366 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



assistance in the collection of additional mate- 

 rials, and granted access to all facilities necessary 

 to the proper conduct of my research. 



John E. Bardach, Chairman of my doctoral 

 committee, planned and supervised my schedule 

 of graduate training; Dr. Ralph Hile directed 

 my research and the preparation of this disserta- 

 tion; other members of my committee gave most 

 useful advice and assistance. 



Various staff members of the Fish and Wild- 

 life Service were helpful in the field and labora- 

 tory. Leonard S. Joeris in particular assisted 

 greatly when I was making a start on the prepa- 

 ration and examination of scales. Howard J. 

 Buettner transferred various kinds of data to 

 punch cards and prepared tabulations of length 

 frequencies and length-weight records. All col- 

 lections before the spring of 1955 were made by 

 Service employees. 



Cecil C. Craig, Director, Statistical Research 

 Laboratory, University of Michigan, advised in 

 problems of multiple correlation and regression 

 analysis. Glenn W. Graves of the laboratory 

 staff programmed the materials for IBM proc- 

 essing. 



Reeve M. Bailey, Curator, Fish Division, Mu- 

 seum of Zoology, University of Michigan, sup- 

 plied small Saginaw Bay perch from the Museum 

 collections for the study of the body-scale re- 

 lation. 



Field work was greatly furthered by the co- 

 operation of: A. J. Neering, Michigan Depart- 

 ment of Conservation; Henry Engelhard, Bay 

 Port Fish Co.; and John Gillingham, R. L. 

 Gillingham Fishing Co. 



MATERIALS AND METHODS 



Collection of Samples 



The study of the Saginaw Bay yellow perch 

 population has been based on the determination 

 of age and the calculation of the growth histories 

 of 4,285 fish, 3,407 of them collected during the 

 spawning seasons of 1943-55 (no fish were col- 

 lected in 1945 and 1952). The term spawning- 

 run sample has a degree of elasticity in its 

 application to the 1955 collections; some tabula- 

 tions included the samples of April 18, May 18, 

 and June 7, whereas others included only the 

 sample of May 18, the one taken nearest the 

 height of the spawning season. The samples 

 through 1954 were gathered by the U. S. Fish 



and Wildlife Service as part of a continuing 

 study of the Saginaw Bay fisheries. In addition 

 to the spawning-run samples, the 1955 collections 

 included data from months outside the spawning 

 period (table 1). All fish were caught by com- 

 mercial trap nets in the Bay Port area. The 

 actual location of the trap nets ranged from 

 Fish Point to Charity Island (fig. 1), but most 

 lifts were nearer Bay Port. Because no evidence 

 of subpopulations was found, the samples all are 

 considered to have been drawn from the same 

 general stock. 



Table 1. — Collections of scales from Saginaw Bay yellow 

 perch 



Records for Individual Fish 



The total length of fish (from tip of the head 

 to tip of the tail, with lobes compressed to give 

 the maximum measurement) was determined by 

 a measuring board to the nearest 0.1 inch. 



Weights were recorded either to the nearest 

 gram or to the nearest 0.1 ounce depending on 

 the kind of balance. Two types of balances were 

 used : A dietary platform balance calibrated by 

 2-gram intervals (weight was estimated to the 

 nearest gram) ; a spring balance calibrated by 

 0.2-ounce intervals (weight recorded to the near- 

 est 0.1 ounce). 



The sex and state of gonads were recorded for 

 all fish except the collections of 1950 and June 7, 

 1955, for which information is available only on 

 sex. 



All samples of the 1943-55 period for which 

 length and weight were recorded (4,285 fish) 

 were used in the study of the length-weight 

 relation. 



Determination of Age 



Preparation and examination of scales 



Scales for age and growth studies were col- 

 lected from all fish (except the 1954 collection) 

 from below the lateral line on the left side. For 

 the 1954 sample, scales were taken from above 

 the lateral line. Because of this inconsistency as 



