FISHERY BULLETIN: VOL. 70, NO. 3 



Table 3. — Back-calculated length of sardines sampled on two dates. The two methods 

 of back-calculating length are discussed in detail in the text; in general, method 1 is 

 by direct proportion, and method 2 is by direct proportion with a correction factor for 

 the body length-scale radius relation. 



DISCUSSION 

 MARK FORMATION ON SCALES 



Walford and Mosher (1943) indicated that on 

 sardine scales an accessory mark was disting- 

 uishable from an annulus by its finer sculpturing 

 and its rare occurrence on all scales of an indi- 

 vidual. The results of our study on sardines 

 reared in the laboratory showed that accessory 

 marks occurred on all scales examined after the 

 fourth month, and they were generally indisting- 

 uishable from annuli. But for some scales in 

 which the accessory mark was distinguishable 

 from an annulus, the identifiable characteristic 

 was the fine sculpturing mentioned by Walford 

 and Mosher. 



Interruptions in the growth pattern are gener- 

 ally assumed to form marks on scales (e.g., see 

 Van Oosten, 1957). It is also widely assumed 

 that the driving mechanism behind mark 

 formation is temperature, through the influ- 

 ence of a fish's metabolism (Brown, 1957). 

 Our results indicate that mark formation is un- 

 related to temperature, but appears to be related 

 to growth rate. Hogman (1968) obtained some- 

 what similar results in his experiments with 

 coregonids. He found that formation of marks 

 was closely related to growth and somewhat re- 

 lated to temperature. However, Hogman indi- 

 cated that light period may be the primary driv- 

 ing mechanism. Bilton and Robins (1971) found 

 that mark formation was correlated with in- 

 crease in food supply, but not with resumption of 

 feeding after starvation in sockeye salmon. 

 Their experiments with light period proved in- 

 conclusive. It thus appears that mark forma- 

 tion in fishes is probably related to growth, al- 



though the actual driving mechanism (s) have 

 so far not been clearly identified. 



ERROR IN AGING 



Scales are routinely used to age and study 

 growth of the Pacific sardine. Our results indi- 

 cate that extreme caution must be exercised in 

 aging because of the presence of accessory marks 

 on scales. Furthermore, since the annulus is 

 formed during the winter, the actual age of a 

 fish at time of first annulus formation may vary 

 depending on its date of hatching. In our study 

 the first annulus was laid down after the sixth 

 month for fish hatched in May. May is the mid- 

 dle of the heavy spawning season for the Pacific 

 sardine, but the season extends from March to 

 October (Kramer and Smith, 1971). 



Kimura (1970)° conducted an experiment to 

 test the consistency of early and recent scale 

 readers in aging sardines from scales. He found 

 that the abrupt increase in Li, as reportedly ob- 

 served in the 1940's by Marr (1960) for sardines 

 landed at San Pedro, may have actually been 

 caused by a change in scale readers and in criter- 

 ia used in reading scales. We compared our av- 

 erage length at annulus formation with the av- 

 erage Li of fish aged by the early (Li = 101.3 

 mm) and recent (Li = 131.5 mm) scale readers, 

 and discovered that the early readers probably 

 aged sardines correctly, whereas the recent read- 

 ers probably underestimated the age of age II 



^ Kimura, M. 1970. Possible errors in locating the 

 first scale annulus and in estimating the length of Pacific 

 sardines. Manuscript filed at National Marine Fisheries 

 Service, Southwest Fisheries Center, La Jolla, CA 92037. 



1050 



