the CalCOFI survey area in 1951-56 was given by Ahlstrom (1956). The temperature range over which anchovy eggs were taken in 1953 and 1954 was 9.9° to 23.3°C; most were taken within a 7°C. range, 12.0-18.9°C. (Ahlstrom, 1956).

The southern extent of anchovy spawning and its offshore extent in the area between San Francisco, Calif., and Magdalena Bay, Baja California, have been adequately circumscribed by CalCOFI survey cruises. The northern extent of anchovy spawning has not

been adequately delimited, however. For example, we do not know how consistently anchovy spawning occurs off Oregon and Washington, or how large the adult stocks are in these waters.

Information from our surveys permits the assessment of the relative abundance of anchovy larvae off California and Baja California (table 4). The majority of anchovy larvae have occurred off Baja California in most years; the exceptions are 1957, 1959, and 1964 (table 5 and fig. 5).

Anchovy larvae were taken to the north of Point Conception, California, in all years; numbers ranged from insignificant (1953, 1955) to consequential (1958-64, especially 1959 and 1964). To evaluate better the importance of anchovy larvae in these waters, data on anchovy larvae have been summarized by 120-mile units (table 6). Each geographic unit contains one cardinal and two ordinal station lines, as 40, 43, 47. The tabulation by year and geographic unit summarizes three kinds of information: 1) number of stations occupied during the year, 2) number of hauls containing anchovy larvae, and 3) total number of larvae obtained (standardized haul totals).

Few anchovy larvae were taken off northern California either on lines 40-47 or lines 50-57. Less than 5 percent of the hauls from these two geographic units contained anchovy larvae, and the average number of larvae per positive haul was slightly less than 10; the average number per occupancy was less than 0.5.

The central California area as a whole was more important for anchovy eggs and larvae than was northern California. The portion of the central California area lying between Point Sur and Point San Luis (station lines 70-77) was consistently more important than the one between San Francisco and Monterey (station lines 60-67). Lines 60-67, San Francisco to Monterey, were occupied almost as

frequently (934 collections in 14 years) as the section between Point Sur and Point San Luis (1,007 collections), but the positive hauls were fewer (167 compared to 335), and the average number of larvae per haul was markedly lower. In both areas the abundance increased sharply during 1958-64. In the San FranciscoMonterey area (lines 60-67, the average number of anchovy larvae per occupancy was only 0.8 during 1951-57, but was 10.6 during the following 7 years. The average number of larvae per positive haul increased from 8.1 to 41.4. The area between Point Sur and Point San Luis had almost 10 times as many larvae in both time periods: 7.1 per occupancy during 1951-57 and 100.5 per occupancy during the following 7 years. The average number of anchovy larvae per positive haul increased from 63.6 to 187.0. Obviously, this latter area is the most important one for anchovy larvae off central or northern California.

The distribution and abundance of anchovy larvae to the north of Point Conception have been presented in some detail to bring out the increasing importance of the region immediately to the north and the comparative scarcity of anchovy larvae off northern California. Markedly more anchovy larvae were obtained off southern California, however, than off central or northern California in all years (table 4 and fig. 5).

SIZE COMPOSITION

The sizes of sardine larvae taken on monthly survey cruises, 1951-59, are summarized in tables 10-18, and of anchovy larvae in tables 38-46. Numbers of larvae of both species decreased markedly with increase in size. These tables are included to make size data available and to demonstrate that about the same proportions of larvae of different sizes are taken each year. To emphasize further the latter point, I have grouped the anchovy

larvae into somewhat coarser size groupings, usually 3 mm., in table 7. This two-part table summarizes numbers of larvae by size group in the upper part and by percentages in the lower. In only 1 year, 1956, were percentages of small larvae conspicuously low and percentages of other size groupings of larvae correspondingly high. The consistency of representation of different sizes of larvae year after year is evidence that total numbers per se are meaningful values for comparing abundance of larvae.

INCREASE IN ABUNDANCE OF ANCHOVY LARVAE SINCE 1951

On the basis of increase in abundance of larvae, the anchovy population has increased markedly since 1951. To make the quarterly survey coverages obtained since 1961 equivalent with the more frequent coverages (mostly monthly) obtained during 1951-60, the following procedure has been adopted. For the decade 1951-60, average quarterly values derived for each year were based on data from the three or fewer comprehensive survey cruises made during the quarter; each yearly total represents the summation of "average" quarterly values (table 8). The data given for 1961-64 represent the quarterly survey totals, summed to obtain the yearly estimates.

The population increased rapidly between 1951 and 1954, reached a plateau that continued through 1957, increased rapidly again through 1962, and since then appears to have maintained a high level. A conservative method of estimating the increase in population size is to use the average of the annual estimates for 3 years at the beginning of the period of increase (1951-53), and of the most recent years (1962-64). The average number of anchovy larvae was 21,875 in 1951-53. In comparable coverage during 1962-64, the average number was 195,000. Thus, approximately nine times as many anchovy larvae were taken per year during the recent period as during the early 1950's. The increase in abundance is illustrated in figure 6.