WOLFF: BEAK KEY FOR EIOHT CEI'HALOPOD SPECIES 



side view 



rw 



top view 



LjwJ 



LOWER 



UPPER 



Figure 1.— Dimensions measured on the upper and lower beak. 



transformation met the criteria for validity as 

 described by Anderson and Lydic (1977). Tukey's 

 (u procedure was used to test for significant dif- 

 ferences (a<0.05) among 21 ratio means from 

 the upper beak and 10 ratio means from the 

 lower beak for each species. This procedure in- 

 volves the computation of a confidence interval 

 from the formula: a» = q a (p, Mi) s?, where w is a 

 range for the treatment means with a given 

 probability level (a<0.05), q is the studentized 

 range, p is the number of treatments, m is the 

 error degrees of freedom and sj is the standard 

 error of the treatment means (Steel and Torrie 

 1960). Simple linear regressions were calculated 

 to express the relationship between a beak 

 dimension and the mantle length and log trans- 

 formed body weight. An AMDAHL 470 V/6 

 computer 2 performed the majority of computa- 

 tions. 



RESULTS 



The results of the ANOVA procedure are sum- 

 marized in Tables 2 and 3. The species' means 

 are ranked and the standard error of the treat- 

 ment mean for each ratio is given. These tables 

 form the basis for the construction of the biomet- 



2 Reference to trade names does not imply endorsement by 

 the National Marine Fisheries Service, NOAA. 



ric portion of the beak key. Combinations of de- 

 scriptive characteristics and significant beak 

 ratios are used to identify the eight species of 

 cephalopods. Separate keys are provided for the 

 upper and lower beak. 



The ratio values presented in the key are mid- 

 points between species' means and often greatly 

 exceed the stated significance level (a<0.05) as 

 indicated by the confidence interval for the spe- 

 cies' means which follows in parentheses. Addi- 

 tional descriptive characteristics and alternate 

 beak ratios are given to corroborate the initial 

 identification. Figures 3-10 show upper and 

 lower beaks for each of the species. A few of the 

 alternate ratios in the upper and lower beak key 

 have species' means which are not significantly 

 different. These ratios can be considered reliable 

 since Hartley (1955) suggested that the experi- 

 mentwise error rate could be relaxed consider- 

 ably below the standard a<0.05 level due to the 

 conservative nature of Tukey's w procedure. 

 Additional alternate ratio values can be deter- 

 mined from Table 2 to distinguish species if the 

 ratios in the key are not satisfactory (e.g., dam- 

 aged beak). The descriptive characteristics fol- 

 low a slightly modified version of Clarke's termi- 

 nology (1962, 1980) with several additions as 

 shown in Figure 2. This key should be used with 

 caution on specimens which are greatly outside 

 the size range of this study. 



359 



