REPRODUCTIVE CYCLES 631 



is very variable, so that some individuals have smaller gonads during 

 months of peak gonad index than some individuals during the months 

 of low gonad index. This raises the question whether the same indi- 

 vidual spawns several times during the year or whether only a single 

 gonad cycle is found for each individual. The data do not permit an 

 answer to this question. It has not been possible to induce spawning in 

 the laboratory, although individuals have been kept in the aquaria dur- 

 ing the spawning months of the year. It would also be difficult to 

 determine to what extent the animal had spawned, although by de- 

 termining the mass of the eggs this might be feasible. It is possible that 

 spawning is related to lunar rhythms and is missed because of failure to 

 make observations during the night (Korringa, 1947). 



From the published data (Bennett and Giese, 1955; Lasker and 

 Giese, 1954) it is quite clear that in two different years the breeding 

 time is quite different, gonads reaching a peak size in November (with 

 a questionable peak in February) in 1952-53, and a plateau for the 

 months of December to May in 1953-54. Unpublished data for the 

 succeeding years further emphasize the uniqueness of each year: 

 February for 1954-55, November-December for 1955-56, and Feb- 

 ruary-April in 1956-57. 



The data for 1953-54 were taken from two points, one exposed to 

 the open ocean (Yankee Point near Carmel). the other from a pro- 

 tected cove (Pescadero Point). Yet the two cycles parallel each other 

 to a remarkable degree and indicate that local differences have very 

 little to do with the onset of the growth in size and of the spawning of 

 the animals; rather, some more general factor is operating. The data 

 from the more distant stations now being tested are awaited with in- 

 terest. 



Pisaster ochraceus 



Data on the ochre starfish are available for over four years, almost 

 two from the work of Feder (1956) and two for succeeding years 

 (Farmanfarmaian et al., 1959, and work in progress). In all cases a 

 single marked cycle is observable and, as seen in the data of Table I, 

 the peak is reached in the spring at almost the same time each year. 

 Extensive spawning was observed in May in 1953-54 and 1954-55 

 by Feder, both in the laboratory and in the field, although the gonad 



