METHODS 173 



(9) The amounts eaten per day are of great interest, e.g. 

 counts of the number of animals brought to its nest by a bird 

 in a given time. 



(10) Finally, the numbers of two species will often give 

 a clue to the fact that one is feeding on the other, e.g. 

 birds attracted by an unusual abundance of caterpillars on 

 oaks. 



13. There are two ways of tackling the problem of food- 

 cycles and community-organisation of animals. One way is 

 to start with one particular species and radiate outwards along 

 its various connections with other animals and follow the 

 train of associations wherever it leads. This was the method 

 described in Chapter V. It is a very fascinating form of 

 ecological work, owing to the variety of interesting facts and 

 ideas which are encountered, and it also has the advantage that 

 it can be carried out without any very elaborate previous survey 

 or listing of all the species of animal in the district. On the 

 other hand, one may list all the animals and then subdivide 

 them according to their place in the community — herbivores 

 and carnivores, key-industries, terminal species, large and 

 small, and so on. The separate food-chains can then be worked 

 out, and in this way one gets a better perspective of the whole 

 community. Perhaps a combination of the two methods 

 would be the best procedure. 



14. Another important subject about which something may 

 be said here is that of numbers. The study of numbers is a 

 very new subject, and perfect methods of recording the numbers 

 and changes in the numbers of animals have yet to be evolved. 

 In practice, we have to deal with two main aspects of this matter. 

 The first question is as to the best way of taking censuses of 

 the animal population at any one time, and the second is the 

 question of recording changes in the numbers from one period 

 to another. With regard to the first, a certain amount of 

 work has been devoted to the methods of estimating the absolute 

 numbers of various animals. Quantitative work on plankton 

 has reached a very high degree of efficiency ; the usual method 

 consists in doing counts of small samples from the whole 

 collection and then multiplying by a factor to get the total 



