692 



Fishery Bulletin 90(4), 1992 



Latreille, with and without access to artificial shelters 

 at different spatial scales in seagrass beds. We then 

 discuss these mortality patterns in terms of the relative 

 importance of lobster size, shelter availability, and 

 distance of lobsters from the artifical shelter. More- 

 over, we use daytime abundance and foraging ranges 

 of shelter-associated predators to speculate on the 

 mechanisms underlying these mortality patterns. 



Juvenile P. argus inhabit shallow bays throughout the 

 tropical and subtropical western Atlantic where they 

 frequently aggregate during the day in crevices of coral 

 and rocky reefs (Berrill 1975, Herrnkind et al. 1975). 

 Gregarious behavior within dens probably enhances in- 

 dividual survivorship because spiny lobsters collectively 

 use their spinose antennae to fend off diurnally active 

 predators (Berrill 1975, Cobb 1981, Zimmer-Faust and 

 Spanier 1987, Eggleston and Lipcius 1992). However, 

 intra- and interspecific competition for suitable dens 

 can force smaller juvenile P. argus out of these dens 

 (Berrill 1975). Predation represents a major source of 

 mortality for juvenile spiny lobsters (Munro 1974, 

 Herrnkind and Butler 1986, Howard 1988, Smith and 

 Herrnkind 1992), and when individuals are displaced 

 or forced to shelter in an inadequate den they may be 

 subject to increased predation rates (Herrnkind and 

 Butler 1986, Eggleston et al. 1990). 



Large juvenile and adult spiny lobsters are the focus 

 of intense commercial and recreational fisheries in 

 south Florida and the Caribbean, with the possibility 

 of regional overexploitation of spiny lobster fisheries 

 (U.S. Agency for International Development 1987). 

 Several Caribbean nations have met increased market 

 demand with the large-scale use of artificial shelters 

 to concentrate lobsters and facilitate harvest (e.g., 

 Mexico-Miller 1989, Lozano-Alvarez et al. 1991; 

 Cuba— Cruz and Brito 1986; Bahamas— R.W. Thomp- 

 son, Dep. Fish., Nassau, Bahamas, pers. commun., May 

 1991). These artificial shelters, commonly referred to 



as "casitas Cubanas" (see Fig. 1), attract and concen- 

 trate a broad size-spectrum of juvenile P. argus, 

 particularly in nursery areas (Eggleston et al. 1990, 

 Lozano-Alvarez et al. 1991). 



Predation intensity in and around artificial shelters 

 is affected by numerous factors including the sizes of 

 predator, prey, and shelter (Hixon and Beets 1989, 

 Eggleston et al. 1990), and distance from the reef 

 (Shulman 1985). Moreover, since most crustaceans 

 have indeterminate growth (Hartnoll 1982), they must 

 continually search for larger shelters as they grow, a 

 process that involves predation risk that is inversely 

 related to body size (e.g., Scully 1983, Reaka 1987, 

 Vermeij 1987). Hence, we hypothesized that (1) the 

 relative importance of a lobster size refuge would vary 

 according to shelter availability, and (2) that the im- 

 pact of artificial shelters upon predation-induced mor- 

 tality of juvenile lobsters would vary according to the 

 distance of unprotected lobsters from these shelters. 

 We tested these hypotheses experimentally in the field 

 by quantifying the survival of tethered spiny lobster 

 juveniles in seagrass beds of Bahia de la Ascension, 

 Mexico. This bay is a productive nursery for juvenile 

 Panulirus argus and supports a commercial fishery for 

 large juveniles and adults (Miller 1989, Lozano-Alvarez 

 et al. 1991). Experimental factors included (1) presence 

 or absence of artificial shelter, i.e., casitas Cubanas, 

 (2) lobster size, (3) site, and (4) distance between 

 tethered, unprotected lobsters and artificial shelters. 



Methods and materials 



Study site 



Tethering experiments were conducted in Bahia de la 

 Ascension, a large bay (~740km") within the Sian 

 Ka'an Biosphere Reserve, Mexico (19°45'N, 87°29'W) 

 (Fig. 2). Two experimental sites with contrasting 



Figure 1 



A large "casita Cubana" 

 constructed with a frame 

 of PVC-pipe and roof of 

 cement (177cm length x 

 118cm width x 6cm 

 height of opening). 



