Effects of Formalin and Alcohol Preservation on Lengths and Weights of Juvenile Sockeye Salmon

Abstract

Preserving fish can induce changes in fish tissue and structure that affect subsequent measurements of length and weight. We conducted 3 different experiments to quantify these changes and allow correction factors to be used as needed in analyses of fish growth and production. The experiments were designed to (1) determine temporal variation in the length and weight of juvenile sockeye salmon Oncorhynchus nerka as a result of storage in either 10% neutral freshwater-buffered formalin or 95% ethanol; (2) describe the amount of observed variation in length and weight measurements that can be attributed to reader variability; and (3) formulate conversion equations that allow for back-calculation to fresh lengths and weights. Fish preserved in alcohol for 70 d revealed a mean weight loss of 0.68 g after 16 d (19.7% of mean fresh weight; P 0.05). Length initially decreased by 1.41 mm after 16 d (2.19% of mean fresh weight; P < 0.0001), remained stable through 42 d, and then increased significantly (P < 0.0001) by day 70. Storage of fish in formalin produced various results. One experiment revealed a nonsignificant change in length (P = 0.114) after 70 d in the preservative; weight increased 0.24 g (7.12%; P < 0.0001) after 16 d of preservation, but then stabilized through 70 d. Additional formalin effects ranged from a minor length loss (P < 0.0001) and a substantial weight gain (P < 0.0001) after 106 d of storage to a significant (P < 0.001) loss in length after 30 d and continued loss through 99 d. Although a weight gain occurred after 30 d of preservation (P < 0.0001), no significant difference from fresh weight was found after 99 d. We examined the amount of variation in these results that may be attributed to reader differences. For the formalin group, maximum differences between readers were 3.5% (P < 0.0001) for lengths and 6.5% (P < 0.0001) for weights. For the alcohol group, differences were 1.9% (P < 0.0001) for lengths and 6.4% (P < 0.0001) for weights. Instructing readers on standardized blotting methods resulted in a maximum difference between readers of only 2.5% for weight measurements. Conversion equations that allow for back-calculation to original live lengths and weights were developed. 81 Authors: PATRICK A. SHIELDS is a fishery biologist and STAN R. CARLSON is a biometrician with the Alaska Department of Fish and Game, Commercial Fisheries Management and Development Division, Limnology Unit, 34828 Kalifornsky Beach Road, Suite B, Soldotna, AK 99669-8367. Acknowledgments: Alaska Department of Fish and Game personnel John Edmundson, Jennifer Brannen, Bill Glick, Morris Lambdin, and Dave Westerman — participated in collecting length and weight data. Gary Kyle — provided the incentive to conduct this work and reviewed an early version of the manuscript. Alaska Fishery Research Bulletin 3(2):81–93. 1996. Copyright © 1996 by the Alaska Department of Fish and Game