CAILLOUET ET AL.: TRENDS TOWABD DECREASING SIZE OF SHRIMP 



Tables. — Trends in slopes (6) of regressions of transformed cumulative percentage (In Fl on size category (C) for brown and white 

 shrimp, Mississippi River to Texas and Texas coast, 1959-76 vs. 1961-76 (based on data from Tables 1, 2; Figures 2. 3). 



' Equals slope of the regression of P on x where x is the last two digits of each year 



-The change in slope (ft) per year was significantly different from zero at the 95% level of confidence. 



There was a positive correlation (r = 0.702) be- 

 tween the slopes of regressions of transformed 

 cumulative percentage on size category for brown 

 and white shrimp from Mississippi River to Texas 

 in 1959-76, that was significantly different from 

 zero at the 99% level of confidence. The same was 

 true (r = 0.742) for brown and white shrimp from 

 the Texas coast. This indicated that the direction 

 of the shift in size composition of reported catches 

 within a given year was usually in the same di- 

 rection for both species in a given State (Tables 1, 

 2). 



For a given weight of reported annual catch, 

 the ex-vessel value of shrimp harvested in 

 Louisiana is considerably less than that in Texas 

 (Caillouet and Patella 1978), and this is largely a 

 function of the size composition of the respective 

 catches in the two States. Our analysis cannot 

 distinguish whether the observed differences and 

 trends in size composition of the reported catches 

 are due to differences and trends in fishing mor- 

 tality, natural mortality, or growth, but we 

 suggest that the predominant causes of the ob- 

 served differences and trends are differences and 

 trends in fishing mortality. There is no evidence 

 to indicate that separate shrimp stocks exist in 

 these two States, or that natural mortality or 

 growth differ between the two States (see 

 Christmas and Etzold 1977). On the other hand 

 the number and size of shrimp fishing craft and 

 other indices of fishing effort are different in the 

 two States and have increased over time (Christ- 

 mas and Etzold 1977; Caillouet and Patella 

 1978). Also, differences and trends in time phas- 

 ing of peak fishing activity in Texas and 



Louisiana within a year could have contributed to 

 the differences and trends in size composition re- 

 ported herein. Regardless of the cause or causes, 

 continued shifts in size composition toward grea- 

 ter proportions of smaller shrimp in the catches 

 can be expected to weaken the ex-vessel value of 

 the catches. 



LITERATURE CITED 



Caillouet, C. W., and F. J. Patella. 



1978. Relationship between size composition and ex- 

 vessel value of reported shrimp catches from two Gulf 

 Coast States with different harvesting strategies. Mar. 

 Fish. Rev. 40(21:14-18. 

 CHRISTMAS. J. Y., AND D. J. ETZOLD (editors). 



1977. The shrimp fishery of the Gulf of Mexico United 

 States: A regional management plan. Gulf Coast Res. 

 Lab., Ocean Springs, Miss., 128 p. 

 HENDERSON, F. 



1972. The dynamics of the mean-size statistic in a chang- 

 ing fishery. FAO Fish. Tech. Pap. 116, 16 p. 

 KUTKUHN,J. H. 



1963. Gulf of Mexico commercial shrimp populations — 

 trends and characteristics, 1956-59. U.S. Fish. Wildl. 

 Serv., Fish. Bull. 62:343-402. 



National Marine Fisheries Service. 



1970-78. Gulf coast shrimp data, annual summary 1967- 

 annual summary 1976. U.S. Dep. Commer., NOAA, 

 Natl. Mar, Fish. Serv.. Curr. Fish. Stat. 5412, 5721, 5925, 

 6126, 6425, 6725, 6925, 7225. 

 RICKER.W. E. 



1975. Computation and interpretation of biological statis- 

 tics of fish populations. Fish. Res. Board Can.. Bull. 

 191, 382 p. 

 U.S. FisH AND Wildlife Service. 



1960-69. Gulf coast shrimp data, annual summary 

 1959— annual summary 1966. U.S. Fish. Wildl. Serv., 

 Curr. Fish. Stat. 3 unnumbered publ., 3358, 3515, 3784, 

 4111,4411,4781,5107. 



989 



