290 BULLETIN OF THE UNITED STATES FISH COMMISSION. 



By means of long-continued observations of fish kept separate in aqua- 

 riums, by trying different methods of feeding, and by chemical analyses 

 of the various articles of food, we shall undoubtedly obtain satisfactory 

 results. This is r : line of investigation which would repay a physiol- 

 ogist for engaging in, and this is a branch of fish-culture where science 

 can be employed to advantage. An attempt in this direction is the 

 standard carp food prepared by Carl Nicklas in his well-known Manual 

 of Pond-culture. This standard food is based too much on unreliable 

 suppositious and on conclusions as to the correctness of which we enter- 

 tain the gravest doubts ; as, for example, when, in order to determine the 

 necessary quantity of albumen, the hog is taken as a standard, because 

 it is likewise omnivorous, which quality alone does not entitle us to 

 draw a comparison between a mammal living on laud and a bsh living 

 in the water. 



Nicklas, however, deserves credit for having been the first to direct 

 attention to the necessity of a rational system of feeding fish, and to 

 lay down some standard rules therefor. By further and careful experi- 

 ments his errors will probably be corrected. Although Nicklas states 

 that by using fish-food prepared according to his rules he had obtained 

 results perfectly satisfactory to himself, and although he accompanies 

 his assertions by figures showing these results, he will doubtless agree 

 with us that by using a method of feeding based on scientific investiga- 

 tions the results jvould have been much greater. 



We desire to know what quantities of the different nutritive sub- 

 stances are needed, and what is the limit of food which should be ex 

 ceeded, if a larger and superior race of fish is to be gradually produced. 

 The proportion of nutritive substances composing fish -food should 

 be very accurate. As is the case with other animals, the needs of fish 

 change with age. During its early stages, when its framework of bones 

 grows very considerably during the course of one year, it needs a dif- 

 ferent food from that required at a later period. All this can only be 

 determined by means of exact scientific experiments. If we are pre- 

 pared to give the fish food which is suitable in every respect, there is a 

 reasonable prospect that we shall be successful in employing the second 

 means for reaching the object in view, namely, the careful selection of 

 breeders which are calculated to produce fish of superior flesh. In this 

 respect it should first be determined what shape is most favorable to the 

 development of flesh. In regard to this we prefer that framework of 

 bones which is shortest in proportion to the size of the body, and the 

 limited development of those parts which are worthless, such as the head, 

 since the food can be given in small particles, and the tail and fins, be- 

 cause a well cared-for pond fish need not be a particularly good swim- 

 mer. Care should also be taken not to lose the power of propagation 

 in the endeavors to form flesh and fat. We also desire fish of rapid 

 growth, so as to reach a good weight as soon as possible. The growth 

 of fish varies very much when left to nature, and is furthered by a suf- 



