Posterior parasplienoid margin variable in shape, 

 rounded, concave, or somewhat angulate (as in T. 

 atlanticus and T. tonggol) hut never with a i)ro- 

 noiniced angle. 



Vertebrae 18+21 (as in all Thnnnus except T. 

 atlanticus). First ventrally directed parapophysis 

 on vertebra (as in all Thutnnis excejit T. thj/nn)is 

 and T. tonggol). First closed haemal arch usually 

 on vertebra 11 (as in T. atlanticus, T. tonggol, T. 

 ohcsus and often in T. thi/nnus). Anteriormost 

 haemal prezygapophyses arising far ventrad on 

 liacnial arch (as in T. atlanticus and T. tonggol). 

 Haemal ])ostzygai)ophyses long (as in T. atlanticus 

 and T. tonggol), the longest about 75 percent of its 

 c(>ntrum length (somewhat shorter than in T. atlan- 

 //(■(/.< and T. tonggol). Anteriormost ventrolateral 

 foramina large, their width three or more times that 

 of haemal spine (as in T. atlanticus and T. tonggol). 

 Least height of centrum of 3nth vertebra 1.2-1.9, 

 usually 1.3-1.') in centrum length (resembling T. 

 alahmga, but in that species the vertebrae taper, 

 whereas in T. albacarrs they are of nearly etiual 

 width throughout). 



Nominal Species 



More names have been proposed for sujjposedly 

 different ])opulations and individual variants of T. 

 albacarcs than for all other species in the genus. 

 Jordan and Evermann (1926) took the most extreme 

 position in recognizing seven species: catalinar, from 

 the Cf.'ifornia coast; vjacroptcrus, from the central 

 and western Pacific; itosibi, a long-finned form from 

 Hawaii and Japan; alhacora, from the eastern Atlan- 

 tic; albacores, from Madeira and the West Indies: 

 allisoni, a western Atlantic long-tinned form; and 

 zacallcs from Hawaii (wtiich has heretofore been 

 considered as most clo.sely related to T. tonggol, see 

 above). The main characters that they used to 

 separate tiiese forms were the lengths of the s(>cond 

 dorsal and anal lobes. I'sing the same characters, 

 Ginsburg (1953) distinguished an eastern Atlantic 

 alhacares, a western Atlantic .suhulatus, an eastern 

 Pacific catalinar, and a central and western Pacific 

 macroplerus. It became apparent to us that T. 

 albacarrs is an extremely variable species mori)ho- 

 metrically, from our own data and from the many 

 detailed moriihometric studies on pojiulations of T. 

 albacarcs, especially in the Pacific, by workers such 

 as Godsil (1948), Schaefer (1948, 1952, 1955), 

 Schaeferand Walford (1950), Godsil and Creenhood 



(1951), Poyce (1953), Tsuruta (1954, 1955, 1961), 

 Km-ogane and Hiyama (1957a, 1958b), Nakagome 

 (1958), Broadhead (1959), Legand (19(10), and Frade 

 (1931a, for the eastern Atlantic). 



Statistical analysis of morphometric data indi- 

 cates that many sul)popuIations of 7'. albacarcs are 

 differentiated, but certainly not to a species or sub- 

 species level. Royce (1965), in a monumental study 

 of the morphometry of T. alhacares, showed con- 

 clusively that it is a single, locally variable, pantrop- 

 ical species. He found that the dilTerences between 

 eastern Atlantic and eastern Pacific specimens were 

 less than the differences between eastern Pacific and 

 Caroline Islands specimens, and that several char- 

 acters change clinally from west to east in the e(iua- 

 torial Pacific. 



Tiiere has been considerable confusion concerning 

 the name Thgnnns argentivitlatns Cuvier. The 

 original description (Cuvier, in Cuvier and Valen- 

 ciennes, 1831: 97-98) was based on two specimens 

 now at the M\HX in Paris: one from the Atlantic, 

 collected by Quoy and Gaimard (MNHN A.5572) 

 and one from the Indian Ocean, sent by Dussumier 

 (MNHX A.5567). Schaefer and AValford (1950) 

 reported that, according to information received 

 from L. Bertin, the description was based on three 

 specimens: the two already mentioned and a third 

 from the Indian Ocean, coast of Malabar, sent by 

 Dussumier (MNHN A. 581 4; given erroneously as 

 A.5816 by Schaefer and Walford). .\.5814, a speci- 

 men in alcohol, was designated the lectotype by 

 Schaefer and Walford (1950, p. 441), who thus recog- 

 nized the Indian Ocean yellowfin as Xrothunnus 

 argentivittatns, the Pacific form as .V. macropirrus, 

 and the Atlantic form as .V. albacora. Based on 

 A.5814 being the lectotype, Uivas (1961) used the 

 name argentivittatns for an Indian Ocean tuna which 

 he tentatively placed in the subgenus Paralhunnus, 

 and regard(>d as different from Neothunnus albacarrs, 

 (lie yellowfin (una, which he considered to be a single, 

 paiitropical species. 



We have examined the supi)os('d lec(otyi)e 

 (A.5814) and believe it, and the other specimens in 

 Rivas' (19GI) account, actually to be T. tonggol. 

 Watson (1964) reached the same conclusion, and 

 suggested that 7'. argrntiritlalus be synonymized with 

 T. tonggol. This aclion, to begin with, is inappro- 

 priate, for argrnliriltatus has ijriority over tonggol. 

 A.581 1, however, (■aiiiio( lie considered as the lecto- 

 type of Thgnnu.^ argrntirillatns, as it is nowhere men- 



106 



U.S. FISII .\ND WII.DI.IFK SKItVICE 



