LAKE HERRING OF GREEN BAY, LAKE MICHIGAN 



93 



Before the plastic-impression method was 

 adopted, careful microscopic comparisons were 

 made of the scales and their impressions to be 

 certain that replication was complete and without 

 distortion. Butler and Smith (1953) who studied 

 the reliability of scale impressions in age and 

 growth studies found that growth calculations 

 made from scale impressions did not differ signif- 

 icantly from those made from the scales them- 

 selves. About 500 of the scales used in this 

 study were mounted in a glycerin-gelatin medium ; 

 plastic impressions were made of the remaining 

 3,900. AH key scales used to establish the body- 

 scale relation were mounted in glycerin-gelatin. 



Scale measurements for growth computations 

 were made from the magnified (X41) scale image 

 projected on the screen of a microprojection de- 

 vice (described by Moffett 1952) and recorded 

 to the nearest millimeter. The scale to be meas- 

 ured was oriented so that a line on the viewing 

 "screen bisected the image at its greatest antero- 

 posterior diameter. Measurements of the total 

 diameter and of diameters of growth fields cir- 

 cumscribed by annuli were made along this line. 

 The total diameter was measured from the extreme 

 anterior to the extreme posterior margins of the 

 scale. Diameters of growth fields were measured 

 from the inside edge of the first complete circulus 

 outside the annulus. 



Scale measurements of each fish were entered 

 on IBM (International Business Machine) cards 

 along with coded information concerning each 

 fish. All subsequent computations and tabula- 

 tions were made by means of the 602A IBM 

 calculator and the 404 IBM tabulator at the 

 Statistical Research Laboratory of the University 

 of Michigan. 



Ages were determined by counting the annuli 

 or year-marks on the scales. Van Oosten (1929) 

 clearly established the validity of this method 

 for the age determination of the lake herring of 

 Saginaw Bay. More recent authors reporting on 

 this species (Carlander 1945; Cooper 1937; Fry 

 1937; Hile 1931, 1936; Pritchard 1931 ; Stone 1938; 

 and others) have accepted the use of scale mark- 

 uigs for age analysis of lake herring. 



Nothing in the data on the Green Bay lake 

 herring gives cause to question the validity of 

 scales for age determination. Nevertheless, cer- 

 tain difficulties of interpretation were encountered. 

 Accessory checks, or false annuli, occurred on 



scales of nearly all fish after the second year of 

 life. The general appearance of these checks and 

 their location with respect to the annuli on either 

 side left little doubt as to their identity; however, 

 the possibility of some errors of age determination 

 cannot be discounted. 



The regular appearance of accessory checks is 

 not confined to the Green Bay stock. These 

 false annuli on cisco scales have been reported 

 by HUe (1936) in the cisco of Muskellunge Lake 

 and by Fry (1937) in Lake Nipissing. Bauch 

 (1949) described a fast-slow-fast growth pattern 

 in a population of "kleinen Marane," Coregonus 

 albula L. (the European coregonid most similar 

 to the lake herring), in Mochelsee. He attributed 

 the midseason check in the scales to oxygen 

 depletion and an accumulation of hydrogen sulfide 

 in the hypolimnion which forced these fish, nor- 

 mally inhabitants of the deeper waters in summer, 

 to live in upper strata where less favorable tem- 

 perature conditions exist. Data on the Green 

 Bay herring are inadequate to show the cause of 

 accessory checks or even the time of their forma- 

 tion. Seemingly the formation of checks varies 

 from fish to fish and possibly' according to season 

 and locality. 



The characteristics of the annulus on the scales 

 of Green Bay lake herring are similar to those 

 described for scales in other populations. The 

 circular ridges on the scale start forming on the 

 anterior margin of the scale and grow posteriorly 

 along the lateral fields. When growth stops com- 

 pletely and resumes again, growth of the un- 

 finished circuli is not completed; instead a new 

 circulus is started which encompasses the ends of 

 those left incomplete at the cessation of growth. 



Fish having scales without an annulus are 

 designated as belonging to age group 0, those 

 with one annulus to age group I, * * *. For 

 convenience, each fish is held to pass into the next 

 higher age group on January 1. Since annulus 

 formation does not actually take place until 

 spring or early summer, the convention requires 

 that a "virtual" annulus be credited at the edge of 

 the scale from January 1 until the new annulus is 

 visible. Year classes are identified by year of 

 hatching (spring) rather than year of egg deposi- 

 tion (fall). Thus, it is always possible to de 

 termine the year class of a fish by subtracting its 

 age from the year of capture; for example, a fish 



