Reproduction in Anthopleura — Ford 
143 
rate from one another and cease to wander 
further within a few days. This would tend to 
suggest that individuals are aggregated into 
large masses only as a result of growth and 
asexual reproduction within the limitations of 
space. 
A third explanation might suggest some fac- 
tor influencing larval settling, such as some 
substance released by larvae which first settled, 
which attracted larvae of the same sex and 
repelled those of the opposite sex. However, 
considering the relatively great distances in- 
volved, and the large volumes of moving sea 
water through which these substances would 
have to pass, this seems to be the most improb- 
able explanation. 
ANNUAL GONAD CYCLE OF A. elegantissima 
The data for the gonad index of the aggregat- 
ing sea anemone for 1959 and I960 are given in 
Figure 1. It is evident that only a single breeding 
cycle occurred during each year, with gonads 
small during the winter months, and increasing 
during spring and summer to a peak before 
spawning out completely in September. Overlap 
of 95% confidence bands about the sample 
means indicates a generally good correlation 
between the cycles of the 2 years, except that 
the maximum gonad index may have been 
reached 2 months earlier in I960 than in 1959. 
The maximum ovum diameters for each sam- 
ple for the two years are presented in Figure 2, 
which shows an almost continuous increase in 
ovarian egg size, from very small during the 
winter months to a maximum size just before 
spawning in September. 
In both 1959 and I960, tailed sperm were 
observed in July in smears of the testicular tissue 
of one or two males, and by mid-August all 
testes were filled with tailed sperm. 
There seems to be some slight correlation 
between size and sexual reproduction, as shown 
in Table 1 and observed throughout the year, 
in that a greater proportion of the smaller in- 
dividuals (weighing 0.3 to 3.0 g) are found to 
be without gonads than are larger individuals 
(3 to 10 g) . However even the smallest animals 
may be fertile and the largest animals may be 
without gonads. 
If any correlation may be safely drawn be- 
tween the events of the gonad cycle and ocean 
temperatures, as shown in Figure 1, it is that 
gonad size and ovarian egg size increase with 
the rising temperatures of the summer months, 
and spawning occurs shortly after temperatures 
have reached a peak in early fall and begun to 
decrease. It must be noted, however, that the 
temperature data were taken at the Golden Gate, 
several miles south of the collecting area. Most 
importantly, due to their position in the inter- 
tidal zone, these animals spend nearly half the 
time out of water, exposed to much greater 
temperature extremes and fluctuations than are 
indicated in Figure 1. 
The data for male and female reproductive 
cycles for 1959 and I960 are given in Figure 3 . 
However, no attempt is made here to compare 
male and female cycles on the basis of these 
data, because of the circumstances of sampling 
as explained above under the topic of the 
distribution of the sexes. 
DISCUSSION 
The present account reports a single gonad 
cycle per year for the aggregating sea anemone, 
Anthopleura elegantissima, with the gonad index: 
and ovarian egg size increasing gradually from a 
low in the fall after spawning, to a high the 
following September, when spawning again oc- 
curs. Spawning was not observed by the author, 
nor are there any published accounts of spawn- 
ing in this species, but reports of such observa- 
tions have been received by personal communi- 
cation. These reports place the time of spawn- 
ing near the end of summer, which agrees with 
the present findings, and state that the majority 
of individuals in the population were shedding 
Fig. 1. Gonad cycle of Anthopleura elegantissima in 1959 and I960, and corresponding ocean tempera- 
tures. Temperature data were obtained from the U. S. Dept, of Commerce, Coast and Geodetic Survey, 
Bureau of Marine Data. Each sample is plotted in a frequency histogram, with heavy lines joining the means 
and 95% confidence limits computed for a "t” distribution for small samples (Wilks, 1948). Below, the 
95% confidence bands for the two years are compared. 
