GENERAL PROPERTIES OF POPULATIONS 



267 



1. Total count of all individuals of all 

 stages or classes; 



2. Total count of all individuals of a cer- 

 tain stage or class; 



3. Determination of population size by 

 the registration method; 



4. Sampling methods (general state- 

 ment); 



5. The method of marking; 



6. Indirect methods; and 



7. Combination of several methods. 



These will be considered in order. 



1. Total count of all individuals of all 

 stages or classes. The total count method 

 gives the only precisely accurate census. 

 For any particular spatial unit at any par- 

 ticular moment of time this technique pro- 

 vides a perfect numerical picture of popu- 

 lation size. Actually, this is a desideratum 

 rarely attained except in a few laboratory 

 population studies. The total census is used 

 for human populations and is there sub- 

 ject to certain obvious errors. These errors 

 are so minor, however, for most civiUzed 

 societies that, for all practical purposes, they 

 can be ignored. Pearl concluded that of the 

 2,069,094,126 persons in the world in 

 1932," 97 per cent, or 2,029,608,900 per- 

 sons, were either counted or computed by 

 extrapolation, while 3 per cent, or 39,485,- 

 226, were estimated purely by guessing. 

 The ecological population student owes a 

 debt to the human demographer for setting 

 up a creditable standard in total census 

 methods and demonstrating the importance 

 of such statistics. The first wide-scale cen- 

 sus was started in 1666 by Canada and 

 adopted by nineteen other countries be- 

 tween that date and 1897. 



2. Total count of all individuals of a cer- 

 tain stage or class. Stage or class counts are 

 used in population studies either because 

 they are preferable or because the total 

 count is impracticable. It is often more 

 meaningful to enumerate a certain compo- 

 nent group of a population than to enumer- 

 ate the population as a whole. Thus, in hu- 

 man populations the group can be dissected 

 into such smaller categories as sex, race, 

 age classes, persons exposed to the risk of 

 contracting influenza, and so on. Likewise, 

 this treatment is valuable in natural and 

 experimental populations. In the salmon, 



• This figure was derived largely by tabulat- 

 ing census reports given in the "Statesmans' 

 Year Book." 



for example, one of the most meaningful 

 statistics is the age-class distribution of the 

 population. In a Tribolium (flour beetle) 

 culture it is usually desirable to know, in 

 addition to the total population size, the 

 number of eggs, larvae, pupae, and ima- 

 goes comprising that population. When 

 data such as these are available, one stage 

 can be studied relative to another, fre- 

 quently to the profit of the analysis. On the 

 other hand, there are times when a total 

 count cannot be made. Then a count of a 

 specific stage or class is substituted. Much 

 of the population work with Drosophila me- 

 lanogaster is based on the imagoes and 

 eggs only; the larvae are particularly hard 

 to count. 



3. Determination of population size by 

 the registration method. The registration 

 method is a theoretically sound, but practi- 

 cally unimportant, method. It requires that, 

 after an initial census has been taken, each 

 birth, death, immigration, and emigration 

 that occurs in a specified population shall 

 be recorded for a stated time interval. These 

 registration data are then treated as follows. 



Registration Summary or X = ( Births -f im- 

 migrations ) — ( Deaths + emigrations ) 



Population size at the time specified then is 

 computed: 



Population size =: Initial size ± X 



This method is rarely used because it is 

 frequently difiicult to get the observations 

 on dispersion, and in many cases it is ob- 

 viously easier to census the population a 

 second time than to compute its size by the 

 registration formula. This method is valu- 

 able, however, in that it systematically 

 places on record many basic data for the 

 observed population. 



4. Sampling methods (general state- 

 ment). The determination of population 

 size by sampling presents these aspects: 

 (a) Commonly this is the only method that 



can be used; (b) for certain populations 

 this method works well when intelligently 

 applied; (c) there are numerous examples 

 in the literature in which sampHng has been 

 inadequately, even fooHshly, appHed; and 

 (d) the investigator must have some knowl- 

 edge as to what constitutes a suitable 

 sample, must know whether the census 

 samples are drawn from a distribution that 



