FOOD, HIBERNATION AND MIGRATION II3 



perceptible at a distance? Except for perhaps carbon dioxide, hydrogen 

 sulphide and ammonia, the inorganic components of pond water are not 

 volatile. Carbon dioxide is so common, and is produced by the frogs 

 themselves in far larger quantities than could be present in the emana- 

 tions from a pond, that it can surely be eliminated. I doubt whether 

 hydrogen sulphide is given up to the air from normally oxygenated 

 ponds, for, in balanced aquaria, lead cements darken below the sand 

 line, but not above it. Ammonia is surely out of the question, not 

 only because it is probably not evolved, but because, if it were attrac- 

 tive to frogs, they would be drawn to manure heaps instead of going 

 to the ponds. So it seems that we must look to organic compounds 

 for any possible stimulus. 



Czeloth (1930) was probably wide of the mark when he suggested 

 marsh gas as a guide to the migrations of the newts he was studying, 

 but much nearer when he mentioned water plants, for it is these that 

 impart the characteristic odour to pond water. It is, however, not the 

 higher plants that do this, but the algae. There is much information 

 on this matter, for it is of economic importance. If certain species of 

 algae grow in drinking-water reservoirs to too large an extent, the 

 consumers complain that the water tastes or smells. The odours are 

 not always unpleasant, but people do not like water that has a strong 

 smell or taste, whatever it is. The odours are due to essential oils 

 elaborated by the algae, and the smells are so characteristic that a skilled 

 person can detect and identify the species of alga sometimes before it 

 can be located under the microscope. I once detected a smell from a 

 pond (Large Totteridge) many yards from the bank, and suspected 

 from the textbook description that it was due to Synura uvella. Micro- 

 scopic examination showed that this species was abundant in the water. 



The fit of this hypothesis with most of the facts in the field is very 

 good. The smells are found in ponds — nowhere else in the whole 

 countryside. Any particular smell is probably only found in a few 

 ponds for there are so many species of algae that, in a limited area, 

 there are hardly any two ponds with the same flora. Ponds tend to have 

 the same species in successive years, but this is not invariably so. The 

 plants are, of course, gready influenced by the mineral salts in the 

 water: it vdll be recalled that potassium and phosphate, both impor- 

 tant plant nutrients, are associated with the preferences of frogs. I 

 shall show in the next chapter that the variation of the spawn date 

 follows a statistical pattern that is similar to that shown by aquatic 



