the first 2 mo. after release, the average 

 length of male shrimp increased by about 

 1-1/4 in., whereas the average length of 

 females increased by about 1-3/4 in. 



Although we have not made a final analysis 

 of information from this experiment, it appears 

 at present that the data can be used to deter- 

 mine the size at which Tortugas pink shrimp 

 should be harvested, 



Charles E. Knight, Project Leader 



SIZE COMPOSITION OF COMMERCIAL 

 SHRIMP LANDINGS 



Reliable information about harvests is 

 necessary to manage any fishery properly. The 

 information we need includes total weights or 

 numbers caught, size composition of the catch, 

 fishing locations, methods of capture, and 

 amount of fishing effort expended. Information 

 of this nature on commercial shrimp har- 

 vests from the Gulf of Mexico is obtained 

 directly from the shrimp industry and is re- 

 ported monthly in the Bureau of Commercial 

 Fisheries publication "Gulf Coast Shrimp 

 Data." 



Because the industry uses two different 

 means for measuring the size composition 

 of landings, the comparability of these methods 

 must be determined. Shrimp unloaded at com- 

 mercial shrimp houses are purchased from 

 fishermen on the basis of the number of head- 

 less shrimp per pound, i.e., 21 to 25,26 to 30, 

 etc. The size of shrinnp in a landing is deter- 

 mined by the box-grading or machine-grading 

 nnethods. Box grading is accomplished by 

 taking several 5-lb. samples of shrimp as 

 they are unloaded. The average number of 

 shrimp per pound in these samples deter- 

 nnines the size category assigned to the land- 

 ing. A box-graded landing is often assigned 

 to one or two size categories when, actually, 

 it consists of shrimp of a large range of 

 sizes. Machine grading is the nnechanical 

 sorting of a landing into several size cate- 

 gories. 



To determine the reliability of size composi- 

 tion data from the two methods of grading, 

 we selected and measured samples of shrimp 

 from landings at a number of ports. At shrimp 

 houses using the box-grading method we meas- 

 ured lengths as the shrimp passed along the 

 conveyor belt between a vessel and the house 

 (fig. 17). Samples of shrimp from machine- 

 graded landings were measured after grading. 

 The length measurements were later converted 

 to number of shrimp per pound for comparison 

 with the graded- size information. 



Figure 18 compares the count size of 

 sampled shrimp with the reported size cate- 

 gories obtained from the box and machine 

 nnethods of grading. Actual size composition 

 of the box-graded shrimp has a much greater 

 range than that implied by the reported size 



Figure 17. — Measuring shrimp from cnnveyor belt. 



BOX -GRADED 

 LANDINGS 



MACHINE -GRADED 

 LANDINGS 



60-70 31-35 21-25 60-70 31-35 21-25 



36-40 26-30 36-«2 26-30 



SI2E GROUP SAMPLED 



Figure 18. — A comparison of the size composition of 

 shrimp from box- and machine-graded landings. Circles 

 represent average size, and vertical lines indicate the 

 range in which two-thirds of the measured shrimp fell. 



groups. The size range for the nnachine- 

 graded shrimp is considerably snnaller, but 

 overlapping of adjacent size groups occurs. 



In the examples shown, the average count 

 size of the measured shrimp usually fell within 

 the size categories reported by the shrimp 

 house. In some box-graded samples (not illus- 

 trated) the average count size fell well outside 

 the reported size categories and a false picture 

 of shrimp sizes was obtained. Machine grading 

 depicted more accurately the true sizes of 

 shrimp in landings. 



Because of the large range of sizes in 

 box-graded landings, we must characterize 

 better the size of the shrimp landed at certain 

 ports. Information from future sampling should 

 enable us to adjust the reported size data for 

 use in population studies. 



James P. Clugston, Project Leader 



23 



