Great Lakes. Trans. Am. Fish. Soc. 86:97-111. 

 1960. Systematics and biology of the gizzard shad {Dorosoma 



cepedianum) and related fishes. U.S. Fish Wiidl. Serv., Fish 



Bull. 60:371-392. 

 WHITWORTH, W. R., p. MlNTA, AND R. Orciari. 



1980. Further additions to, and notes on, the freshwater ich- 



thyofauna of Connecticut. In P. M. Jacobson (editor), 



Studies of the ichthyofauna of Connecticut, p. 27-29. Storrs 



Agricultural Experiment Station, University of Connecticut, 



Storrs, Bull. 457. 



John O'Leary 



Department of Forestry and Wildlife Management 

 University of Massachusetts 

 Amherst, MA OlOOS 



Douglas G. Smith 



Museum of Zoology 

 University of Massachusetts 

 Amherst, MA 01003-0027 



RELATIONSHIP OF OTOLITH LENGTH TO 



TOTAL LENGTH IN ROCKFISHES FROM 

 NORTHERN AND CENTRAL CALIFORNIA 



Knowing the relationship between otolith length and 

 total length of a fish is useful for two reasons: 1) 

 Fish size can be estimated from otolith lengths 

 measured from otoliths encountered in predator 

 stomachs, in core samples, archaeological sites, etc., 

 and 2) the length of a fish can be verified when the 

 age determined from the otolith lies outside ex- 

 pected values. 



The otolith/total length relationship is useful in 

 predator-prey and archeological studies if fish size 

 can be extrapolated from otolith length. Otoliths are 

 often the only part of a prey fish remaining in a 

 predator's gut (Ainley et al. 1981; Treacy and 

 Crawford 1981) or at cooking sites of archeological 

 middens (Fitch 1972). Fish lengths could be esti- 

 mated from otoliths found as remains of prey or in 

 coastal archaeological excavations (Fitch and 

 Brownell 1968). Existing keys (e.g., Morrow 1979) 

 allow identification of fish species from otoliths. 

 With these keys, personal reference collections, and 

 the length relationships described in this paper, in- 

 vestigators will be able to verify species and size 

 data collected in field sampling, and obtain more 

 complete knowledge of prey species of marine mam- 

 mals, birds, and fishes. 



Large-scale surveys, such as the California coop- 

 erative survey (Sen 1984) that samples commercial 



rockfish landings in northern California, are prone 

 to errors at several levels. Problems that may be 

 encountered in collecting otoliths and measuring fish 

 lengths include errors in recording lengths and the 

 mixing up of otoliths. Some errors can be corrected 

 by measuring the otolith and estimating the size of 

 the fish it came from. Every effort should be made 

 to eliminate erroneous data from the database 

 before curves are constructed or cohort analysis is 

 performed. 



In this paper, I report the results of my investi- 

 gation of the relationship between otolith length and 

 total length for 30 rockfish species of the genus 

 Sebastes. Linear regression statistics are presented 

 for all fish of the species encountered. 



Methods 



Specimens were collected during a life history 

 study on the rockfishes of northern and central 

 California conducted at the Southwest Fisheries 

 Center Tiburon Laboratory. Fish were sampled 

 from the commercial trawl fishery, the commercial 

 sport fishery, skiffs, and research cruises from 1977 

 to 1980. Specimens were identified to species, and 

 then total lengths of frozen— then thawed— car- 

 casses were measured on a meter board in milli- 

 meters (mm). Otoliths were measured to the near- 

 est 0.1 mm with an ocular micrometer. The greatest 

 length of the otolith was measured from the ante- 

 rior tip to the most posterior projection (Kimura et 

 al. 1979) (Fig. 1) as if the otolith were flat, without 

 compensating for the curvature. Linear regressions 

 were run on total length (y) versus otolith length 

 (x) for 30 rockfish species. Outliers (±3.0 standard 

 deviations) from the line were assumed to result 

 from measurement or recording errors and were 

 discarded (2% of the observations). 



Figure 1.— The length of an otolith is measured from the anterior 

 tip to the posterior projection. 



FISHERY BULLETIN: VOL. 85, NO. 2. 1987. 



383 



