216 



Fishery Bulletin 94(2), 1996 



readers. The results of the statistical tests were 

 equivocal but suggested that there was a potential 

 bias between the readers' counts. However, the ap- 

 parent bias was relatively small and did not suggest 

 that there were substantial differences between the 

 readers' recognition of increments. 



An examination of L. pealei marked with oxytet- 

 racycline supported the hypothesis that statolith in- 

 crements are formed daily (Macy, 1995). In particu- 

 lar, one marked squid that did grow appreciably dur- 

 ing a 21-d period was found to have a total of 21 sta- 

 tolith increments (Macy, 1995). To date, the daily 

 increment hypothesis has been verified by statolith 

 marking for several squid species, including Illex 

 illecebrosus (Dawe et al., 1985), Alloteuthis subulata 



Figure 2 



(Al Schematic view of posterior face of Loligo peaU'i statolith 

 with a ventro-dorsal cross section indicated from points a to 

 a,. (B) Cross section (a, a,] from (A) showing the curved plane 

 where growth increments are counted. ( C I After initial grind- 

 ing removes the cross-hatched area to reveal the nucleus ( 1 ), 

 the statolith is turned over 1 2 1. (D) After the anterior side of 

 the statolith is oriented with the dorsal dome elevated and 

 the cross-hatched material removed 1 1 ), the statolith is turned 

 over again to show the posterior face (2). IE) The lateral dome 

 is elevated while the resin cools and the cross-hatched mate- 

 rial is removed. iFl The polished statolith is covered with mol- 

 ten resin and allowed to cool with the lateral dome elevated. 



(Lipinski, 1986), Loligo opalescens (Jackson, 1994b), 

 Sepioteuthis lessoniana (Jackson, 1990a; Jackson et 

 al., 1993), Loliolus noctiluca and Loligo ehinensis 

 (Jackson, 1990b), Todarodes pacific-its (Nakamura 

 and Sakurai, 1991), Abralia trigonura (Bigelow, 

 1992), Ommastrephes bartramii (Bigelow and 

 Landgraf, 1993), Onychoteuthis borealjaponica 

 (Bigelow, 1994), and validation is in progress for 

 Loligo plei. 4 The daily increment hypothesis has also 

 been verified by using laboratory-reared squid of 

 known age for Loligo opalescens (Jackson, 1994b) and 

 for Sepioteuthis lessoniana (Jackson et al., 1993), both 

 loliginids, and is considered to be the best working hy- 

 pothesis for statolith increment formation in L. pealei 

 on the basis of limited marking data and by analogy 

 with other loliginid species. 



Month of hatching was backcalculated for all 

 samples by subtracting the estimated age, based 

 on statolith analysis, from the date of sample col- 

 lection. Samples were categorized by sex (indeter- 

 minate, female, and male) and maturity stage (im- 

 mature, maturing, and mature) to examine 

 whether size at age was related to maturity stage. 

 Samples were also categorized by month and year 

 of hatching to examine monthly and seasonal pat- 

 terns in size at age. Average monthly growth rates 

 of individual squid were computed to examine in- 

 dividual variation in growth, where the growth 

 rate in length was ML (mm) divided by age 

 ( months ), and the growth rate in weight was total 

 wet weight (g) divided by age. Summary statistics 

 of the average monthly growth rate of individuals 

 in ML and in weight were computed for squid cat- 

 egorized by sex, maturity stage, hatch month, and 

 hatch year. Variances of growth rate in length and 

 weight were log-transformed and tested for homo- 

 geneity by sex, maturity, hatch month, and hatch 

 year by using Bartlett's test for homogeneity of 

 variances (Sokal and Rohlf, 1981). Unplanned 

 multiple comparison tests appropriate for equal 

 variances and unequal samples sizes (Sokal and 

 Rohlf, 1981 ) and for unequal variances and sample 

 sizes (Games and Howell, 1976; Day and Quinn, 

 1989) were applied to test whether growth rate in 

 length and weight differed by sex, maturity stage, 

 hatch month, and hatch year. 



Growth model 



We applied the general growth model of Schnute 

 ( 1981 ) to quantify the relationship between length 



4 Jackson, G. 1994. Department of Marine Biology, James 

 Cook Univ., Townsville, Queensland 4811, Australia. 

 Personal commun 



