Abstract. — A principal mechan- 

 ism underlying a production hypoth- 

 esis that artifical reefs increase en- 

 vironmental carrying capacity and 

 eventually the biomass of reef-asso- 

 ciated organisms is that these struc- 

 tures reduce predation on reef res- 

 idents. We tested this predation 

 mechanism with a series of field ex- 

 periments at two sites (inner-bay 

 sand-seagrass flat, and outer-bay 

 seagrass bed adjacent to coral reefs) 

 in Bahia de la Ascension, Mexico. We 

 examined survival of two size-classes 

 of juvenile Caribbean spiny lobster 

 Panulirus argus tethered in sea- 

 grass beds with and without access 

 to artificial lobster shelters, and at 

 different distances from the shelters. 

 The artificial shelters were concrete 

 structures (casitas) that simulate 

 lobster dens. Large juvenile lobsters 

 (56-65 mm CL) attained a relative 

 size refuge when tethered 60 m away 

 from casitas compared with smaller 

 (46-55 mm CL) lobsters. Conversely, 

 the small lobsters survived better 

 beneath casitas than did large lob- 

 sters. Small juveniles also survived 

 better at casitas or 30 m away from 

 casitas than at 15 m or 70 m away. 

 Observations indicated that the day- 

 time predator guild, composed pri- 

 marily of snappers (family Lutjani- 

 dae), seldom foraged more than 60 m 

 from casitas and were typically with- 

 in 15 m of casitas. There was also a 

 significant positive correlation be- 

 tween predation-induced lobster 

 mortality and numbers of snapper 

 associated with casitas at the inner- 

 bay site. Thus, tethering lobsters 70 

 m away from casitas appeared ade- 

 quate to examine survival of lobsters 

 in an environment uninfluenced by 

 daytime predators aggregating to 

 casitas. These results indicate that 

 (1) the relative importance of a lob- 

 ster-size refuge from predators 

 varies according to shelter availabil- 

 ity, and (2) that there is a nonlinear 

 relationship between predation risk 

 and distance from an artifical 

 shelter. Our results demonstrate that 

 casitas increase survival of small 

 juvenile lobsters but reduce survival 

 of larger juveniles. Small casitas 

 scaled according to body size may 

 enhance survival of large juvenile 

 lobsters in nursery habitats where 

 large conspecifics are removed from 

 large casitas. 



Manuscript accepted 13 July 1992. 

 Fishery Bulletin, U.S. 90:691-702 (1992). 



Artificial shelters and survival 

 of juvenile Caribbean spiny lobster 

 Panulirus argus: Spatial, habitat, 

 and lobster size effects* 



David B. Eggleston 



The College of U/illiam and Mary, Virginia Institute of Marine Science 



Gloucester Point, Virginia 23062 

 Caribbean Marine Researcfn Center, Lee Stocking Island. Exuma Cays, Bahamas 

 Present address: College of Ocean and Fishery Sciences WH-10 



University of Washington, Seattle, Washington 98195 



Romuald N. Lipcius 



The College of William and Mary, Virginia Institute of Marine Science 



Gloucester Point, Virginia 23062 

 Caribbean Marine Research Center, Lee Stocking Island, Exuma Cays, Bahamas 



David L. Miller 



Department of Geography, State University of New York, Cortland. New York 13045 



Artificial reefs are in use worldwide 

 as a means of increasing local abun- 

 dance of finfish and invertebrates 

 (see reviews by Bohnsack and Suth- 

 erland 1985, Grove and Sonu 1985, 

 Mottet 1985, Bohnsack 1989). The 

 use of artificial reefs to increase fish- 

 eries production remains controver- 

 sial because it is unknown whether 

 these structures (1) provide critical 

 resources that increase the environ- 

 mental carrying capacity and even- 

 tually the biomass of reef-associated 

 organisms (production hypothesis), 

 or (2) merely attract and aggregate 

 organisms from surrounding areas 

 without increasing total biomass (at- 

 traction hypothesis) (Bohnsack 1989). 

 The attraction hypothesis is an im- 

 portant consideration for artificial- 

 reef-based fisheries that may be vul- 

 nerable to overexpioitation. Thus, 

 there is a need for ecological inves- 

 tigations capable of assessing the im- 

 pact of artificial reefs upon species 

 distribution, abundance, and survival 



* Contribution 1725 of the Virginia Institute 

 of Marine Science. 



patterns, and the processes underly- 

 ing these patterns. 



Artificial reef technology has tradi- 

 tionally been based on the assump- 

 tion that obligate reef dwellers (e.g., 

 reef fishes and lobsters) are limited 

 locally or regionally by the availabil- 

 ity of shelter (Bohnsack 1989, Hixon 

 and Beets 1989, Eggleston et al. 

 1990 and references therein). Con- 

 versely, artificial reefs also concen- 

 trate numerous potential predators 

 (Hixon and Beets 1989, Eggleston et 

 al. 1990); increased predation pres- 

 sure at or near these structures could 

 outweigh the benefits from increases 

 in production. For instance, fishes 

 and lobsters normally dispersed over 

 a wide area could be concentrated 

 and consumed by predators more 

 rapidly in a smaller area. Thus, arti- 

 ficial shelters may either enhance or 

 reduce the survival of their inhabi- 

 tants, depending upon predator 

 responses. In this paper, we present 

 the results of a series of field ex- 

 periments comparing survival rates 

 of two size-classes of juvenile Carib- 

 bean spiny lobster Panulirus argus 



691 



