ROSKNHU'M and NIKSKN: SPAWNINC.CVC'LK OF A/y'.4 ARKNAHIA 



during May and June. By the end of June, the sample 

 was roughly divided into thirds among the partially 

 spawned, spent, and inactive stages. Some spawning 

 could still be seen through the end of August, but by 

 late September 85% were inactive, 10% partially 

 spawned, and 5% spent. 



Foster City 



September-October 1979 marked the end of the 

 Foster City population spawning, with most clams 

 (70%) being inactive during this period (Fig. 3). 

 From early November through the end of December 

 1979, no activity could be found. The late January 

 1980 sample showed a 90% active:10% spent ratio 

 with no intermediate stages represented. It is 

 unclear whether this indicated a rapid maturing and 

 spawning of a few precocious individuals or an over- 

 wintering of residual gametes. By mid-March, every 

 stage was represented, most (50%) being in the ripe 

 stage. Peak spawning extended from the beginning 

 of April through May, with a complete maturation- 

 spawning season extending from late February 

 through May. From July to late September no appre- 

 ciable gonadal activity could be discerned. This lack 

 of activity from mid- to late summer distinguished 

 the Foster City population by its short spawning 

 season relative to other populations (Fig. 3). 



Burlingame Lagoon 



Sampling of the Burlingame Lagoon population 

 did not begin until November 1979. At the time 85% 

 of the clams sampled were spent and only 15% were 

 inactive (Fig. 3). This suggests a protracted spawn- 

 ing in 1979, similar to that which was seen in 1980. 

 By December all but 5% were inactive and by 1 

 January 1980 all were inactive. Gonadal activity ap- 

 peared again by the end of January 1980. Peak 

 spawning occurred during May and June; however, 

 spawning continued well into September and Octo- 

 ber with a full 30% of the clams of each sample being 

 in the partially spawned stage. The Burlingame 

 Lagoon population had the longest spawning season 

 of the populations examined. 



Point Isabel 



Sampling of the Point Isabel population began in 

 September 1979. The September and October sam- 

 ples still contained ripe individuals (< 10%), but most 

 of the clams sampled were spent or inactive. By 

 November, 70% of the clams sampled were inactive 

 and by the end of December all were inactive (Fig. 



3). The 1979 spawning season had ended for the 

 Point Isabel population by early November. 



Gonadal activity had resumed by late January. By 

 March, 80% of the clams sampled were active. Only 

 ripe and partially spawned individuals could be found 

 in mid-April, and by May inactive individuals were 

 being found (10%). Peak spawning occurred during 

 May. Spawning continued through June and by July 

 almost 80% were inactive and 20% partially spawn- 

 ed or spent. Spawning appeared to be over; however, 

 the August sample contained almost 35% partially 

 spawned and 5% ripe individuals. It is tempting to 

 suggest a possible second spawning in August, but 

 the May and June samples lacked active or ripe in- 

 dividuals, which suggest the August observation be 

 attributed to sampling error. Spawning was still 

 occurring in late September, as 15% were still in the 

 partially spawned stage. At this time, however, 60% 

 were inactive and 25% spent. 



The four study sites showed similar trends (Fig. 4) 

 in temperature and salinity. Temperatures fell from 

 November through mid-Januar^' and rose from mid- 

 January through the beginning of April and then 

 stabilized. Temperature then climbed again from 

 mid-May through July. The July-September temper- 

 atures were steadier at the Foster City and Bur- 

 lingame sites than at Point Isabel or Candlestick 

 Point, where they dropped markedly during this 

 period. Salinity followed a similar trend, falling dur- 

 ing the late winter months, and rising during spring 

 and summer (Fig. 4). Salinity ranged from 8 to 33 

 ppt and temperature from 12° to 23°C. 



DISCUSSION 



The spawning cycle of M. arenaria in San Fran- 

 cisco Bay in 1980 was an extended one. Gameto- 

 genesis had begun by late January for three of the 

 four populations sampled (Candlestick Point, Bur- 

 lingame Lagoon, and Point Isabel), and by mid- 

 March all five stages of gonadal development were 

 represented in the Burlingame Lagoon, Point Isabel, 

 and Foster City populations. Spawning had begun at 

 all four sites by April; over 20% of the individuals 

 from each sample were in the partially spawned or 

 spent stages. The number of clams in these spawning 

 stages reached a maximum during May and June 

 1980. Spawning continued through September and 

 October and then ceased. 



Only a single clam < 25 mm in length was found to 

 have active gonads. This suggests that 25 mm might 

 be generally recognized as the size at first repro- 

 duction for San Francisco Bay M. arenaria. 



409 



