82 ECOLOGY AND LIFE HISTORY OF THE COMMON FROG 



as free from bacteria as possible before he sterilizes it. If he does not 

 do this, satisfactory processes may fail — that is, steriHzation is popu- 

 lation control by inversely density-dependent means. 



It seems almost certain that the destruction of eggs and embryos by 

 heat would follow the same laws, so that the more eggs an individual 

 frog laid in the regions where heat death is important, the larger the 

 chances of the survival of some offspring. Now, Boulenger (1897), 

 almost certainly working on British frogs, counted 1,155 to 2,044 

 eggs in five frogs. Heron-Royer (1885), certainly working on French 

 frogs, counted from 2,856 to 4,005, almost twice as many. It is not 

 known exactly where Heron-Royer obtained his frogs, but he pub- 

 Hshed much of his work in the proceedings of a scientific society in 

 the provincial town of Angers, about 150 miles north of the southern 

 hmit of distribution, and perhaps within the area where heat death 

 might well be found. As a pure conjecture, I suggest that, in this area, 

 natural selection has favoured the frogs having the larger number of 

 eggs, and that this is a case in which the reproductive rate has become 

 adapted to the death rate, through the operation of an inversely 

 density-dependent factor. This is not at all contrary to Lack's hypo- 

 thesis, for he merely points out that death rates cannot directly control 

 reproductive rates — there must be some other factor working in such 

 a way as to give a selective advantage to an individual with the higher 

 reproductive rate, and all I have done is to show that such a factor may 

 well exist. It is even possible that there is another factor working in 

 the same direction. In France, but not in England, frogs including 

 this species are used as human food. Man is a predator in France but 

 not in England, at least for this reason. If it is unlikely that the popu- 

 lation of ducks is increased or diminished in proportion to the popu- 

 lation of frogs or tadpoles, then it is ludicrous to suppose that the 

 population density of Frenchmen is dependent on the population of 

 animals that merely provide a tasty dish from time to time. At the 

 very least, therefore, we cannot look for directly density-dependent 

 relations between frogs and Frenchman. More probably, we should 

 expect that a shortage of frogs would result in redoubled efforts by 

 frustrated Frenchmen, so that we should have another inversely 

 density-dependent factor. The custom of eating frogs may well be 

 very ancient, and it is surely not too far-fetched to consider the effects 

 of an additional and ingenious predator. 



To summarize my conclusions on the factors responsible for the 



