FISHERY BULLRTIN: VOL. 86. NO. 4 



No quantitative information is available to correct 

 the estimates for this effect. 



The analyses of the growth data, including both 

 the single molt increments and the ARGRs, clearly 

 indicated high variability in growth rates between 

 age groups, and that site (reef) and season varia- 

 tions (recapture interval) were compounding factors. 

 Estimates of growth made during the 1-yr period 

 of the study were similar to those obtained on reef 

 I in previous studies at Seven Mile Beach (Chittle- 

 borough 1970, 1975, 1976; Chittleborough and 

 Phillips 1975; Joll and Phillips 1984). However, 

 despite the fact that the population densities in 1981 

 were the highest ever recorded, the molt increments 

 of the 3-yr-old P. cygnus were significantly higher 

 than at the low densities in 1971-74. 



Comparisons of the growth data for P. cygnus 

 juveniles showed no significant differences in 

 growth within any age group between reefs I, III, 

 and V. The reduced densities of juveniles on reef 

 III had no apparent effect on either their overall 

 growth rates or their molt increments. This suggests 

 that the food resources on the surrounding seagrass 

 beds may not limit growth within the range of P. 

 cygnus densities present on these reefs during the 

 field experiments. However, other factors may be 

 involved. The foraging ranges of juveniles on reef 

 III may overlap those animals from reef V and other 

 nearby patch-reefs. More recent acoustic tracking 

 studies by Jernakoff (unpubl. data) suggest that this 

 is probably the case. If so, then reducing the den- 

 sity of P. cygnus on reef III might not produce a 

 significant increase in their growth rates, because 

 they could still be sharing their food resources with 

 animals from nearby reefs. 



The best tests of hypotheses about the effects of 

 limited resources are those where the densities of 

 P. cygnus are experimentally manipulated in repli- 

 cated experimental areas and which incorporate ap- 

 propriate controls (Connell 1974; Underwood 1979). 

 This attempt has highlighted a number of problems. 

 Nevertheless, it may be useful to conduct modified 

 manipulation experiments of this kind in the futiire. 

 Obviously, one of the problems with the present ex- 

 periment was the lack of replication, and replication 

 should be incorporated in the design of any future 

 experiments. This was not possible in the present 

 study because of high time and manpower require- 

 ments associated with the use of trapping and mark- 

 recapture techniques. Furthermore, our observa- 

 tions indicate that it will be extremely difficult to 

 find a series of patch-reefs similar enough in size, 

 structure, and other features to serve as true rep- 



licates. As has been shown in this study, even small 

 differences in water depth, or as yet unidentified 

 characteristics, make the selection of reefs as equi- 

 valents very difficult. Selection of such reefs also 

 will not be easy because often several reefs are 

 within the known foraging range of the juveniles. 

 Evaluation and refinement of the visual estima- 

 tion technique also will be necessary before further 

 manipulation experiments are undertaken, as the 

 usefulness of mark-recapture techniques is dubious. 

 Without the development of such a refined method 

 which would permit rapid and frequent estimates 

 of population size, the effect of subsequent changes 

 in population levels cannot be properly monitored. 

 Without such a method it also would not be possi- 

 ble to determine the extent to which migrations of 

 P. cygnus juveniles to and from the reefs are induced 

 by the use of the baited traps or by handling and 

 other disturbances during the mark-recapture 

 process. 



ACKNOWLEDGMENTS 



We thank David Wright, David Evans, Simon 

 Braine, and Leo Olsen of the CSIRO Division of 

 Fisheries Research, Marmion, Western Australia, 

 for their assistance with field and laboratory work. 

 We also thank Frank W. Reneke of San Diego State 

 University and R. Sandland of the CSIRO Division 

 of Mathematics and Statistics for their assistance 

 with computer data summaries and statistical anal- 

 yses. This study was sponsored by the National Sci- 

 ence Foundation U.S. -Australia Cooperative Science 

 Program through NSF grant INT 7927203 to 

 Richard F. Ford and by the CSIRO Division of 

 Fisheries Research. 



LITERATURE CITED 



Bailey, N. T. J. 



1951. On estimating the size of mobile populations from 

 recapture data. Biometrika. 38:293-306. 

 Chittleborough, R. G. 



1970. Studies on recruitment in the Western Australian rock 

 lobster Panulirus longipes cygnus George: density and 

 natural mortality of juveniles. Aust. J. Mar. Freshwater 

 Res. 21:131-148. 



1974a. Home range, homing and dominance in juvenile west- 

 ern rock lobsters. Aust. J. Mar. Freshwater Res. 25:227- 

 234. 



1974b. Development of a tag for the western rock lobster. 

 Rep. Div. Fish. Oceanogr. CSIRO Aust. No. 56, 19 p. 



1975. Environmental factors affecting growth and survival 

 of juvenile western rock lobsters Panulirus longipes (Milne- 

 Edwards). Aust. J. Mar. Freshwater Res. 26:117-196. 



1976. Growth of juvenile Panulirus longipes cygnus George 



786 



