With the rise in teii?)erature of the water and the growth of fish, the food demands of 

 the fish increased more and more, as did the inroads made by predatory aquatics. The results 

 are that the numerical maximum of Chironomidae is already reached between the middle of May 

 and the middle of July. 



The more fish we have upon a certain unit of surface the sooner this maximum will be 

 reached and at a so much lower point (pcxid 6). 



Equally decisive factors causing the vary early attainment of a maximum are the pupation 

 and hatching v;hich proceeds with a certain suddenness with the entry of warm May days. Of 

 course, the weijht quantity of the Chironomidae larvae, which alone is sufficient for the 

 nutrition of the fishes, rises numerically beyond the peak and advances until July, as the 

 remaining larvae, due to warn weather, mature rapidly, before the progeny of freshly hatched 

 midges again becomes noticeable in the pond. 



At the beginning of August, though, a very marked slide toward a minimum of Chironomidae 

 larvae — in nvimber as well as in weight — becomes apparent. P. Schiemenz (1931) confirms this 

 upon the strength of his own experiments. 



The fishbreeder, for pi^actical reasons, must prepare to counter balance this lossj he 

 will now resort to artificial feeding and also will prepare the developanent of large Cladocerae 

 through rational fertilization of his breeding pcxids. 



The number of Chironomidae larvae now remains more or less ccaistant until September, i.e. 

 upon a minioal scale. Increase and decrease are obviously equalized. 



The grown-up fish now consume all, up to an always present unusable remainder. But, 

 this remainder of not consumed aquatics is relatively small in coo^jarison with the early 

 summer months, due to a far greater voracity on the part of the fish, during the fall. 



If the ponds are not all too greatly overstocked, a slight rise in the number of larvae 

 and a somewhat greater rise in their weight occurs again by the end of September and the 

 beginning of October. 



The decrease, due to pupation and consumption now becomes less on account of the falling 

 tascperature . But a corresponding almap in the increase must also be taken into account. 



It has just been mentioned that the non-consumed remainder of aquatics is Inversely 

 proportional to the amount of fish stock, kept in the pond at different tines. 



'iYe notice a more rapid decrease in food-providing aquatics in pond 6 (Fig. 6), planted 

 sixteen times above normal, than in the other and less densely planted ponds. 



The part of aquatics, consvmied in pond 6 was greater, while the nonconsumed remainder 

 is smaller. This means that a larger stock of fish will exhaust the available stock of 

 aquatics to a greater extent, in other words the consunption quotient, the evaluation 

 factor or the food detection factor really rises in proportion to the amount of fish stock 

 present. 



Watching a feeding pigeon, we notice that it picks up a grain, here and there. It 

 does not continuously consume the grains adjacent to each other. The same applies to fish 

 which go after their food in a similar way, picking up £md gulping down a morsel of food 

 here and there. Consequently, the greater the number of fish feeding in such manner, the 

 amaller will be the nianber of naidetected morsels. 



But, at the same time, we learn from the chart (Jig. 6) that a complete cleaning up of 

 available larvae does not result even from overstocking (16, 2 year carps per 100 sq. m.), 

 not even at the time vihen further increase of this kind of "provender" is out of the 

 question . 



Although this was presumed by many authors, experiences have proven otherwise, just as 

 P. Schiemenz has maintained all along. 



38 



