FRY: NATURAL STABLE CARBON ISOTOPE TAG 



-23i 



-21 



COMPOSITE 



-13 -15 -17 -19 -21 -23 

 AVERAGE INDIVIDUAL 6^\ 



Figure 2. — Agreement between two methods of estimating 

 mean shrimp 8'^C at 12 bay stations. The mean 8^^C value of five 

 or six individuals is plotted as the x coordinate; the S"C value of 

 the composite sample from which the individuals were subsam- 

 pled is plotted as the y coordinate. The 45° reference line indi- 

 cates complete agreement between the two methods. 



cle tissue taken from the same shrimp showed a 

 maximum 8^^C range of 0.3L (three shrimp, each 

 sampled thrice), and subsampling one piece of tis- 

 sue per shrimp was therefore considered adequate. 



Shrimp were sorted by species by hand, and 

 identification checked with a dissecting micro- 

 scope on 5-10 specimens per composite sample 

 using several keys (Zamora and Trent 1968; 

 Perez-Farfante 1970; Wood 1974). 



Composite and individual samples were dried in 

 an oven or freeze dryer, then shredded through a 

 #40-mesh screen with a Thomas-Wiley MilP or 

 ground to a fine powder with mortar and pestle. 

 Samples were combusted in a modified Leco radio 

 frequency furnace (Parker et al. 1972) and ana- 

 lyzed for 8^^C values using a Nuclide 6-60 dual 

 collector isotope ratio mass spectrometer. All re- 

 sults are expressed relative to the international 

 PDB^ standard where: 



13ri/12 



8''C = 



^' *^sample 



-1 X lo^ 



standard 



Based on analysis of 53 paired replicate samples, 

 the mean error ±95% confidence limits was 

 0.45±0.77'L. 



RESULTS 



Bay Shrimp 

 Figure 3 shows the 8^^C values of brown shrimp 



from all bay stations sampled during the spring 

 and early summer. The sampling effort was un- 

 equal at these stations, primarily due to fluctua- 

 tions in shrimp numbers from week to week. At 

 most stations, three or more composite samples 

 were collected at irregular intervals over the 2-mo 

 period; individual subsamples of pooled composite 

 samples were collected between 1 and 15 May. 



Analysis of variance showed highly significant 

 differences between station means computed from 

 composite samples (P<0.01). The LSD multiple 

 range test showed that bay stations formed at 

 least three significantly different groups on the 

 basis of their mean S^^C values (Figure 3; vertical 

 bars at right margin). Most stations were open bay 

 stations and had mean 8^^C values between - 17.9 

 and -19.6L (stations 60-84). Mean S'^C values 

 were higher in sea grass meadows, —12.8 to 

 - 15.41.. Several open bay stations formed a third 

 group, having more negative average values of 

 -20.1 to -2L7'L (stations 80-13). 



Analysis of the open bay stations showed that 

 the more negative shrimp 8^^C values tended to 

 occur in lower salinity bays which have been 

 historically well flushed by freshwater inflows (Ta- 

 ble 1). 



Table l. — Brown shrimp 8'^C and salinity at open bay stations 

 and freshwater flushing rates for five Texas bays and the Gulf of 

 Mexico. 



'Based on composite samples only 



^Salmity taken at the time of shrimp collection (mId-Aprll to mid-June 1978). 

 ^Turnover rate Is the ratio of annual freshwater inflow volume to the volume of 

 a bay Calculated from Table 8, DIener 1975. 

 ■•Station 13 In the Old Arroyo Colorado is not Included In this average. 



Size and Bay Brown Shrimp 8^^C 



Systematic 8^^C changes with size might be ex- 

 pected if shrimp foods available to different size 

 classes possessed different 8^^C values. Shrimp 

 sizes ranged from 20 to 110 mm total length (TL), 



^Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



■"Abbreviation in the isotopic literature for a fossil belemnite, 

 Belemnitella americana, from the PeeDee formation in South 

 Carolina. 



339 



