FOOD HABITS AND FEEDING CHRONOLOGY 



OF RAINBOW SMELT, OSMERUS MORDAX, 



IN LAKE MICHIGAN' 



Rainbow smelt, Osmerus mordax Mitchill, in 

 Lake Michigan originated from a planting in 

 Crystal Lake, Mich., in 1912 (Van Oosten 1937). 

 Since its introduction in Lake Michigan, the smelt 

 has become abundant, serving as forage for larger, 

 predatory species (Wright 1968; Harney and Nor- 

 den 1972) and sustaining a small seasonal sport 

 and commercial fishery. There has been consider- 

 able controversy regarding the smelt's role as a 

 piscivore. Food studies of smelt in Saginaw Bay, 

 Lake Huron (Gordon 1961) and Lake Erie (Price 

 1963) have shown that smelt consumes fishes, but 

 not the alewife, Alosa pseudoharengus . Recently, 

 a fall collection of smelt revealed that it consumes 

 young-of-the-year alewives (O'Gorman 1974). 



The food habits of Lake Michigan smelt have 

 not been studied on a seasonal basis and little 

 information exists concerning its food habits dur- 

 ing the winter months. Also, feeding chronology 

 has never been considered. The purpose of this 

 study was to examine food habits of smelt during 6 

 mo representing all four seasons and to consider 

 feeding chronology during two representative 

 months. 



Materials and Methods 



Rainbow smelt were collected along the western 

 shore of Lake Michigan by gillnetting or trawling 

 on six dates between March 1973 and June 

 1974 (Table 1). Gill nets were placed on the bottom 

 overnight and 45-min trawl hauls performed at 

 4-h intervals over a 24-h period. Gill nets were set 

 and retrieved at 4- to 6-h intervals over a 24-h 

 period on 13 October 1973 in order to examine 

 feeding chronology. Smelt were collected at differ- 

 ent depths during the course of the study because 

 of their seasonal inshore-offshore movements. 

 Fish were frozen shortly after capture. 



Feeding chronology was examined on 23 March 

 and 13 October 1973. Stomachs of smelt used in 

 this portion of the study were dissected out and the 

 contents removed. Fish and stomach contents 

 were dried for 48 h at 60°C and weighed to the 



TABLE 1. — Dates, locations, depths, and methods of capture of 

 Lake Michigan smelt examined in this study. 



'This research was supported hy the University of Wisconsin 

 Sea Grant Program. Contribution No. 154, Center for Great 

 Lakes Studies, University of Wisconsin-Milwaukee, Milwaukee, 



Wis. 



'All collections made on the bottom 

 2 Feeding chronology examined 



nearest milligram. Dried stomach contents were 

 expressed as a percentage of dry body weight. The 

 significance of time of day on the amount of food in 

 stomachs was ascertained with analysis of var- 

 iance (ANOVA). Means and the ANOVA were 

 calculated from arcsine transformed data (Sokal 

 and Rohlf 1969). A chi-square contingency test 

 was used to ascertain the significance of time of 

 day on the occurrence of empty stomachs. Sig- 

 nificance testing was performed at the 0.05 error 

 level. 



Separate smelt were examined for food habits. 

 These fish were measured to the nearest millime- 

 ter in length. Stomachs were removed, contents of 

 each stomach were placed in a Petri dish with 

 water, and the organisms enumerated. Food 

 habits were defined in terms of percentage num- 

 bers and percent dry weight of stomach contents 

 (Wells and Beeton 1963). Dry weight indices used 

 were fish, 176; Mysis, 3; Pontoporeia, 1; fingernail 

 clam, 1; Tendipedidae, 0.4; and Cladocera- 

 Copepoda, 0.003 (Morsell and Norden 1968). 



Results 



Stomachs of 515 smelt were examined. Food of 

 smelt included Mysis; Pontoporeia; alewives 

 (young-of-the-year and yearlings); and to a lesser 

 extent, fingernail clams; Tendipedidae pupae; 

 cladocerans; and copepods (Table 2, 3). A marked 

 increase in piscivorous food habits was observed in 

 smelt longer than 180 mm. For this reason, smelt 

 were divided into two size groups. 



Smelt shorter than 180 mm consumed primarily 

 Mysis during October, February, and March (Ta- 

 ble 2). Smelt were found in shallower water during 

 May, June, and August and their stomachs con- 

 tained yearling alewives, Pontoporeia, and Ten- 

 dipedidae. Pontoporeia were consumed most fre- 

 quently during August, when they represented 

 35% dry weight of the diet. Tendipeds represented 

 25, 6, and 2 percentage numbers of the diet during 

 May, June, and August, respectively. However, 



637 



