MATSUMOTO ET AL.: LARVAL TUNA IDENTIFICATION 



EVALUATION OF CHARACTERS 



In identifying fish larvae collected in plankton 

 nets, the easiest and perhaps the only recourse 

 is to identify the largest stage and work down 

 to the smallest. Unfortunately, very few tuna 

 larvae above 9 mm SL are taken in plankton 

 .j\et tows so that this process cannot be followed 

 W all times and identification, therefore, must 

 depend upon those nonadult characters that are 

 the most distinctive and consistent throughout 

 the size range. 



Characters that have been used in the past 

 were reviewed and evaluated. A resume of the 

 usefulness of the various characters follows. 



Meristic 



The number of myomeres is useful in sepa- 

 rating Katsutvomcs pelamis (42-43) and Euthi/n- 

 nus lineatus (38-39) from other tunas, including 

 other species of Euthynnus, all of which have 

 similar numbers of myomeres (40-41). The 

 number of fin rays and spines are not useful for 

 separation of Thunmis because all species are 

 similar in this respect. 



Morphological 



Shape of first dorsal fin, when completely 

 formed, is useful to distinguish late larval stages 

 of K. pelamis, Euthynnus, and Auxis from those 

 of Thunnus. Preopercular spines are unreliable 

 . because they undergo rapid growth changes and 

 position of eye relative to longitudinal axis of 

 body needs to be determined more accurately. 

 Distribution (number and position) of pterygio- 

 phores in the second dorsal fin in relation to 

 neural spines is useful in separating several 

 Thunnus species, but only after these bones 

 have ossified in larvae longer than 10 mm SL. 

 Other characters of the axial skeleton useful in 

 identification, such as the position of the first 

 haemal arch and the position of the zygapophyses 

 on the vertebrae, also form late. 



Measurements 



Morphometries have not been used extensively 

 to date, although there may be some with good 

 possibilities, such- as the relations of body depth 

 to standard length, snout length to head length, 



and snout length to orbit diameter. Some of 

 the reasons for not using measurement data are 

 that the larvae not only shrink in preservatives, 

 but the degree of shrinkage varies in different 

 preservatives and with duration of preservation; 

 the distortion of the body at the time of fixing 

 cannot be controlled; and, more important, there 

 are too few larvae in undistorted condition for 

 reliable measurements. Added to these are other 

 sources of variability such as rapid changes in 

 body parts due to growth, changes which often 

 occur in spurts, and distension of the abdomen, 

 as well as stretching of the body at each feeding. 



Pigmentation 



For the most part black pigment patterns 

 have been the most widely used and accepted 

 character in identifying tuna larvae. There are 

 variations and changes in black pigment patterns 

 on tuna larvae due to growth, but in certain 

 areas of the body these patterns have been found 

 to be consistent enough for identification pur- 

 poses. This is particularly true of pigment pat- 

 terns on the first dorsal fin, posterior half of the 

 trunk, forebrain, and tips of both jaws. The 

 larval size at which black pigment cells appear 

 in certain areas of the body, especially at the 

 upper and lower jaw tips, may be useful in sep- 

 arating T. albacares from T. alahinga. Red pig- 

 ment patterns, although not species specific, have 

 been useful in confirming the identification of 

 T. alalunga when used in conjunction with black 

 pigment patterns. 



Of all the characters reviewed and examined, 

 pigment patterns, both black and red, were con- 

 sidered to be the most reliable for identification 

 of the larval stages, despite their known varia- 

 bility, when supplemented by the use of certain 

 morphological characters such as the distribu- 

 tion of pterygiophores in the second dorsal fin 

 and characteristics of the vertebral column, 

 whenever these are developed. 



VERIFICATION OF RED PIGMENTATION 



Ueyanagi (1966) reported on the usefulness 

 of red pigmentation in identifying tuna larvae. 

 Up to then identification of tuna larvae by pig- 

 mentation had been based on black pigment only. 



3 



