Lovrich and Vinuesa: Reproduction of Paralomis granulosa 



673 



by larger males, and 3) crabs near ecdysis and berried 

 females are less vulnerable to trapping since they do 

 not feed (Miller, 1990). Size at maturity of P. granu- 

 losa may be overestimated because the 57.5-mm-CL 

 size class was poorly sampled; this would particularly 

 affect size at gonadal maturity. Also, the frequency of 

 occurrence of females with late embryonic stages, es- 

 pecially those in pre- and postmolt conditions, may be 

 underestimated by trap sampling. Unfortunately, we 

 have no way to assess these possible biases because 

 we have no trawl surveys with which to compare data. 



Compared with other shallow-dwelling lithodids, P. 

 granulosa has low fecundity and large eggs, resem- 

 bling in this respect the deep-water species: Lithodes 

 ferox (8,000 eggs maximum, 1.97 mm egg diameter; 

 Abello and Macpherson, 1992); L. couesi (5,000, 2.3 mm; 

 Somerton, 1981); L. murrayi (4,200, 2.45mm; Miquel 

 and Arnaud, 1987; Miquel et al., 1985). By contrast, 

 the shallow water Paralithodes camtschaticus and P. 

 platypus carry up to 350,000 and 280,000 eggs, respec- 

 tively, with an average egg diameter of 1.2 mm 

 (Matsuura et al., 1971, 1972; Somerton and Macintosh. 

 1985). Fecundity off! granulosa was less in females 

 with ES V or with asynchronously developing clutches 

 for a constant carapace length (Table 2), because of 

 egg loss. At the end of embryogenesis about 10-12% of 

 the initial brood was lost. Diseases and egg predators 

 are frequent causes of egg loss in crab species (Kuris, 

 1991). However we did not find evidence of epibiosis in 

 broods of P. granulosa. Exceptionally small broods, out- 

 liers in Fig. 7, may have resulted from delayed mat- 

 ing, lack of mates, or to small size of mating males, as 

 occurs in other lithodids (McMullen, 1969; Powell et 

 al, 1973; Paul and Paul, 1990). This question requires 

 further investigation. 



As the energy expended on each offspring increases, 

 the number of offspring that parents produce decreases 

 (Smith and Fretwell, 1974). Thus, in two related spe- 

 cies similar energetic investment may result in many 

 small or few large offspring. However, considering the 

 two lithodids of the Beagle Channel, one finds that P. 

 granulosa biennially produces eggs which are fewer 

 but not larger than those of L. santolla, which annu- 

 ally produces up to 59,000 eggs (Guzman and Cam- 

 podonico, 1972) of 2. 1-2.2 mm in diameter (Vinuesa, 

 1987). 



There is no evidence that the biennial reproductive 

 cycle of P. granulosa is more advantageous (i.e., an 

 adaptative strategy) than the annual cycle of L. 

 santolla. Paralomis granulosa larvae pass through 

 fewer molting events (Campodonico, 1971; Campodonico 

 and Guzman, 1981); thus mortality due to ecdysis is 

 reduced. Shorter larval development would also re- 

 duce predation risks in the plankton. Inhabiting the 



layer of water closest to bottom (Lovrich, unpubl. data) 

 allows larvae to find refuge and thus reduces losses to 

 predation. We speculate that the lesser fecundity of 

 P. granulosa may be partially compensated for by a 

 high survival during their development. In Paralithodes 

 platypus, Jensen and Armstrong ( 1989) interpreted bi- 

 ennial reproduction as a consequence of physiological 

 and energetic constraints incurred by the species in a 

 harsh environment. 



King crab species can be categorized into three 

 groups on the basis of their reproductive cycles: 

 Paralithodes camtschaticus and Lithodes santolla 

 spawn annually, Paralomis granulosa and Paralithodes 

 platypus (except primiparous females) spawn bienni- 

 ally, and finally L. aequispina, L. couesi, and L. ferox 

 spawn asynchronously. All Lithodes species, with the 

 exception of L. santolla, inhabit deep waters whereas 

 Paralithodes species inhabit shallow waters. Otto and 

 Cummiskey (1985) hypothesized that king crabs in- 

 habiting shallow waters (L. santolla, P. camtschaticus, 

 and P. platypus) spawn synchronously during spring 

 while deep-sea king crabs (L. aequispina and L. couesi) 

 have protracted spawning periods. These authors sug- 

 gest that this pattern could be related to a dependence 

 on food sources by shallow water species. Paralomis 

 granulosa inhabits shallow waters and spawns syn- 

 chronously every two years, but we suppose that in 

 this species synchronicity is not related to food depen- 

 dence because larval hatching occurs mainly during 

 winter when neither food nor potential competitors are 

 abundant (Lovrich, unpubl. data). 



Paralomis is a deep-water genus (Takeda et al. 1984; 

 Macpherson, 1988) and P. granulosa is the only spe- 

 cies that inhabits shallow waters. This species prob- 

 ably colonized the Beagle Channel relatively recently, 

 i.e., 8,500 years ago after the last deglaciation occurred 

 (Rabassa et al., 1986). This species still retains certain 

 features of its deep-water relatives: low fecundity, large 

 eggs, protracted reproductive cycle, and independence 

 of larval hatching from food availability. 



Acknowledgments 



This study was funded by CONICET (PID # 3154700- 

 88). We thank the fishing companies Mar Frio S.A. 

 and Pesquera del Beagle S.A. for allowing us to sample 

 on board. Special thanks to Jose T Garcia, Hector 

 Monsalve, and the crew of the fishing boats. Beatriz 

 Gonzalez (Biological Dept, F.C.E.y N., University of 

 Buenos Aires) supervised statistical analysis; Mercedes 

 Cattaneo assisted in correction of the early English 

 version; and Bernard Sainte-Marie (Dept. Fisheries and 

 Oceans, Canada) as well as two anonymous reviewers 



