Brodziak and Macy: Growth of Loligo pealei 



215 



75 74 



Figure 1 



Collection sites for long-finned squid, Loligo pealei, samples off the northeast coast of the 

 United States. 



Statolith ageing 



A video image processing system with high-resolu- 

 tion monochrome video camera (MTI-Dage 65, 

 Newvicon) mounted on a compound microscope (Zeiss 

 Universal) was used to count increments (Macy, 

 1995). Because of the low contrast seen in L. pealei 

 statolith microstructure, image enhancement was 

 employed to identify growth marks. Statolith incre- 

 ments were counted by using transmitted light with 

 the microscope stage condenser (brightfield) set to 

 the smallest aperture to obtain maximum depth of 

 field. A rotatable polarizing filter was also placed 

 above the field condenser to improve contrast be- 

 tween increments. Within a prepared L. pealei sta- 

 tolith, a series of paired bands that alternate from 

 light to dark is observable in the LD, beginning at 

 the nucleus and continuing to the LD margin. Two 

 fractures that radiate outward from the nucleus are 

 also apparent. These fractures are not the result of 

 statolith preparation because they are also found in 

 unprepared statoliths. In this study, the successive 

 circumferential light bands in the statolith (appear- 

 ing black in enhanced images) were counted and will 

 be referred to hereafter as "growth increments." 



Counting began with the first check or natal ring 

 (Lipinski, 1986; Natsukari et al., 1988, Saleh-Eddine, 

 1991), which is a distinct oblong-shaped mark 



roughly the same size (100-120 pm) as the major 

 axis length of a statolith from a L. pealei embryo. Al- 

 though prehatching increments can sometimes be ob- 

 served (Saleh-Eddine, 1991; Macy, 1995), their period- 

 icity, if any, is unknown. Typically, counting proceeded 

 along a series of transects extending outward from the 

 first check across the LD to the edge. Up to six fields of 

 view were used to enumerate the increments. 



The consistency of increment counts between the 

 reader of this study and another experienced reader 

 was evaluated post hoc by using a sample of 76 L. 

 pealei collected during 1994. Three statistical tests 

 were applied to determine the consistency between 

 age readings (cf. Campana et al., 1995): a linear re- 

 gression analysis of paired counts; a paired t-test; 

 and a paired Wilcoxon rank test. Results of the lin- 

 ear regression of one reader's counts on the other's 

 indicated that the regression slope was not signifi- 

 cantly different from 1 (6=1.046, ct 6 =0.033, P=0.17) 

 and that the regression intercept was not signifi- 

 cantly different from 0(o = 1.655, cr a =4.697, P=0.73). 

 This suggested that the readers' age determinations 

 were consistent. Results of the paired i-test, how- 

 ever, indicated a significant mean paired difference 

 of7.6days(a=2.1,P<0.001). Similarly, results of the 

 Wilcoxon signed rank test indicated a significant 

 median paired difference (P<0.001) existed between 



