FISHERY BULLETIN: VOL. 80, NO. 3 



perhaps more suitable for large-scale applica- 

 tions. 



There are obvious similarities between this 

 classification and that of Vovk (1972), and it ap- 

 pears that, except for stage-5 spent squid, the 

 maturity stages are comparable in both systems. 

 It should be reiterated, however, that the stage 

 characteristics and average parameter values 

 given in this report (Fig. 3a, b; Table 3) were 

 identified after the classification functions had 

 been determined, whereas in the Vovk system 

 the stage indicators form the basis for classifica- 

 tion. The offshore squid sampled by Vovk also 

 appear to have been larger by at least 3-5 cm at 

 each stage than those used in this study. Thus, 

 NGL indices in the Vovk study are 6.4-25% great- 

 er than those given in Table 3. A TLI, surpris- 

 ingly, was not used. Vovk did employ hectocoty- 

 lus length, but in L. pealei the hectocotylized arm 

 may be difficult to identify, particularly in small 

 males, and, more important, may be lost or dam- 

 aged during trawl capture of squid. These and 

 other smaller differences between the methods 

 appear to result mainly because Vovk did not 

 sample the actively spawning inshore popula- 

 tion. His method generally works well for experi- 

 enced personnel, but dissection and weighing of 

 the reproductive tracts and weighing the whole 

 squid take more time and equipment than may 

 be available in the field. At least two more char- 

 acters must be recorded for each sex as well. 



Two simple but objective classifications were 

 developed for Japanese squid, Todarodes pacifi- 

 cus (Hayashi 1970) and for female short-finned 

 squid, Illex illecebrosus (Durward et al. 1978). 

 Hayashi computed a numerical index, M, equal 

 to the weight of Needham's sac (NW) divided by 

 NW plus the testis weight, or to the oviduct 

 weight (odW) divided by odW plus the ovary 

 weight. If the computed value of M is <0.5, equal 

 to 0.5, or between 0.5 and 1.0, the squid is consid- 

 ered immature, mature, or spent, respectively. 

 Since spent L. pealei are at least rare, the system 

 reduces to a two-stage classification which offers 

 no obvious advantage over the immature-mature 

 distinction used by Summers (1968). The dissec- 

 tion and weighing of the two tissues are addi- 

 tional drawbacks. 



Durward et al. (1979) used data initially ob- 

 tained from six /. illecebrosus which matured in 

 captivity but had not spawned to develop a five- 

 stage maturity scale for females. These investi- 

 gators showed (by scatter plots and regression 

 analyses) that relative NGI's correlated well 



with identifiable stages of ovarian development. 

 Thus only two parameters, DML and NGL, were 

 needed for classification. As Durward et al. have 

 pointed out, the critical values of the NGI for 

 Illex for the first four stages are very similar to 

 those for Loligo (Table 3). However, in L. pealei 

 NGI ranked last of four variables in importance 

 (Table 2), and even the most significant variable 

 (AEO) alone yielded only 77.8% correct classifi- 

 cation accuracy. This simple and objective classi- 

 fication for Illex is now widely used ( Amaratunga 

 and Durward 1979). 



ACKNOWLEDGMENTS 



This research was funded in part by an NSF 

 Grant for Improving Doctoral Dissertations. The 

 author gratefully acknowledges the vital com- 

 puter and boat time provided by the University 

 of Rhode Island and the Graduate School of 

 Oceanography, the able drafting of Betsy Wat- 

 kins, and the assistance of S. B. Saila during the 

 field studies and preparation of the manuscript. 



LITERATURE CITED 



Amaratunga, T., and R. D. Durward. 



1979. Standardization of data collection for the short- 

 finned squid, Illex illecebrosus. ICNAF Sel. Pap. 5, p. 

 37-41. 



Anderberg, M. R. 



1973. Cluster analysis for applications. Acad. Press, 

 N.Y., 359 p. 



Anderson, T. W. 



1958. An introduction to multivariate statistical analy- 

 sis. John Wiley & Sons, N.Y., 374 p. 

 Arnold, J. M., and L. D. Williams-Arnold. 



1977. Cephalopoda: Decapoda. In A. C. Giese and J. S. 

 Pearse (editors), Reproduction of marine invertebrates, 

 Vol. IV, p. 243-290. Acad. Press, N.Y. 

 Brown, M. B. (editor). 



1977. BMDP-77 Biomedical computer programs P-ser- 

 ies. Univ. California Press, Berkeley, 880 p. 



BURUKOVSKI, R. N., AND A. N. VOVK. 



1974. Some questions of oogenesis in the Northern Amer- 

 ican squeed (Loligo pealei Les.) from the Georges Sand- 

 bank. Arkh. Anat. Gistol. Embriol. 66:44-50. Fish. 

 Mar. Serv. (Can.), Transl. Ser. 3711. 



Cohen, A. C. 



1976. The systematics and distribution of Loligo (Cepha- 

 lopoda, Myopsida) in the western North Atlantic, with 

 descriptions of two new species. Malacologia 15:299- 

 367. 



Durward, R. D., T. Amaratunga, and R. K. O'Dor. 



1978. Maturation index and fecundity for the female 

 squid Illex illecebrosus. Fish. Mar. Serv. (Can.), Tech. 

 Rep. 833, p. 24.1-24.10. 



Fields, W. G. 



1965. The structure, development, food relations, repro- 

 duction, and life history of the squid Loligo opalescens 



458 



