While annual fluctuations in catch per unit of effort denote 

 changes in the size of the population, seasonal variation seems to be 

 caused by changes in the behavior of the mussels which affect their 

 catchability. Intensive brail dragging over the beds in the early part 

 of the season may have a disturbing influence and reduce catch success 

 during the latter part of the season. As previously described, the sea- 

 sonal catch success (see table 6) for both 1956 and 1957 was highest in 

 April, declined sharply in May, increased slightly in June, and gradually 

 declined as the season advanced. The trend in fishing effort followed a 

 similar pattern. Unfavorable weather prevented fishing during March of 

 1956 and restricted the amount of fishing done in March of 1957- Weather 

 was not a serious factor, however, during the remaining months of either 

 year . 



Certain species are harvested at a_ higher rate in proportion to 

 their relative abundance than are other species in the popiolation. The 

 pigtoe appears to be especially vulnerable, since the rate of harvest 

 for this species in test areas was higher than the average rate for all 

 other species combined. The crowfoot brail appears to be biased in 

 catching certain sizes of pigtoe mussels. The brail tends to catch size 

 groups larger than 50 mm. in length with a higher frequency rate in pro- 

 portion to relative abundance than it does the smaller size groups. This 

 indicates the smaller and therefore younger stocks, being less vulnerable, 

 are offered some natural protection from fishing pressiire. 



A valid statistical comparison between the two methods used to 

 estimate the size of the pigtoe population in Wheeler Reservoir is not 

 possible because of the different procedvires used. A simple comparison, 

 however, shows that the estimate by SCUBA sampling is about 70 percent 

 as high as that for the tag and recovery method. It was impossible to 

 achieve strict accuracy by either method because of the variables invol- 

 ved, foremost of which was the non-uniform distribution of the mussels 

 on the beds. The actual co^mt method by SCUBA sampling is presumed to 

 provide the most reliable estimate. 



Using the number estimated by SCUBA sampling, population of pig- 

 toe mussels at the end of the 1957 harvest season was 20,564,000 in the 

 productive area of Wheeler Reservoir. Since catch records provided an 

 estimate of 6,024,000 harvested during the previous 2 years, the esti- 

 mated number available at the start of the I956 harvest season was 

 26,590,000, if natural mortality is disregarded. These figiores indi- 

 cate that 23 percent of the population was exploited d\iring the two 

 years. When calculated for each year, the' harvest rates were Ik percent 

 in 1956 and 9 percent in 1957. 



It is obvious that with the high rate of mortality attributed to 

 commercial fishing, an equally high rate of replacement is needed for the 



35 



