FOOD OF BULLHEADS 33 



and preserve them for examination later. They were sealed in glass 

 fruit jars with data consisting of the date and method of catching, 

 depth of water, or other pertinent information. The fingerlings 

 were preserved whole. This material was placed at the disposal of 

 the writer in September. 



In the laboratory the stomachs and intestines were opened sepa- 

 rately and the contents spread out in watch glasses containing water. 

 Examination was made under a binocular microscope, and the com- 

 pound microscope was used for the examination of very small organ- 

 isms. The contents were identified as nearly as possible and the 

 specimens of each kind present were counted and the number 

 recorded. Filamentous algse, Copepoda, and Cladocera were meas- 

 ured in cubic centimeters by dropping them into small-bore glass 

 graduates containing water and allowing them to settle to the bottom, 

 which reciuired from 30 to 45 minutes. By actual count it was found 

 that the average number of Daphnia longispina was 1,000 and of the 

 smaller Clad#)cera 1,750 per cubic centimeter, and there were 1,312 

 copepods per cubic centimeter. Plant stems were measured in 

 centimeters of length. Larger crustaceans, mollusks, insect larvae and 

 pups', etc., were counted as they were lifted from one glass to another. 



DESCRIPTION OF FOODS 



To obtain accurate knowledge of the food eaten b^^ the bullheads 

 from the lakes mentioned earlier in this paper, the alimentary tracts 

 of 106 specimens from these lakes were examined. The data in the 

 following tables are arranged according to the lakes investigated, 

 which have been numbered consecutively with reference to the dates 

 on which they were visited. These numbers correspond to the num- 

 bers of the lakes on the accompanying map. The number following 

 the date of the collection indicates the length of the bullheads in 

 that collection, measured from the tip of the snout to the posterior 

 end of the caudal fin. 



The tables have been compiled to show the approximate per cent 

 in actual volume occupied by the various organisms comprising the 

 food of the specimens examined. Bait, blood cells,^ and parasites 

 have not been included in the tables because they are not food in 

 the sense in which the term is here used, though they were found in 

 the alimentary tract and record was made of them wherever found. 

 Unrecognizable fragments, which in most cases were well-digested 

 bits of food of the same kinds as that which was still in a condition 

 to be identified, also are omitted. 



The per cent volume was obtained by using the Cladocera as a 

 basis of computation. The number of Cladocera required to fill 1 

 cubic centimeter was learned by actual count. Each organism was 

 assigned a number representing the number of Cladocera needed to 

 make an equal mass volume. This number was then multiplied by 

 the number of organisms of the same species and size. All num- 

 bers for organisms from the same fish were added together and the 

 per cent volume of each organism found by dividing its number by 

 the sum total for all the organisms found in the alimentary tract 



3 These cells would not be found in the alimentary tract under normal conditions. When the stomach 

 and the intestine were removed from the specimen the blood escaped into both ends of the digestive tube. 



106460—28 2 



