Arkhipkin et al.: Distribution, stock structure, and growth of Berryteuthis magister 



Microstructure of the middle layer Laminae were 

 visible as faint microstructural ridges on the dorsal 

 surface of the proostracum (Fig. 6E). Each lamina 

 consisted of a broad growth layer with a rather 

 smooth surface confined by a narrow layer with a 

 convoluted surface. Average width of laminae varied 

 (depending on squid size) from 0.4-0.5 mm at 50 mm 

 ML to 0.7-0.9 mm at 250-300 mm ML. In most speci- 

 mens the laminae were best resolved in the middle 

 part of the gladius. The general illegibility of lami- 

 nae in the proostracum made it impossible to calculate 

 individual growth rates with this gladius structure. 



Statolith microstructure 



Growth increments were observed in all statolith 

 preparations. The nucleus was oval-shaped and had 

 a maximum diameter of 40-45 \im along the rostrum 

 axis (Fig. 7). The postnuclear zone lay beyond the 

 nucleus; its color resembled that of the dark and pe- 

 ripheral zones. Approximately 10-12 narrow and 

 rather uniform increments were observed within the 

 postnuclear zone, each with a maximum width of 4.3- 

 6.5 |im. Maximum radius of the postnuclear zone was 

 65-75 p.m. The postnuclear zone was separated from 

 the dark zone by the first prominent growth incre- 

 ment which was assumed (after Natsukari et al., 

 1993) to be the "natal ring." Within the dark zone 

 and sometimes in the peripheral zone, three types of 

 growth increments were revealed. The first-order 

 growth increments occurred mainly within the inner 

 part of the dark zone. They were very narrow (ca. 1— 

 1.2 um) and faint, corresponding to the "fine rings" of 

 Natsukari et al. (1993). However, they could not be 

 seen in all statolith preparations, especially within the 

 outer part of the dark zone and in the peripheral zone. 

 The second-order growth increments (corresponding 

 to the "heavy rings" of Natsukari et al., 1993) were 

 prominent and seen both in the dark and peripheral 

 zones. Their width varied from 4.7-7.1 um in the 

 inner portion of the dark zone in the direction of the 

 lateral dome to 9.5-12.2 um in the proximal portion 

 of the rostrum, from 3.5 to 4.3 um in the midpart of 

 the rostrum, and from 4.7 to 5.4 um in the distal 

 portion of the rostrum. Usually 4-6 first-order growth 

 increments were observed within each second-order 

 increment in the inner portion of the dark zone, and 

 3 or 4 increments within the peripheral zone (if they 

 were present). In both the middle and distal portion 

 of the rostrum, second-order increments were 

 grouped within third-order increments: 4-7 second- 

 order growth increments in each third-order incre- 

 ment. These third-order growth increments were not 

 noted in Natsukari et al. (1993). 



Validation of statolith growth increments 



The presence of three different types of growth in- 

 crements within statoliths prevented us from assum- 

 ing that any of them were deposited daily. There- 

 fore, we attempted to validate the periodicity of 

 growth-increment deposition on the basis of two in- 

 dependent and indirect methods. 



Comparison of growth increment numbers in sta- 

 toliths and gladii We investigated statoliths and 

 gladii within the entire size range of B. magister 

 sampled from early juveniles (35^10 mm ML) to large 

 mature adults (280-330 mm ML). The total number 

 of laminae within the inner layer of the gladius cor- 

 responded well with the total number of second-or- 

 der growth increments within the statoliths (Fig. 8A). 

 The estimated value of the slope was close to 1 (Table 

 2), which indicates that growth increments in sta- 

 toliths and laminae within the inner layer of gladii 

 were laid down synchronously. The results of incre- 

 ment validation from the first method were checked 

 by the following second method. 



Growth-increment analysis of successive modal 

 classes The method of comparison of difference be- 

 tween mean ages of successive modes with elapsed 

 days is relevant when 1 ) good resolution of the modes 

 is possible, and 2) these modes represent the same 

 cohort of squid, without substantial emigration and 

 immigration (Caddy, 1991). It is known that sexual 

 maturation affects the growth of both sexes and that 

 during maturation squid either slow or cease somatic 

 growth (Forsythe and van Heukelem, 1987). Hence, 

 data on B. magister males were inappropriate for our 

 purpose because most of them were mature during 

 the period of our study (Figs. 3 and 4) and, there- 

 fore, a mode of male length composition might con- 

 sist of several cohorts. Length-frequency composi- 

 tions of females in Olyutorsky Bay were rather com- 

 plex, especially during the summer ( Fig. 3 ), and thus 

 we were prevented from choosing progressive modal 

 classes. Unfortunately, we had no statoliths from 

 well-distinguished modal classes of females in the 

 Navarin-St. Matthew region (Fig. 5). However, in 

 the Navarin-Olyutorsky region, a clearly distin- 

 guishable modal group of females (immature in 

 June— July, gradually maturing in August-Septem- 

 ber) was traced within the catches from June to Sep- 

 tember. The modal sizes of these females increased 

 progressively from 180-190 mm in June to 250-270 

 mm in September, suggesting that these females 

 belonged to the same cohort (Fig. 4). This group of 

 females was analyzed from four successive monthly 

 samples taken from June to September. 



