Natural Mortality 203 



information on numerical changes in a fish population. This does not 

 minimize the value of underwater observations for many other purposes. 



A comprehensive measurement of fish mortality, however, requires 

 either a comj)lete inventory of a fish population at specific intervals 

 (which is usually impossible) or a mathematical approach, either where 

 returns from marked fishes over several seasons are employed to estimate 

 natural losses for the entire population during that time, or where num- 

 bers of fish caught (separated into age classes) are used with data on 

 eflFective eflFort, to estimate natural mortalities.'*^ A dependable creel census 

 is essential for furnishing information on fishing mortality. 



If successive annual broods of young of a given kind of fish were nearly 

 the same size, a comparison of the numerical sizes of all of the year classes 

 present in a lake in any year would give accurate information on total 

 annual mortality over the life span of that species. Thus a consideration 

 of the relative abundance of successive year classes of a species in any 

 mixed population may show total mortality, although it may show little 

 more than length of life of that species in the lake in question. 



If fish could be marked in sufficient numbers, a measure of the returns 

 from fishing for these marked fish over a period of several years would 

 give an estimate of natural mortality. This method was developed by 

 Ricker ^^ who investigated the mortality rates of bluegills in several 

 Indiana lakes, by fin marking these fish prior to the opening of the fishing 

 season (June 16) in two or more successive years. Catch records of 

 marked fish over a period of two or more years furnished data on rate 

 of exploitation. The relationship between the number of fish captured 

 and marked the first year, and recaptured in the first and second years, 

 allowed calculations of total mortality from which angling mortality could 

 be subtracted to give natural mortality. Ricker also discussed the "in- 

 direct" method of determining total mortality through a consideration of 

 the relative numerical abundance of successive year classes, and he 

 describes the weaknesses of such a system. 



His estimates of total mortality for bluegills in three Indiana lakes 

 ranged from 60 to 77 per cent per year. Of this range, fishing accounted 

 for 19 to 36 per cent, leaving 40 per cent to about 50 per cent for natural 

 mortality. Ricker's calculations for total annual mortality, rate of ex- 

 ploitation, and natural mortality (two methods of calculation) are shown 

 in Table 7.2. 



It is perhaps presumptuous to separate fishing mortality and natural 

 mortality. Certainly, some of the fish that fall prey to anglers might die 

 during the same period from natural causes, and fish might be caught by 

 anglers if they had not previously died from natural causes. The mortality 

 rate could be as high with no fishing as with the exploitation rates shown 

 in Table 7.2. However, it probably would not be so high because fish 



