16 VICTOR I. SHELFORD. 



is based on the general impression which ha- been acquired in 

 taking classes to these and other ponds of similar character 

 several times per year during six years. Secondly, by taking 

 the time required to make a representative collection from the 

 different ponds. On the basis of this experience, the figures given 

 in the table are thought to be very conservative. That there is 

 a far greater number of animals and a greater volume of animal 

 substance in the old ponds is very easily demonstrated to any 

 one by inspection. 



TABLE XX. 



SHOWING AN ESTIMATE OF THE RELATIVE NUMBERS OF THE CHIEF ITEMS OF FISH 



FOOD IN THE DIFFERENT PONDS. 



Previous to being drained pond 140 should be rated at 70 for fishes. 



While the results here presented are not such as to justify 

 conclusions concerning details, we may state that the amount of 

 life per unit volume unquestionably increases as the ponds grow 

 older, at least up to stages like 146. Qualitative differences are 

 shown in the Tables VII. and XIV., and the total number of 

 species recorded in each pond is about the same, tin- actual 

 quantity is far greater in the older. 



IM. THE CAUSES OF SUCCESSION 01 FISH. 



A discussion of succession must be made 1 with reference to all 

 the organisms of the habitat, or at least a large number of them 

 considered in mass. Succession of one group of organisms t.ikin^ 

 place without the succession of others in the ^une environment 

 seems improbable. A discussion with reference to tish UHIM 

 take other organisms into consideration. 



i. Statement of the Problem. A rlrar understanding of the 

 problem at hand will perhaps be facilitated bv a careful Mate- 



