TABLE 2. — Food habits of Lake Michigan smelt shorter than 180 mm total length. Upper values for food 

 organisms represent dry weight and values in parentheses are the percentage numbers. 



i„ D r 3 n» an n Food organisms 



TABLE 3. — Food habits of Lake Michigan smelt 180 mm total 

 length and longer. Upper values for food organisms represent 

 percent dry weight and values in parentheses are the percentage 

 numbers. 



these numbers never exceeded 1% dry weight of 

 the diet. Alewives were consumed most frequently 

 during the June collection when yearlings com- 

 posed 58% dry weight of the diet. Small alewives 

 constituted 30 and 10% dry weight of the diet 

 during August and October, respectively. 



Food eaten by smelt 180 mm and longer con- 

 sisted principally of small alewives and Mysis, but 

 included small numbers of Pontoporeia and occa- 

 sionally fingernail clams (Table 3). Yearling ale- 

 wives represented 94 and 99% dry weight of the 

 diet during the May and June collections, respec- 

 tively. 



Smelt examined for feeding periodicity aver- 

 aged 158 mm total length. Weight of stomach con- 

 tents differed statistically over the 24-h period 

 during the October collection (F = 9.99, P^O.001, 

 df = 5, 82). Stomachs contained the most food 

 (1.5% body weight) at 2430 h and decreased to 

 0.2% by 0400 h (Figure 1). In addition, the occur- 



1200 



1600 



2000 



2400 



0400 



0800 



TIME OF DAY 



1. — Feeding periodicity of Lake Michigan smelt col- 

 lected on 13-14 October 1973. Dry weights of stomach contents 

 are expressed as a percentage of dry body weight. Vertical lines 

 represent ±2 SE of the mean and the horizontal black bar the 

 hours of darkness. The number of empty stomachs and number of 

 stomachs examined are given near each average. 



rence of empty stomachs was dependent upon time 

 of day (x 2 = 31.51, P^O.001, df = 5). Only 1 out of 

 45 stomachs was empty in the collections between 

 1600 and 2430 h. In contrast, 23 out of 43 stomachs 

 were empty between 0400 and 1200 h (Figure 1). 

 The March collection showed no significant differ- 

 ences in weight of stomach contents over a 24-h 

 period. 



Discussion 



Smelt examined in this study were piscivorous, 

 consuming young-of-the-year and yearling ale- 

 wives. Food habit studies of smelt in Saginaw Bay, 

 Lake Huron (Gordon 1961) and Lake Erie (Price 

 1963) have shown that smelt consume fishes, but 

 not alewives. Smith (1970) hypothesized that dif- 



638 



