382 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



sizes in Louisiana landings provides a synoptic 

 picture of population size structure in what is 

 considered the nucleus of these stocks. Height- 

 ened spawning in November-December and in 

 June-July may be inferred, respectively, from 

 offshore recruitment surges in May-June (ligtt 

 arrows) and again in November-December (dark 

 arrows). This pattern is quite similar to that 

 described for populations in the northeastern Gulf. 

 Year-to-year variation in extent of maximum 

 spawning activity and timing of recruitment is 

 again obvious; but a major distinction when com- 

 paring reproductive patterns for stocks in both 

 areas is the enhanced significance of late-season 

 broods (light arrows) to offshore and inshore fish- 

 eries in the northwestern Gulf areas. Secondary 

 yield and biomass modes occurring in May or 

 June (figs. 28A and 28B) are attributed in large 

 part to late-season broods supplementing rem- 

 nants of the prior year's early-season brood (fig. 

 28C). Populations giving rise to early- and late- 

 season broods are believed to be predominated by 

 survivors of the previous year's corresponding 

 broods. The degree of predominance appears to 

 vary widely, however, being largely dependent 

 upon the relative initial strength and subsequent 

 survival of each brood comprising a spawning 

 population. 



The foregoing description of the white shrimp's 

 seasonal reproductive pattern agrees to some ex- 

 tent with that already given by Lindner and 

 Anderson (1956). Also in general accord with the 

 findings of these authors are gross conclusions 

 that may be drawn from figure 28C regarding 

 growth in recruited (offshore) population phases. 

 If this figure gives a reasonably faithful picture 

 of spawning class progress, note on curves tracing 

 broods in populations fished off Louisiana that 

 growth in weight is, on the average, compara- 

 tively slow during the period November- April. 

 This agrees with statements made by Lindner 

 and Anderson (1956) who used increase in body 

 length rather than increase in weight as the growth 

 criterion. By means of tagging experiments they 

 showed that white shrimp of most sizes (105-175 

 mm. total length) ordinarily fished by the offshore 

 fleet experienced reduced growth during winter 

 months, and that growth during this season was 

 approximately constant regardless of size. Over 

 the remainder of the year, growth rates, as would 

 be expected, were much greater in the smaller 



sizes (105-125 mm. total length at release) than 

 in the larger sizes (155-175 mm. total length at 

 release). 



Compared to rate of growth measured in terms 

 of length, rate of growth in weight is fairly low 

 in the small sizes, increases to a maximum some- 

 where in the middle of the shrimp's overall size 

 range, and then tapers off as the maximum attain- 

 able size is approached. Using increase in weight 

 as the growth criterion and maintaining corre- 

 spondence with Lindner and Anderson's results 

 based on length increments, note in figure 280 

 that seasonal growth varied from year to year. 

 Thus the late-season spawning class of 1955 (Lou- 

 isiana Coast, brood B) apparently grew more 

 rapidly the following November-April (1956-57) 

 than did those of 1956 and 1957 (broods E and G) 

 during the winters of 1957-58 and 1958-59, re- 

 spectively. Average size in the 1956 class, for 

 example, only increased from that equivalent to 

 24, to not quite 16 whole shrimp to the pound 

 over the period December- April, about a .30 per- 

 cent weight increase. This is contrasted to a 150 

 percent increase in average weight for the 1955 

 class during the corresponding season a j^ear earlier. 



The principal lesson derived here is that popu- 

 lation growth in white shrimp (and very likely 

 other species as well) is dynamic and therefore 

 difficult to predict. Mark-recapture studies can 

 only contribute growth estimates derived over 

 short periods of time from a limited number of 

 individual animals. Such estimates may be ques- 

 tionable not only from the standpoint of overall 

 representativeness, but also from the standpoint 

 of expected consistency in space and time. Before 

 resource productivity can be projected, average 

 growth in populations treated as units and broken 

 down insofar as possible according to their com- 

 ponent age classes, is the factor demanding meas- 

 urement. This is best achieved in the case of 

 exploited populations through analyses of appro- 

 priate commercial statistics. Current statistics, 

 unfortunately, provide only a crude or "qualita- 

 tive" picture of population growth. Progressive 

 elimination of data biases should provide increas- 

 ingly accurate growth parameter estimates to- 

 gether with some indication of their expected 

 variability. For the present, however, eye esti- 

 mates of optimum growth from serial alignment 

 of what are considered representative weight- 

 frequency distributions must suffice. 



