NOTE May Ku et a\ Morphological differentiation between small [uveniles of Farfantepenaeas biasiliensis and F notialis 



307 



lagoon (21'^^26'-21°38'N, 87°30'-88°15'W) was 

 sampled from November 1996 to April 1997, 

 whereas the Yalahau lagoon (21''26'-21°36'N, 

 87°08'-87°29'W) was sampled from June 2001 

 to May 2002. These coastal lagoons are consid- 

 ered the main nursery areas for F. brasiliensis, 

 accounting for 80-95% of the total shrimp abun- 

 dance, followed by F. notialis (May-Kii, 1999; 

 May-Kii and Ordofiez-Lopez, 2000). Shrimps 

 (4-10 mm CL) were preserved in 10% formal- 

 dehyde, prior to examination. Rostrum length 

 (RL; distance from the tip of the rostrum to the 

 postorbital margin of the carapace) and carapace 

 length (CL; distance from the postorbital margin 

 to the posterior margin of the carapace) were 

 measured with an ocular micrometer to the near- 

 est 0.1 mm. Furthermore, the shape and position 

 of the rostrum with respect to the segments 

 and flagellum of the antennule were considered 

 (Perez-Farfante, 1969; 1970, 1971a). Shrimps 

 ranging from 8 to 10 mm CL were identified to 

 species level according to the characters out- 

 lined by Perez-Farfante (1970, 1971a). The rela- 

 tionship between rostrum length and carapace 

 length (RL-CL) was determined for each spe- 

 cies (F. brasilie?isis. F. notialis) in each lagoon, 

 and fitted to the linear function RL = a + b CL, 

 where a and b are coefficients. One-way analysis 

 of covariance (ANCOVA) was used to evaluate 

 differences between species in the RL-CL rela- 

 tionship for each site, with RL as the dependent 

 variable and CL as the covariate. Assumptions 

 of homoscedasticity and homogeneity of slopes 

 (parallelism) were met after log-transformation 

 of RL. Lastly, the RL/CL ratio was estimated 

 to remove the differences in individual sizes 

 between species, then plotted against CL and 

 determined for each lagoon. A one-way analy- 

 sis of variance (ANOVA) was conducted to test 

 interspecific differences between mean RL/CL 

 in each lagoon. The original RL/CL values were 

 log-transformed to fulfill ANOVA assumptions. 



Results 



A total of 13,234 shrimps in the size range 

 4-10 mm CL were collected. Of the 727 organisms 

 measured in the Rio Lagartos lagoon, 74.6% were 

 F. brasiliensis and 25.4% F. notialis. In Yalahau 

 lagoon, from the 12,507 organisms measured 

 91.7% were F. brasiliensis and 8.3% F. notialis 

 In the two lagoons, both species showed a signifi- 

 cant positive linear RL-CL relationship (r->0.92 

 P<0.001). Significant differences were detected 

 between the species (ANCOVA: P<0.001; Fig. 2 

 Table 1) — F. brasiliensis having a longer rostrum 

 than F. notialis for a given CL. 



The RL/CL ratio for both species ranged from 

 0.42 to 0.94. In both lagoons the scatter plots 



Figure 1 



Location of Rio Lagartos and Yalahau lagoons off the Yucatan Penin- 

 sula. The distribution of Farfantepenacuf brasiliensis and F. notialis 

 along the Mexican coast of the Gulf of Mexico and Caribbean Sea is 

 indicated by oblique lines. 



9- 

 8- 



7 

 6 

 5 

 4 



I '^ 

 £ 2-1 



01 = -0 427 + 840 CL, f^ = 0.96 . i ."tl 1'':'m''»*»5 ' 



. ■;;,;i;iii"^-- ^ V •: 



* . ' ■'»'' 



.!-' 



:t:,Mi« 



RL = -0 321 + 679 CL; f2 = 0.94 



° ft 



7- 

 6- 



5- 

 4- 

 3- 

 2- 



6 7 8 



Carapace lengtfi (mm) 



10 



Figure 2 



Relationships between rostrum length (RL) and carapace length 

 (CL) fitted by the linear model RL = a + h CL for juveniles of 

 Farfantepenaeus brasiliensis (▲) and F. notialis (0) from (A) 

 Rio Lagartos and (B) Yalahau lagoons. All regressions are 

 highly significant (P<0.001l. 



