A BEAK KEY FOR EIGHT EASTERN TROPICAL PACIFIC 



CEPHALOPOD SPECIES WITH RELATIONSHIPS BETWEEN 



THEIR BEAK DIMENSIONS AND SIZE 



Gary A. Wolff 1 



ABSTRACT 



A method of identifying the beaks and estimating body weight and mantle length of eight common 

 species of eastern tropical Pacific cephalopods is presented. Twenty specimens were selected from 

 each of the following species: Symplectoteuthis oualaniensis, Dosidiem gigas, Ommastrephes bar- 

 tramii, Onychoteutkis banksii, Abralwpsis affinis, Pterygwteuthis giardi, Liocranchia reinhardti, 

 and Loligo opalescens. Seven dimensions measured on the upper beak and five dimension^ measured 

 on the lower beak are converted to ratios and compared individually among the species using an 

 analysis of variance procedure and Tukey's to. Significant differences (a<0.05) observed among the 

 species' beak ratios means, in addition to structural characteristics, are used to construct artificial 

 keys for the upper and lower beaks of the eight species. Upper and lower beak dimensions are used as 

 independent variables in a linear regression model with mantle length and body weight ( log trans- 

 formed). Two equations are given for estimating the length and weight for each species from the 

 upper or lower beak. One uses the rostral length dimension because of its durability and the second 

 uses a dimension derived from a stepwise regression procedure. 



The importance of cephalopods as prey is well 

 documented for whales (Gaskin and Cawthorn 

 1967; Clarke et al. 1976; Clarke 1977), seals (Aus- 

 tin and Wilki 1950; Laws 1960), seabirds (Ash- 

 mole and Ashmole 1967; Imber 1978), tunas 

 (Pinkas et al. 1971; Matthews et al. 1977), tunas 

 and porpoise (Perrin et al. 1973), and sharks 

 (Clarke and Stevens 1974; Tricas 1979). Due to 

 the rapid digestion of the softer body parts, how- 

 ever, the cephalopod's beak is often the only iden- 

 tifiable structure remaining in these predator's 

 stomachs as evidence of feeding on cephalopods. 

 Consequently, the accuracy of specific identifica- 

 tions and estimates of cephalopod biomass con- 

 sumed by these predators often suffers. 



Two methods have generally been used to ap- 

 proach the problem of characterizing cephalo- 

 pod beaks. A descriptive method was used most 

 notably by Clarke (1962, 1980), Mangold and 

 Fioroni(1966), and Pinkas etal. (1971). Families, 

 genera, and occasionally species were identified 

 from structural characteristics of the beak. A 

 biometric method was used by Wolff (1977) and 

 Wolff and Wormuth (1979) to separate two spe- 

 cies of ommastrephid squid with beak dimen- 

 sions. It was suggested that the method could be 



'Department of Oceanography, Texas A&M University, Col- 

 lege Station, TX 77843; present address: Environmental Engi- 

 neering, Texas A&M University, College Station, TX 77843. 



Manuscript accepted October 1981. 

 FISHERY BULLETIN: VOL. 80, NO. 2, 1982. 



expanded to include other species of cephalo- 

 pods. 



This study presents a key based on structural 

 and biometric differences among the beaks of 

 eight species of cephalopods. The species of ceph- 

 alopods examined were: Symplectoteuthis oua- 

 laniensis (Lesson), Dosidicus gigas (d'Orbigny), 

 Ommastrephes bartramii (Lesueur), Onychoteu- 

 this banksii (Leach), Abraliopsis affinis (Pfeffer), 

 Pterygioteuthis giardi Fischer, Liocranchia 

 reinhardti (Steenstrup), and Loligo opalescens 

 Berry. Regression equations of body weight and 

 mantle length from beak dimensions are also 

 presented. 



MATERIALS AND METHODS 



The cephalopods for this study were obtained 

 from Southwest Fisheries Center, National Ma- 

 rine Fisheries Service, and Invertebrate Collec- 

 tion, Scripps Institution of Oceanography, La 

 Jolla, Calif. Twenty specimens of each species 

 were selected in the maximum mantle length 

 range available. Table 1 shows the ranges for 

 mantle length and body weight and collection 

 locations for the cephalopods. The buccal masses 

 were removed, after the specimens were mea- 

 sured and weighed, and placed in a solution 

 saturated with sodium borate and trypsin (8 g 

 trypsin/1 sodium borate solution) for 6 to 10 d. 



357 



