates, based on essentially the same data, and the result was to increase the estimate by about 10 per cent. In the last two years Cook has begun a more intensive investigation, the results thus far being new estimates for the San Joaquin Valley (1955) and for the Northern California coast (1956). The upshot of these last papers has been to double Kroeber's estimates in the areas under consideration. The basis of the new estimate suggested by Cook is a more intensive use of historical sources and readier acceptance of the observations found there. He says, "Evidence of misstatement should be looked for and, if found, should be discounted or discredited. Otherwise it should be admitted at face value."

Kroeber has recognized the discrepancy between his estimates and those based on historical statements. He agrees that, if the extrapolations from the latter are accepted, the Merriam figure of 260,000 persons would probably be more accurate. The difficulty there is that "if we accept 260,000, one-quarter of all United States Indians were in California; and this seems unlikely enough. Shall we then assume that Mooney and practically all American anthropologists computed far too low?" (1939, p. 179). Kroeber leaves the question unanswered but Cook's recent work carries the implication that the answer is decidedly affirmative.

The estimate in this paper of the population of the California Athabascans agrees with Cook's results, raising Kroeber's estimates; in fact, it goes even further than Cook in that direction. But the estimates here, with one exception, have been based on village counts by ethnographers rather than on historical data. The fact that the estimates run so high tends to bear out Cook's contention that the Kroeber estimates should be raised.

In basing population estimates on village counts there are several sources of error. Among these are assumptions regarding the number of persons per house and the number of houses per village. I believe that all the assumptions I have made in this regard have been conservative and therefore would not result in overestimates. The number of houses per village can sometimes be calculated rather closely from the number of house pits seen in the sites. That is, the houses can be calculated closely if the assumption is correct that four-fifths of the number of house pits in a site represents the number of simultaneously occupied houses. Admittedly, this figure is rather speculative, but the best opinions I have been able to get grant that it is probably conservative.

A more serious possible source of error concerns the question of which and how many sites were simultaneously occupied. When there is a complete village count, I have excluded from consideration known summer villages, villages not on main salmon streams, and other villages of doubtful status. Even so, the villages run about one per mile along the salmon streams and the possibility presents itself of movement from site to site, perhaps in response to varying fishing conditions. If this was the practice, then the population estimates might have to be reduced by half or even more. But there is no concrete evidence to support such a theory and it is a fact that the Goddard material gives quite complete information of this kind. Therefore, if the present calculation is an overestimate, it is not a very great one.

ESTIMATES BASED ON VILLAGE COUNTS

Wailaki (Eel and North Fork).—The present list gives a total of 67 villages among the Eel River and North Fork Wailaki. For purposes of calculating population I have excluded 13 of them (nos. 6, 9, 16, 31, 38, 40, 51, 57, 58, 59, 61, 66, 67) because they are summer camps in the hills, rock shelters used only briefly, or specialized fish-drying camps. These places do not seem to have been used simultaneously with the main villages. This list appears to be a substantially complete count from Horseshoe Bend south, but it is clear that neither Merriam nor Goddard visited the area north of this, and the village count suffers as a result. There are about 16 river-miles south of Horseshoe Bend, including both the main Eel and North Fork, and there are 49 main villages on this stretch, yielding an average of 3.1 per river-mile. If we apply this figure to the 7 river-miles above Horseshoe Bend, we get 21.7 villages for that stretch rather than 5, as given by ethnographers. We may reduce this figure to 15, because this stretch of the river appears to offer a less desirable location (Goddard, 1923a, p. 107).

This calculation gives a total of 69 villages for the entire group, considerably less than Cook's total of 87 (Cook, 1956, p. 104). The reason for the difference is that Cook bases his estimate on Goddard's data, with the territory of the Wailaki extending above Kekawaka Creek, whereas I have taken Kekawaka Creek as the boundary.

The house count per site for this group must be extrapolated from Goddard's house-pit counts (1923a, pp. 103, 105) on the sites of two of the tribelets. This figure has been calculated by Cook, who takes Goddard's house-pit count for 20 sites as "92 pits." For two localities, however, Goddard 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" (Cook, 1956, p. 104). Cook then reduces the figure by 20 per cent to allow for the probability that not all the house pits represent simultaneously occupied houses. His average number of houses per site is 4, which would not appear to be an overestimate. If we take this figure, we have a total of 276 houses for the Wailaki as against Cook's figure of 348, which was based on a greater area.

Cook takes 6 persons per house as the average density for the Wailaki. This figure is arrived at in several ways. The figure of 7.5 per house is well established for the Yurok and sets an upper limit for the Wailaki area. Goddard appears to have based his population estimate on a mean of 4.5 persons per house, almost certainly too low, and Cook compromised at 6 per house. This figure is supported by independent observation by Foster on the Round Valley Yuki (Cook, 1956, p. 107). The social organization and the habitat of the Yuki and Wailaki are nearly identical, so the population per house should be the same for both groups.

Accepting the figure of 6 persons per house, we get a total population of 1,656 for the Eel Wailaki and the North Fork Wailaki, as compared with Cook's figure of 2,315 and Goddard's figure of between one and two thousand.

Pitch Wailaki.—Goddard (1924) records 33 villages for the Pitch Wailaki. For two of the four tribelets, the count is virtually complete. For a third tribelet, the T'odannaÑkiyahaÑ, Goddard lists 6 villages and indicates that there were probably more (1924, p. 225). If, to allow for these possible villages, we add 5 to the total above, we get a total of 38 villages for three tribelets, or an average of 12.7 per tribelet. Although the fourth tribelet, the TchokotkiyahaÑ, had a poorer habitat than the other three (Goddard, 1924, p. 222), we may assume that it had at least 8 villages, an estimate which is probably conservative in view of its extensive territory. We then get a total of 46 villages for the Pitch Wailaki.

Goddard counted house pits in 22 village sites and got an average of 5 per site. If we reduce this to 4 to account for unoccupied pits, we have an estimate of 184 houses for the Pitch Wailaki, as against 172 estimated by Cook. On the basis of 6 persons per house this gives a population of 1,104 as against 1,032 by Cook and between 650 and 800 by Goddard.

For all Wailaki combined we get a total of 2,760. Cook's figure is 3,350, Kroeber's is 1,000, and Goddard's is between 1,650 and 2,800—average of 2,225. The difference between the figure presented here and Cook's figure is mostly due to the adjustment I have made in the Wailaki boundary from the one used by Goddard.

Mattole.—The village lists of Merriam and Goddard give a total of 42 villages for the Mattole. I have excluded 5 of these from calculation of population estimates, one because it is a summer camp and four others because the frequency appears too great, in places along the coast, to make simultaneous occupation likely. This leaves a total of 37, very likely a conservative estimate since Goddard gives a number of names of villages not located and therefore not included in our calculations.

Cook estimates 6 houses per village for the Mattole on the basis of comparison with the Wiyot, Yurok, Tolowa, and Chilula. Goddard counted house pits for a few sites of the Mattole and they appear to average less than that. Not much reliance can be placed on this average, because the sample was very small. However, the number of houses per site is probably not as high as among the Yurok. I have compromised with a figure of 5.4, the same as the estimate for the Sinkyone, the eastern neighbors of the Mattole.

Cook takes Kroeber's Yurok figure of 7.5 persons per house in calculating Mattole population. The social organization here is more nearly like that of the southern Athabascans, so I have used 6 per house. This figure gives a total population of 1,200 as against 840 figured by Cook for the Mattole exclusive of Bear River. The difference here is due to the fact that Goddard's village lists were not available to Cook. If they had been, he would have obtained a figure of 1,665, or nearly double his actual estimate.

Lolangkok Sinkyone.—For the Sinkyone on the northern part of the South Fork of the Eel we have a nearly complete village count. South of Larabee Creek Goddard and Merriam give a total of 46 villages. North of Larabee Creek on the main Eel the village count is incomplete, but Merriam gives 8 place names. That these place names represent village names is clear from the Merriam place names farther south which can be checked against Goddard's data. Together, these give a total of 54 villages but leave out the areas of Bull Creek and the upper Mattole River. We may assume 5 villages in each of these, surely a conservative estimate in view of the density of sites on Salmon Creek and South Fork. We thus have an estimate of 64 villages for the Northern Sinkyone.

Goddard counted house pits in 24 of the sites he recorded. They come to a total of 162 or 6.7 per village. If we reduce this by 20 per cent to account for unoccupied pits, we get an average of 5.4 houses per site or a total estimate of 346 houses among the Lolangkok Sinkyone. At 6 persons per house this estimate yields a total population of 2,076.

Hupa.—In the present village list there are 11 villages in Hoopa Valley and 16 above the valley on the main Trinity and on South Fork. Of these sixteen, three have been rejected as being in Chimariko territory (nos. 25, 26, 27). Cook has argued, reasonably, it appears, that the villages in Hoopa Valley average 11 houses, whereas the villages above the valley average 4.5 houses each. This average gives a total of 193 houses for the Hupa.

Cook has estimated that there is an average of 10 persons per house among the Hupa. This figure is arrived at by the following line of reasoning: according to a census taken in 1870 there was a total of 601 persons in 7 villages at that time, of which 232 were male and 359 were female. This count indicates a disproportionate number of males and Cook therefore calculates a population of twice the number of females, or 718, as a more normal population. Goddard's data give the number of houses for these villages as 92, a figure Cook takes as representing the situation in 1850. This combination yields an average of 7.8 persons per house. Since there had certainly been a decline in population between 1850 and 1870, Cook proposes that the figure for the density of population be raised to 10 persons per house.

But Goddard does not say what period his figures represent, so I propose to follow a line of reasoning similar to that of Cook but to use different figures. The number of houses for 6 villages in 1851 is reported by Gibbs (see map, pl. 9). We may compare these to the 1870 population estimates as given by Kroeber (1925a, p. 131). If we adjust for male attrition by calculating population as twice the female population, or 640 (see table 1), we get a density per house of 7.8, exactly the same figure that Cook gets.

TABLE 1

Hupa Population, 1870[1]

That there was a decline in population between 1850 and 1870 is agreed by all authorities. This fact makes it very attractive to accept Cook's proposed density of 10 persons per house for the Hupa in aboriginal times. But there are two objections to this procedure. For one thing, the population figures for 1870 may be inaccurate. In the census of that year, there were reported 874 Indians of all tribes on the Hoopa Reservation (Kroeber, 1925a, p. 131). But in the same year another agent reported only 649 Indians on the reservation. This is a 25 per cent reduction, and if we reduce the population estimate of 640 by 25 per cent, we get 480 as the estimate for 1870 and a density per house of 5.9. If we raise the population of 480 to account for the 1850-1870 reduction, we are again close to the figure 7.5 persons per house. This calculation is presented merely to indicate that the figures are not reliable.

The other objection to accepting Cook's proposed figure for density is that the established figure for the Yurok is 7.5 persons per house. According to Cook, this figure was based on an underlying assumption that "the social family in the usual monogamous tribe included the father, mother, children, and occasional close relatives" (Cook, 1956, p. 99). As a matter of fact, Kroeber's estimate is not based on this assumption but is an empirical estimate based on population counts and house counts (Kroeber, 1925a, pp. 16-19), and the figure is accepted wholeheartedly by Cook for the Yurok (1956, p. 83). But what is certainly clear is that the social organization, house type, and environment of the Hupa was virtually the same as that of the Yurok and therefore the population density per house must have been the same. It is therefore clear that we must accept either 7.5 persons per house or 10 persons per house as the population density for both the Hupa and the Yurok, and the question becomes one of comparing the reliability of the figures given for the Yurok with those given for the Hupa. Yurok figures appear to be intrinsically more reliable and are also earlier and I have therefore taken 7.5 persons per house as the density.

The population for the Hupa then comes to 1,475 as compared to 2,000 estimated by Cook and to less than 1,000 estimated by Kroeber.

Whilkut.—The number of permanent villages among the Whilkut has been estimated here at 69. This estimate excludes known summer camps and other villages away from the main salmon streams. For the Chilula Whilkut there are 23 villages. For the Kloki Whilkut there are 16 villages, including several which are not shown on the map but which are listed by Merriam as being on upper Redwood Creek. Ten villages have been taken from the North Fork Whilkut. Twenty villages are taken from the Mad River Whilkut even though only 16 are given in the village lists. Wherever both Merriam and Goddard worked the same area the latter has recorded substantially more villages than the former. I have therefore added 4 to the village count to make up for the presumptive lack, thus bringing the total up to 69.

House-pit counts from the Chilula Whilkut are listed for six villages by Kroeber (1925a, p. 138) as 17, 7, 4, 2, 4, 8, or an average of 7 per village. Kroeber reduces this average by a third, on the basis of his estimates for the Yurok and Hupa, to arrive at a figure of 5 houses per village. Cook (1956, p. 84) says the reduction should be only about 10 per cent, calculated on the basis of Waterman's study of the Yurok (Waterman, 1920), and he compromises, making a reduction of a seventh to use 6 as an average number of houses per village.

The sample used by Kroeber and Cook is so small that an estimate based on it of the average number of house pits per village is liable to considerable error. If we look at the figures for some of the surrounding groups, we find an estimate of 11 houses per village for the Hupa in Hoopa Valley, 4.5 for the Hupa outside the valley, 4 for the Wailaki, 4.5 for the Wiyot (Cook, 1956, p. 102), and 5.4 for the Lolangkok Sinkyone. The Whilkut terrain and culture is certainly more nearly like the region outside Hoopa Valley than inside it, so we are scarcely justified in estimating more than 5 houses per village.

On this basis we get a total of 345 houses for the Whilkut. Both Kroeber and Cook use the Yurok figure of 7.5 persons per house in calculating the population of this group. This figure may well be too high, and perhaps it should be more nearly the same as the estimate for the southern groups, but since I have no concrete evidence to support such a contention, I have also used the Kroeber and Cook figure. This gives a total population of 2,588 for the Whilkut.

Cook's figures for the groups which were formerly listed under the Chilula and Whilkut were 800 and 1,300 making a total of 2,100. Kroeber's figures were 600 and 400 for a total of 1,000. The difference between Cook's figures and those given here is partly due to the fact that Cook took the group on the North Fork of the Mad to be Wiyot, whereas I have them as Whilkut. Also Cook made a reduction of a ninth in his Mad River estimates because of the poor environment there. I have not done this because the Mad River region does not seem to me noticeably poorer than that along Redwood Creek.

ESTIMATES BASED ON FISH RESOURCES

For the six tribes just discussed, the ethnographic notes at our disposal offer a means of estimating the population, but we have also another basis for our calculations. Fishery was the most important single factor in the California Athabascan economy, hence the fish resources of the region undoubtedly exerted a marked influence on population size. Therefore, before attempting to estimate the population of the remaining groups, for which we have scanty ethnographic information, I would like to present some data on the fish resources of the region.

I have attempted to calculate the number of stream miles of fishing available and thereby to form some estimate of the economic basis of each of the groups. Most of my information comes from Mr. Almo J. Cordone, Junior Aquatic Biologist of the California Department of Fish and Game, who was kind enough to gather the relevant data from the records of that organization. I have not included material on the freshwater trout, which was apparently too scarce to be important, or on the lamprey eel, on which we do not have sufficient information, although it was of some importance, especially in the Eel River and its tributaries.

The available stream miles of fishing may seem insufficient material on which to base estimates of fish resources and unquestionably it would be desirable to have some idea of the fish population per mile of stream in order to estimate the food value of the resources available to the people. On the other hand, this point may not be as crucial as it seems, for apparently the fish population was not a governing factor in the number of fish taken by the Indians. According to Rostlund (1952, p. 17), the aboriginal fishermen of California did not even approach overfishing. If this is so, then there must have been fish left uncaught even in the smaller salmon streams and it would therefore seem that one stream was nearly as good as another, if it carried salmon at all. An exception would be the Trinity River and its tributaries, the only streams in the Athabascan area with both spring and fall runs of salmon. In other streams there is only a fall run.

The lists that follow include data, not only for the six tribes previously discussed (Wailaki, Pitch Wailaki, Mattole, Lolangkok Sinkyone, Hupa, and Whilkut), but also for the Nongatl, Kato, Shelter Cove Sinkyone, Lassik, and Bear River groups. The fish species is recorded, when it is known; when our source gives no identification of species, however, the generic term is used.

Available Stream Miles for Fishing in Tribal Territory

KATO 29 mi.

South Fork Eel R.—19 mi. Quantities of steelhead and silver salmon go up at least to Branscomb and King salmon go at least to Ten Mile Cr. (Dept. of Fish and Game).

Hollow Tree Cr.—5 mi. There was fishing on this stream (Gifford, 1939, p. 304). Fish not specified, probably steelhead and salmon.

Ten Mile Cr.—5 mi. This stream appears to be large enough for salmon and there were villages on it. Also the Fish and Game information for South Fork implies fish in the stream.

WAILAKI (Eel R. and North Fork Wailaki) 23 mi.

Eel R.—16 mi. There are good runs of salmon as far up as Lake Pillsbury (Dept. of Fish and Game).

North Fork Eel—7 mi. Salmon go up North Fork farther than 7 mi. (see Pitch Wailaki).

PITCH WAILAKI 15 mi.

North Fork Eel—12 mi. See below.

Casoose and Hulls creeks—3 mi. The Dept of Fish and Game states that salmon do not ascend North Fork above Asbill Cr. but Goddard's informant (see Pitch Wailaki Village no. 21) said that fish got up into Hulls and Casoose creeks, the mouths of which are above Asbill Cr. The Dept. of Fish and Game information may refer to a more recent situation.

LASSIK 25 mi.

Eel R.—17 mi. (See Wailaki.)

Dobbyn Cr.—8 mi. There would seem to have been fish in Dobbyn Cr., since it is a fair-sized stream and there were many villages on it.

SHELTER COVE SINKYONE 67 mi.

South Fork Eel—39 mi. There were a good many fish in South Fork as far up as Branscomb (Dept. of Fish and Game).

Redwood Cr.—5 mi. According to Merriam the region around Redwood Cr. was a center for the Shelter Cove Sinkyone; therefore there must have been fish in the creek.

Mattole R.—11 mi. There is a partial barrier to salmon at the community of Thorn but some fish get up even beyond this (Dept. of Fish and Game).

East Branch, South Fork Eel—4 mi. King salmon and silver salmon go up at least to Squaw Cr. (3 mi.) and steelhead go up at least to Rancheria Cr. (4.5 mi., according to the Dept. of Fish and Game).

Sea Coast—8 mi. The Shelter Cove Sinkyone have 16 mi. of sea coast. The only reliable data on the density of sea coast population in relation to the riverine population are given by Kroeber (1925a, p. 116). According to his figures, the seashore is about half as productive as the rivers and I have therefore halved the sea coast mileage in the calculation of available fishing miles.

LOLANGKOK SINKYONE 63 mi.

Eel R.—27 mi. (See Wailaki.)

South Fork Eel R.—16 mi. (See Kato.)

Bull Cr.—6 mi. According to Merriam, there was a large settlement on Bull Cr. It could not have been supported without fish.

Salmon Cr.—5 mi. Goddard mentions fishing on at least part of this stream.

Mattole R.—10 mi. The fish go beyond this stretch at least as far as Thorn (Dept. of Fish and Game).

MATTOLE 38.5 mi.

Mattole R.—25 mi. The fish go considerably beyond here in the Mattole.

North Fork Mattole—5 mi. North Fork is a sizable stream and there were several villages along it, so it probably had fish in it.

Sea Coast—8.5 mi. The Mattole have 17 mi. of sea coast. This has been halved in accordance with the principle stated above.

BEAR RIVER 21 mi.

Bear R.—18 mi. This figure is rather arbitrary since the information is poor for this stream. It is known that silver salmon and steelhead are caught there and that there is a fall run of King salmon (Dept. of Fish and Game).

Sea Coast—3 mi. The Bear River group has 6 mi. of sea coast, halved for present purposes.

NONGATL 85 mi.

Van Duzen R.—40 mi. Steelhead go up as far as Eaton Roughs (40 mi.). Silver salmon go up as far as Grizzly Cr. (21 mi.) and probably as far as Eaton Roughs. There are no data on King salmon but it is known that there is a fall run of them here. Information from Dept. of Fish and Game.

Eel R.—5 mi. All 5 mi. of the Eel in Nongatl territory should provide excellent fishing.

Larabee Cr.—20 mi. There is no direct information on this stream, but it is of considerable size and there were many villages at least 20 mi. up.

Yager Cr.—20 mi. Again we have no direct information but there are many villages far up on this stream. Twenty miles of available fishing is probably a conservative estimate.

Mad R.—0 mi. There is a long stretch of Mad R. in Nongatl territory but, according to the Dept. of Fish and Game, no fish go up so far.

WHILKUT 70 mi.

Mad R.—27 mi. There is a 12-ft. falls at Bug Cr. which represents a nearly complete barrier to salmon. This means that there are salmon in nearly all the territory of the Mad R. Whilkut.

North Fork Mad R.—8 mi. According to Merriam, there were fishing camps nearly this far up on North Fork.

Redwood Cr.—35 mi. There is no direct information on this stream. I have attributed salmon to nearly its whole length because of the size of the stream and the large number of villages along its upper course.

HUPA 39 mi.

Trinity R.—27 mi. There are fish in this whole stretch (Dept. of Fish and Game).

South Fork Trinity—12 mi. There are known to be salmon in South Fork, and presumably they go up as far as the border of Hupa territory.

TABLE 2

Area, Fishing Miles, and Population Estimates

TABLE 3

Area and Fishing Miles

GROSS ESTIMATE

From the preceding data we have obtained population estimates for certain of the California Athabascan groups. If these estimates are judged reliable, it would be desirable to use them as a basis for estimating the population of the remaining groups. When a detailed analysis of the ecological or demographical factors involved is lacking, it is sometimes necessary to fall back on rather simplistic assumptions to attain the desired end. Cook goes rather far in this direction, using simply the average population density per square mile of the known groups to estimate the population of the unknown groups.

It appears to this writer that a somewhat more satisfactory method of estimation would be based on simple linear regression theory. It is a fact that pertinent relationships in population studies can often be expressed in terms of simple exponential functions or in linear combinations of logarithms. Thus we might propose a relationship such as the following:

population = a + b (ln area)

or

population = a + b (ln fishing miles)

where a and b are constants to be determined and ln is the logarithm to the base e.

Of course we would not expect these relationships to be precise. The lack of exactness might be due to the crudeness of the various measurements involved or perhaps to the fact that population depends on more than one such factor. To account in some way for the uncertainty, we might make a further assumption and propose the following relationships:

population = a + b (ln area) + X
population = a + b (ln fishing miles) + X

where X has a normal probability distribution with mean = 0 and some unknown variance = s². X is then, roughly speaking, the error involved in each observation. That the error would be distributed normally is quite reasonable under the circumstances. In situations where the uncertainty of the observation is due to measurement error or to a multiplicity of factors, the distribution obtained often assumes a normal form or a form sufficiently normal so that the normal distribution can be used as an approximation.

One additional assumption is necessary. We must assume that the sample used is taken in a random fashion from the population to be studied. In the present investigation, the sample is definitely not taken at random, since we are using all groups for which we have population estimates based on ethnographic information. The question is, then, whether this selection of groups would result in some bias. For instance, the groups for which we have ethnographic data might be the most numerous in the first place and might thus cause us overestimate the population of the remaining groups. On the whole, it would seem to me that there is no such bias and that the assumption of a random sample is therefore not misleading, at least in the direction of overestimation. If we now consider each group for which we have no ethnographic data, we can see whether the lack of such data is due to an initially small population or to mere luck.

Kato: The reason Kato population is being estimated in gross rather than from ethnographic data is that Goddard (1909, p. 67) obtained a list of more than 50 villages which are not available for calculation.

Bear River: Here the lack of information is due simply to the fact that it was not collected. There have been several informants living until recently (see Nomland, 1938).

Lassik: There was at least one good informant living until recently (Essene, 1942), but Merriam worked with her only briefly. Goddard evidently recorded a number of villages from this group, but his notes are lost.

Nongatl: Goddard seems to have worked with at least two informants from this group, but he spent a very brief time in the area and some of his notes may have been lost.

Shelter Cove Sinkyone: Several informants from this group have been alive until recently (see Nomland, 1935). No one saw fit to collect the appropriate data.

It is obvious from this summary that the main reason for our lack of information on these groups is the loss of Goddard's notes. If those were at hand, we would probably have complete information on the Kato, the Lassik, and probably the Nongatl. The absence of data on the Bear River and Shelter Cove Sinkyone is due to the ethnographers' oversight. None of these groups, therefore, seem to have been selected because of their small aboriginal population. If the following estimates are in error because the sample is not a random one, then the error is probably one of underestimate rather than overestimate.

Given the foregoing assumptions, the least squares estimate of the normal regression line may be obtained with the following formula.

P: population. A: area. F: fishing miles.

The equations of the lines are:

P = a + b (ln A)
P = a' + b' (ln F)

the estimate of b is (Bennett and Franklin, 1954, p. 224)

and of a is

 = Y - b^X

where X_i = ln A for each group with known population and Y_i = P for each known group.

Similarly the estimate of b' is

and of a' is

Â' = Y - b^'X

where X_i = ln F for each known group and Y_i = P for each known group. These calculations are shown in table 4.

TABLE 4

Calculation of Regression Lines Shown in Figure 2

The results are the following equations, which are shown, together with the points from which they were calculated, on figure 2.

P = 1.02 (ln A) - 4.06
P = .75 (ln F) - 1.00

Thus, given either the area of a group or the fishing miles of a group habitat, we may estimate its population. From the diagram in figure 2 it appears that the estimates based on area have greater dispersion than those based on fishing miles and are therefore less reliable. This fact can best be made precise by using the above assumptions to obtain the confidence intervals for each of the estimates. The confidence intervals for the area estimates are given by the following formula (Bennett and Franklin, 1954, p. 229).

where the symbols have the following values and meanings:

[10.6] X_o: the log of the area of the group for which the population is being estimated.

X_i: the log of the area of each of the groups for which the population is already known.

X: the average of the X_i.

t_?: the upper ?-point of the t-distribution (Bennett and Franklin, 1954, p. 696) where 1-? is the confidence coefficient.

{1 }
S_a = v{- × S(Y_i + 4.06 - 1.02X_i)²}
{4 }

where Y_i is the population of each of the groups for which population is known. This is the estimated standard deviation of population where the estimate is made from area.

The confidence intervals for the fishing-mile estimates may be obtained in similar fashion—simply substituting the words fishing mile for area and S_f for S_a.

For calculating the confidence intervals for area we have the following quantities:

X = 5.56
t_.2 = 1.533
S(X_i - X)² = .859
S_a = .3594

The calculations are shown in table 5.

The comparable quantities in calculating the confidence intervals for fishing-mile estimates are:

X = 3.70
t_.2 = 1.533
S(X_i - X)² = .932
S_f = .394

The calculations are shown in table 6.

TABLE 5

Calculation of Confidence Intervals for Area

TABLE 6

Calculation of Fishing-Mile Estimates

The results of the calculations are given in table 7. The figures are point estimates with 80 per cent confidence intervals. This means that under the assumptions given earlier we expect that the tabled intervals will contain the true population 8 times out of 10. I have accepted the estimates derived from fishing miles because their confidence intervals are a bit shorter on the average.

TABLE 7

Population Estimates and Confidence Intervals

The question of whether the fishing-mile estimates yield shorter confidence intervals than the area estimates brings up an entire range of problems pertaining to economy, settlement pattern, and the like. The obvious interpretation of the shorter confidence intervals would be that the economy of the people in question depended more on fish and fishing than on the general produce over the whole range of their territory. The question then becomes one of quantitative expression—we would like to have some index of the extent of dependence on various factors in the economy. This might best be approached from the standpoint of analysis of covariance, where we would obtain the "components of variance." This technique is a combination of the methods of regression used in this paper and those of the analysis of variance. It would evidently yield sound indices of economic components, but it involves, for myself at least, certain problems of calculation and interpretation which will have to be resolved in the future.

Another problem of this kind turns on the question of which factors are important in which area. Considering the State of California, for instance, we might want to know about such factors as deer population, water supply, the quantity of oak trees, etc. Any one of these factors or any combination of them might be important in a particular area; the problem of gathering the pertinent information then becomes crucial. Moreover, because the situation has changed since aboriginal times, we must combine modern information with available historic sources. S. F. Cook has shown that energetic and imaginative use of these sources yields very good results (e.g., Cook, 1955).

Finally, there is the problem of the assumptions we were required to make in order to obtain our population estimates. Although many of the assumptions in the present paper are difficult to assess, the two which I would like to discuss here were particularly unyielding—the assumptions of the number of persons per house and the assumptions of the number of houses per village.

The question of how many persons there were per house has been dealt with extensively by both Kroeber and Cook. There is also a great deal of random information in the ethnographic and historical literature. I believe there are enough data now at hand to provide realistic limits within which we could work, at least for the State of California. This information should be assembled and put into concise and systematic form so that it would be available for use in each area. It would also be of interest in itself from the standpoint of social anthropology.

For the number of houses per village we have also a considerable body of information, but here we are faced with a slightly different problem. It often happens that we know, from ethnographic information or from archaeological reconnaissance, how many house pits there are in a village site but do not know how many of the houses which these pits represent were occupied simultaneously. In the present paper it has been assumed that four-fifths of the house pits represents the number of houses in the village occupied at any one time. This, however, is simply a guess, and one has no way of knowing how accurate a guess. The solution to this problem is simple but laborious. From each area of the State a random sample of villages with recorded house counts should be taken. Each of these village sites should then be visited and the house pits counted. A comparison of the two sets of figures would give us a perfectly adequate estimate, which could then be used subsequently over the entire area.

TABLE 8

Population Estimates

The corpus of information provided by the methods outlined above would be useful in two ways. First, it would clarify our definitions of the economic factors in the lives of hunter-gatherers. Functional hypotheses which postulate dependence of social factors on economy would be subject to objective, quantitative tests of their validity.

Second, the corpus of information would afford a suitable basis for inference from archaeological data. If we can determine what were the major economic factors in the lives of a prehistoric people, then we can make assertions about population, settlement pattern, and the like. Conversely, information about population and settlement pattern would imply certain facts about the economy. This technique has already been developed to some extent. For instance, Cook and Heizer, depending on assumptions derived from ethnographic data (Cook and Treganza, 1950; Heizer, 1953; Heizer and Baumhoff, 1956), have made inferences concerning village populations. These methods have such great possibilities for the conjunctive approach in archaeology that their use should be extended as much as possible.