468 



Fishery Bulletin 100(3) 



SCL size classes and were censored (randomized selection 

 of n=105) in each of the two size classes (Table 1). 



Strandings analyses 



The Sea Turtle Stranding and Salvage Network (STSSN) 

 documents dead or injured sea turtles along the coasts 

 of the eastern United States and the U.S. Caribbean 

 (Schroeder, 1989). The STSSN relies on a trained group 

 of volunteers, including state and federal employees and 

 private individuals, to collect basic biological data on 

 each stranded turtle. Each animal is identified to species, 

 the condition or state of decomposition is determined, 

 standard carapace measurements are taken, and any 

 obvious wounds, injuries, or abnormalities are noted and 

 described. Volunteers who have received additional train- 

 ing may also perform necropsies, or internal exams, on 

 a carcass to determine the general state of health of the 

 animal prior to death, to determine sex, and to locate any 

 obvious internal abnormalities. Data are recorded on stan- 

 dardized report forms that are submitted first to a state 

 coordinator and then to the national STSSN coordinator 

 at the National Marine Fisheries Service, Southeast Fish- 

 eries Science Center, Miami, Florida. 



The species-specific predictive regression equations 

 from the morphometric analyses were used to estimate 

 the carapace width for each turtle in the STSSN database 

 for which this measurement had not been taken and to 

 estimate the body depth for each turtle. For turtles with 

 curved measurements only, straight line carapace lengths 

 were estimated from curved carapace lengths (CCL) be- 

 fore estimating body depth and carapace width by apply- 

 ing equations reported by Teas ( 1993). 



Within each region (Fig. 2) carapace widths were com- 

 pared with the currently required minimum widths of 

 TED openings and body depths were compared with the 



currently required minimum heights of TED openings. 

 Stranded turtles that were reared in captivity, cold- 

 stunned, or known to have been captured incidentally were 

 censored. 



Results 



Morphometric analyses 



Loggerhead sea turtles The relationships between cara- 

 pace width and carapace length and between body depth 

 and carapace length were linear. Coefficient of determina- 

 tion (/■-) values of regressions with log-transformed data 

 were slightly ( <0.002 ) higher than values based on untrans- 

 formed data. Regression of each of the morphometric values 

 on carapace length was highly significant (P<0.0001) and 

 resulted in the following predictive equations: 



In sew = -0.0225 + (0.9507 x In SCL) [n=250, /-■-=0.9891, 



In BD = -0.5682 -t- (0.9100 x In SCL) \n='250, r2=0.9661. 



Straight line carapace lengths corresponding to turtles 

 with carapace widths of 81.28 cm (32 inch; the minimum 

 width ofTED openings in the Gulf of Mexico) and 88.90 cm 

 (35 inch; the minimum width of TED openings in the At- 

 lantic) were 104.5 cm and 114.9 cm, respectively. Straight 

 line carapace lengths corresponding to turtles with body 

 depths of 25.40 cm (10 inch is the minimum height of TED 

 openings in the Gulf of Mexico) and 30.48 cm (12 inch is 

 the minimum height of TED openings in the Atlantic) 

 were 65.3 cm and 79.8 cm, respectively. 



Green sea turtles The relationships between carapace 

 width and carapace length and between body depth and 



