358 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 





JAN FES MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 



Figure 5. — Relative increment of the last ring plotted by 

 month of capture of the fish. 



of ring formation. Similarly, a plot of the abso- 

 lute increment against time of capture resulted 

 in a random scatter of points. 



DISCUSSION 



Previous investigators working on albacore age 

 and growth have reported the presence of definite 

 rings on the inner surface of the centrum of alba- 

 core vertebrae which, they believed, could be 

 used as age indicators. They concluded that 

 each of these rings represented an annulus and, 

 therefore, assigned fish to specific age groups after 

 making counts of these rings. Aikawa and Kato 

 (1938) found that Uno's mean lengths (Uno 

 1936a, b) for age groups IV, V, and VI fell within 

 the length range that they had established for 

 these same age groups. Partlo (1955), on the 

 other hand, found his ring radii values to be at 

 variance with those observed by Aikawa and 

 Kato, but stated that the disagreement was 

 largely caused by the latter 's not recognizing the 

 ring closest to the vertex of the cone. Aside from 

 this discrepancy, there was remarkable agreement 

 between their ring radii values. Figueras (1955), 

 who based his work largely on Partlo 's technique, 

 concluded that his results were in fair agreement 

 with the latter. 



Aikawa and Kato, Figueras, and Uno, so far as 

 can be determined, have not satisfactorily estab- 

 lished or presented any evidence showing that 

 the rings on the vertebrae are age rings and annual 

 in nature. Partlo, however, hypothesized that 

 the rings are true year marks by the following: 

 (1) by assuming that the length groups in the 

 size frequencies are age groups, he found that 



there was acceptable agreement between the mean 

 length of fish assigned to each vertebral ring-class 

 and the mean length of corresponding length- 

 groups in the sample, and (2) there was agree- 

 ment between the lengths of young fish calculated 

 from vertebral measurements of older fish and the 

 observed lengths of young fish. For the latter, 

 Partlo, working with a subsample of 98 fish, found 

 that the observed lengths were consistently greater 

 than the calculated lengths because the fish were 

 captured some time after the last ring was formed. 

 Partlo concluded that "from the above it seems 

 clear that the third and subsequent vertebral rings 

 are produced annually. It is assumed, in the 

 absence of contrary evidence, that the inner two 

 rings are annual also." 



While Partlo 's results indicated that the verte- 

 bral rings were true year marks, it was felt that 

 more direct evidence would be desirable. One 

 possible approach to obtaining such evidence is 

 presented on page 357. 



In this study, ring counts were not reproducible 

 with consistency, due largely to the difficulty in 

 formulating suitable criteria for the identification 

 of the rings. Since it was demonstrated that there 

 was a linear relation between the vertebral radius 

 and the body length of the albacore (fig. 4), it was 

 difficult to make a reading without first forming 

 an opinion as to how many rings a certain-sized 

 vertebra should have. Also, in making ring meas- 

 urements the selection of a ring often depended on 

 whether or not its radius fell within a certain 

 range which encompassed all previous radii for 

 that ring. Any extreme deviation was inadvert- 

 ently questioned, and a subsequent remeasurement 

 was made on a more "reasonable" ring, which was 

 invariably found after closer scrutiny. For ex- 

 ample, if the investigator measures the radius of 

 the second ring at 5 mm. in a particular vertebra, 

 and if all previous second-ring radii have fallen 

 between 3 and 4 mm., he re-examines the vertebra 

 for a second ring which he may have "missed." 

 More often than not he will "find" the "missing" 

 ring. In fact, it was observed that "rings" could 

 be located at will almost anywhere on the centrum. 



Aikawa and Kato (1938) state that "where the 

 variation of the radius in comparison with other 

 rings is very marked, the line cannot be considered 

 an annual ring." Figueras (1955) found that in 

 40 percent of the specimens he examined, the first 

 annulus as well as the succeeding annuli were lo- 



