156 ANIMAL ECOLOGY 



observations, which is quite justifiable in the case ot definite 

 experiments, or of the descriptions of dead structures which 

 will wait for you to observe them completely, and which can 

 be checked by other observers if they are sufficiently in- 

 terested. But in accumulating the life-history records of wild 

 animals it is essential to publish any tiny fact about them which 

 seems unlikely to be encountered in the near future, or which 

 helps to provide another piece in the complete jig-saw puzzle 

 which ecologists spend their time putting together. A good 

 example in this direction has been set by C. B. Williams, ^^^ 

 who has clearly demonstrated in his studies on the migration 

 of the painted-lady butterfly the value of collecting together 

 small and apparently isolated pieces of information, both from 

 his own notes and from the observations of others, which, 

 when carefully collated, throw a great deal of light on the 

 process of dispersal. 



10. We have said earlier that when an animal has accom- 

 plished the business of dispersal from its starting-point, the 

 next process consists in the establishment of itself as an 

 individual and then as a species, in its new home. There is, 

 however, this exception to be noted. It seems highly probable, 

 although difficult in the present state of our knowledge to prove 

 conclusively, that many animals migrate on a large scale in 

 order to get away from a particular place rather than to go 

 towards anywhere in particular. That is to day, there are often 

 very cogent reasons why a large section of the population should 

 migrate somewhere else, the most common one being over- 

 population. We see such pressure of numbers acting in the 

 case of lemmings, locusts, sand-grouse, and aphids, when they 

 suddenly depart in huge swarms for new lands. The point 

 here is that the immediate reason for migration may be to 

 relieve the situation in their normal habitat, and although it 

 is all to the good for each species to establish itself from one 

 of these migrating swarms in another locality (as happens in 

 the case of the Caucasian locust migrations), yet the main 

 effect of the migration is to remove a surplus population 

 from the area usually inhabited by the species, and not to 

 colonise other areas. It should, therefore, be borne in mind 



