chapter lo 



METHODS 



In this chapter, I find for the first time that I am not talking to myself, 

 but addressing at least three kinds of people. Some zoologists may 

 find some of the methods of use to them, and, in particular, I can draw 

 attention to some neglected fields of study in which they could be 

 applied. Then there is the growing group of people who, in this 

 technical age, are qualified in some science, but have considered that 

 there is recreational value in another field of study, in which, perhaps, 

 their own special knowledge could make a novel contribution. 

 Finally, there is that large body of people who, while not likely to 

 write scientific papers, have a great interest in the lives of animals, and 

 can contribute very important data for the use of others. Chapter 8 

 is almost entirely built on their labours. I have not attempted to 

 separate these classes, for they overlap, and some of the things I shall 

 say might interest any of them. 



I begin by commending to notice the comments of Andrewartha 

 and Birch (1954) on pp. 10 and 11, paragraph (c), and elsewhere in their 

 book, in which they stress the need for statistics in ecology, and 

 criticize what they consider the misuse of other forms of mathematics. 

 As an ecologist, I think with them that it is my task to find out what 

 actually does happen : not what might happen under certain artificial 

 conditions, or in exceptionally simple natural circumstances. The lives 

 of animals are usually so complicated that we must study as many 

 variables as we can all at the same time. There is no way of doing this 

 except that which the statisticians have pointed out to us, and usually 

 we need more data than we can possibly collect for ourselves. It is 

 therefore remarkable that, as I pointed out in Chapter 8, so httle has 

 been done to use the very great amount of information that has been 

 collected together by various organizations such as the Royal Meteoro- 

 logical Society in their Phenological scheme. All I have done is to 

 pick out from the eighty or so animals in that scheme the one that 

 interested me. There is every reason to suppose that the others could 

 be investigated in a similar way, using, no doubt, different variables. 

 Williams (195 1) in a paper that is not itself a phenological research, but 



I2A— {T.914) ^77 



