26 



Fishery Bulletin 88(1), 1990 



4-33 mm CL. The percentage contribution by size class 

 to the total beam trawl catch showed relatively little 

 change between 8 and 18 mm CL (Fig. 6) suggesting 

 that selection bias was not a problem within this range. 

 On this basis, we are confident in using a restricted 

 range of the catch (7-17.66 mm CL) for the growth 

 estimates. However, our growth-rate estimates for 

 sizes above this range may be influenced by emigra- 

 tion and avoidance, and should be used with caution. 

 Similarly, the growth rates determined for small 

 shrimp between the time of recruitment into the estu- 

 ary and vulnerability to our trawl should be regarded 

 as rough approximations because of the reduced effi- 

 ciency of our gear and absence of a precise immigra- 

 tion time. 



Growth 



The mean growth-rate estimates obtained in our study 

 are lower than those reported for P. indicus in other 

 areas. This may be a consequence of the lower tem- 

 peratures which prevail at the edge of this species' 

 distribution. Parrack (1979) suggested that differences 

 in the growth rates of P. aztccus populations from 

 different latitudes could be explained by temperature 

 differences in the bottom waters of each region. Juve- 

 nile P. i7idicvfi grew at 0.102 mm CL per day in Singa- 

 pore ponds (Hall 1962) while Le Reste and Marcille 

 (1976) recorded more rapid growth of 0.125 mm CL 

 per day in Madagascar. A review of the growth rates 

 recorded for this species (Champion 1983) suggested 

 an average growth of 30 mm total length per month 

 which equates to 0.176 mm CL per day when the con- 

 version factor of Prahhakara Rao (1967) is applied and 

 a 30-day month assumed. 



Application of the von Bertalanffy growth function 

 may be justified for some penaeid shrimp (Garcia and 

 Le Reste 1981); however, its use is questionable in 

 areas where growth is reduced at times by low temper- 

 atures. The reduced growth of overwintering cohorts 

 accounts for the relatively long residency period of the 

 shrimp in St. Lucia. Between August and October 

 1983, the water temperatures in the Narrows were 

 18-23°C. The lowest growth rates from our study oc- 

 curred over 18-22° C, which may provide an estimate 

 of the low-temperature growth threshold for P. indicuf^ 

 in St. Lucia. Comparable data for P. indicus were 

 unavailable; however, our estimate of the low-temper- 

 ature gi'owth threshold is similar to estimates reporteci 

 for other penaeid species. Latapie et al. (1972) found 

 negligible growth in P. aetiferus below 20°C, and 

 Phares (1980) used 17°C as the zero growth tem- 

 perature in a model for the same species. Below 22°C, 

 the growth of juvenile P. vannamei was reduced in 

 Mexico (Edwards 1977). In contrast, Zein-Eldin and 



Aldrich (1965) reported slow growth of postlarval 

 P. aztecuH at 18°C and a lower growth threshold of 

 11°C. 



In several instances, the interval growth rates were 

 inversely related to temperature. This may be a con- 

 sequence of size-specific changes in growth rates over- 

 riding the effects of temperature. Hall (1962) reported 

 linear growth up to 25 mm CL for P. indicus; however, 

 there is no evidence that more complex growth models 

 were fitted. In juvenile P. vannamei (Menz and Blake 

 1980) and juvenile P. sefifenis (Phares 1980) growth 

 was negatively correlated with size. While growth rates 

 probably decline with increasing size in P. indicus, the 

 observed fluctuations may also be due to changes in 

 the thermal history of each cohort not manifested in 

 the mean interval water temperature. 



Emigration 



The life cycle of P. indicus in St. Lucia appears to 

 follow the generalized pattern outlined for penaeids by 

 Garcia (1985). There appear to be two cohorts each 

 year, the first entering during early summer and 

 emigrating from the system over autumn and early 

 winter; the second recruits during autumn, also over- 

 winters in the estuary, and emigi'ates during the follow- 

 ing spring and summer. A similar pattern of recruit- 

 ment and emigration was observed for P. indicus in 

 Singapore by Hall (1962) and in Madagascar by Le 

 Reste (1978). This pattern has also been noted for 

 P. /*/f'r(/«(>riSi's in Australia (Dredge 1985, Rothlisberg 

 et al. 1985). In an earlier study of the St. Lucia pen- 

 aeids, Joubert and Davies (1966) reported emigration 

 through summer and autumn but did not indicate that 

 this protracted emigration actually represented the 

 overlapping departures of two cohorts. 



Penaeus indicus appears to emigrate from St. Lucia 

 and other systems over similar size ranges. In Singa- 

 pore, Hall (1962) attributed a decline in the frequency 

 of 20-25 mm CL shrimp to emigration, and in Mada- 

 gascar P. indicus began to emigrate at approximately 

 20 mm CL. Prabhakara Rao (1967) suggested that 

 P. indicus left Chilka Lake, India, between 18.3 and 

 22.9 mm CL. 



Jayakody and Costa (1988) repotted that P. indicus 

 may emigrate from Sri Lankan estuaries in response 

 to osmotic stress imposed by monsoon rains. The onset 

 of emigration by P. mer-guiensis has been linked to 

 elevated rainfall and declines in salinity (Rothlisberg 

 et al. 1985, Staples and Vance 1986). Joubert and 

 Davies (1966) suggested a link between migration and 

 salinity in St. Lucia but had no experimental data to 

 support this connection. Emigration appeared to be 

 related to late-summer and autumn water cooling; 

 however, the apparent correlation does not inifjly 



