OSTEOLOGICAL DEVELOPMENT AND VARIATION IN YOUNG 

 TUNAS, GENUS THUNNUS (PISCES, SCOMBRIDAE), FROM THE 



ATLANTIC OCEAN' 



Thomas Potthoff^ 



ABSTRACT 



The development and variability of osteological and meristic features obtained from 427 

 juvenile Thunnus from 8 to 117 mm SL are described. The juveniles of T. atlanticus, T. 

 thynnus, and T. alalunga can be identified. Thunnus obesus cannot be separated from T. albacares , 

 but both are separable as the "Thunnus spp. complex" from the other three species. Identification 

 methods are discussed with emphasis on the features of the axial skeleton, number of gillrakers 

 over the ceratobranchial bone, pterygiophore pattern under the second dorsal fin, and the shape 

 of the lateral line. 



This paper describes the development of osteo- 

 logical features and their variability for the iden- 

 tification of juvenile Thunnus from 8 to 100 mm 

 standard length (SL). To date, this has not been 

 attempted in an orderly and systematic fashion 

 for the species in the Atlantic Ocean, mainly 

 because all juveniles have the same general 

 external appearance as T. atlanticus shown in 

 Figure 1. Nevertheless, some species have been 

 previously identified by using mostly external 

 characters: Sella (1924), Schaefer and Marr 

 (1948), Wade (1950, 1951), Mead (1951), Padoa 

 (1956), Jones (1960), Matsumoto (1961), Marchal 

 (1963a, b), and Yabe, Ueyanagi, and Watanabe 

 (1966). The identifications were probably correct, 

 although other species of juvenile Thunnus 

 could fit these same descriptions. Scaccini (1961) 

 published on a series of juvenile Thunnus from 

 the Mediterranean and stated that they were 

 T. thynnus, but he did not reveal how he arrived 

 at his identifications. Ueyanagi (1967) mentioned 

 the occurrence of small T. alalunga in the North 

 and South Atlantic Oceans, but he gave no identi- 

 fication methods. Klawe and Shimada (1959) and 

 Klawe (1961) examined juvenile Thunnus ma- 

 terial, using external and osteological characters, 

 but they had doubts as to the correctness of their 

 identifications. Watson and Mather (1961) used 

 the soft X-ray method to distinguish between the 

 species. One of their major characters was the 



vertebral position of the first ventrally directed 

 parapophysis. I found' this character to be of 

 limited value, because the vertebral position of 

 the first parapophyses changes with growth. 

 Potthoff and Richards (1970) used osteological 

 characters to identify two species of juvenile 

 Thunnus from bird stomachs, and Juarez (1972) 

 described larvae of T. atlanticus, also on the 

 basis of osteological methods. Other researchers 

 have used adult osteological characters on speci- 

 mens larger than 100 mm (Yabe et al., 1958; 

 Nakamura and Kikawa, 1966). 



I was not entirely successful in separating all 

 species. Using osteological characters, T. alba- 

 cares and T. obesus were not separable from each 

 other as juveniles (from 8 mm to about 100 mm 

 SL), but together they can be separated from the 

 other three Thunnus species in the Atlantic 

 Ocean. Thus, I have lumped them together as the 

 "Thunnus spp. complex." 



Adult characters from the works of Kishinouye 

 (1923), Frade (1932), Godsil and Byers (1944), 

 de Sylva (1955), Watson (1964), Nakamura 

 (1965), and Gibbs and Collette (1967) formed the 

 basis for my study. At first I identified the largest 

 fish in my collection and then worked down to 

 the smaller sizes, noting changes that occurred. 



METHODS 



'Contribution No. 229, Southeast Fisheries Center, National 

 Marine Fisheries Service, NOAA, Miami, Fla. 



^Southeast Fisheries Center, National Marine Fisheries 

 Service, NOAA, Miami, FL 33149. 



Manuscript accepted October 1973. 

 FISHERY BULLETIN: VOL. 72. NO. 



1974. 



The specimens used in this study were first 

 measured with dial calipers or calibrated ocular 

 micrometers and the standard length (SL) from 



563 



