Macewicz et al.: Fecundity, egg deposition, and mortality of Loligo opolscens 



323 



ent study all atretic losses would be attributed, of course, 

 to ovulation and spawning but the chances of this being a 

 major error seem low. Because we counted atretic as well 

 as normal oocytes, atretic losses would be erroneously 

 attributed to spawning only if atresia had proceeded to 

 the point that the atretic structure could not be identified 

 as that of an oocyte in whole-mount preparations under a 

 light microscope ( 64x power). The time at stage for atretic 

 oocytes in L. opalescens ovaries, as well as other squid, is 

 unknown. The duration of alpha-stage atresia of yolked 

 oocytes in anchovy is about a week at 16°C (Hunter and 

 Macewicz, 1985b) and we suspect for the larger L. opal- 

 escens yolked oocyte that the alpha-stage duration may 

 be even longer. The disappearance of unyolked atretic oo- 

 cytes, as an oocyte-like structure that would be counted, 

 is more difficult to dismiss because so little is known 

 about this atretic stage and its duration. If our estimate 

 of the average longevity of spawning female is only about 

 1.67 days, then atretic losses of even small unyolked oo- 

 cytes is probably not an important bias. It would be useful 

 if a way could be found to estimate oocyte resorption rates 

 in squid although it may be very difficult. It seems more 

 important to validate our preliminary estimate of the 

 average longevity of spawning squid, because if true, any 

 concerns regarding atresia could be dismissed. 



Mature females without postovulatory follicles in their 

 ovaries made up only 6% of the 247 females examined 

 histologically. The rarity of these females in our collections 

 reduced the precision of our potential fecundity estimate. 

 Only thirteen of the fifteen females classed as a mature 

 preovulatory female were usable for estimating potential 

 fecundity, further reducing the sample size. Such a small 

 sample size not only results in a low precision but raises 

 the concern that the sample may not be representative 

 of the stock as a whole. The fact that the average total 

 fecundity of females with high mantle condition from the 

 catch was close to the predicted value based on the thir- 

 teen females, indicates that the latter estimate may not be 

 biased. Clearly a larger sample size is needed, particularly 

 if egg escapement is used to monitor the fishery. It would 

 be helpful, in obtaining more samples, if we knew the 

 reason for the apparent rarity of mature preovulatory L. 

 opalescens females. One possibility is that females might 

 pass rapidly from the initial vitellogenesis to ovulation, 

 perhaps in the course of a single day or some fraction of it, 

 and ovulation might begin sometime in the evening when 

 L. opalescens are the most vulnerable to fishing. Another 

 possibility is that mature preovulatory females aggregate 

 in regions not heavily fished by either our trawl or the 

 fishery. 



Egg escapement 



A practical suggestion from this study is the idea of man- 

 aging spawning-ground loliginid fisheries by monitoring 

 the fecundity of the catch and computing the fraction of the 

 potential fecundity spawned. Monitoring the escapement 

 of eggs from the fishery is an attractive approach for Loligo 

 opalescens because costs are moderate, unlike the high 

 cost for monitoring egg beds that cover many locations 



offshore and occur at any time of the year, and because 

 traditional fishery assessment models are difficult to 

 apply or inappropriate at the present time (PFMC, 2002). 

 To proceed with escapement fecundity as a management 

 tool, it would be necessary to set a target level for egg 

 escapement and to relate escapement to egg-per-recruit 

 analysis so that fishing effort could be adjusted to alter 

 egg escapement rates. Conceptual work along these lines 

 has been completed (Maxwell 3 ) 



As mentioned earlier, as a practical matter in applying 

 the egg escapement method, one would need to use Q SP , 

 the mean fraction of the potential fecundity escaping (Eq. 

 1), as a proxy for the more comprehensive and more use- 

 ful measure of egg escapement R e tmaz , the fraction of the 

 expected lifetime fecundity deposited (Eq. 12). Obviously, 

 Q SP will always be lower than R e-tm „ because the denomi- 

 nator of Q SP (the fraction E SP IE P ) is potential fecundity 

 which will always be larger than the denominator for 

 R e t mal , which is expected lifetime fecundity (E). Although 

 quite a different value, Q SP is a useful proxy for R e tnai . If 

 natural mortality (m ) and egg deposition rates (v) are con- 

 stant, changes in fishing mortality will result in changes 

 in Q SP that are proportional to the change in R eJmiI . 



However, changes will not be proportional if either v 

 or m varies. If there is reason to believe that m and v are 

 varying significantly, the use of Q SP as a proxy for R etm!Lll 

 should be undertaken with caution. 



A point of concern in applying this method is that it may 

 be difficult to substantially change escapement of eggs by 

 regulating fishing effort. Our model indicated that egg es- 

 capement may be relatively insensitive to changes in fish- 

 ing mortality if natural mortality rates are as high as we 

 believe them to be. Of equal importance to management is 

 the need to protect egg beds from damage by nets and to 

 monitor the catch to prevent any change that might result 

 in the capture of significant numbers of female L. opal- 

 escens before they begin to deposit eggs. Thus the fraction 

 of the catch that is immature females must be monitored 

 if the stock is managed by using the egg escapement 

 method. For simplicity, our calculations of escapement 

 were based on only mature females because immature 

 females were only 2.6% of the females in the catch 

 (1998-99) and their inclusion had little effect on param- 

 eter estimates. Egg escapement would decrease with an 

 increase in the fraction of immature in the catch. As none 

 of the fecundity of a captured immature female escapes 

 the fishery, a relatively small increase in the fraction 

 of immature animals in the catch can have significant 

 consequences. 



From the standpoint of fishery management, the most 

 important unanswered question regarding the reproduc- 

 tive biology of L. opalescens is "how long do they remain on 

 the spawning grounds?" or the equivalent question "what 



3 Maxwell, M. R.. L. D. Jacobson, and R. Conser. Unpubl. 

 data. Managing squid stocks using catch fecundity in an 

 eggs-per-recruit model. Southwest Fisheries Science Center, 

 National Marine Fisheries Service, 8604 La Jolla Shores Dr., 

 La Jolla, CA 92037. 



