INSECTA 435 



Thus when we compare the cockroach with such forms as the tsetse 

 fly (Glossina) and the blow-fly Calliphora^ we find these latter deficient 

 in certain enzyme classes, the former in carbohydrases, the latter in 

 tryptases and peptidase. The evolution of the habit of feeding on 

 blood (which consists so largely of proteins) involves the loss of the 

 enzymes which digest carbohydrates and fats. Similarly the blow-fly 

 which exists on a diet in which carbohydrates are predominant has 

 to a certain extent lost its proteolytic and lipolytic enzymes. 



This principle has an even wider application. In the leaf-mining 

 caterpillars of the Lepidoptera, certain species are restricted to the 

 upper and others to the lower parenchymatous layer of the leaf. If 

 an egg of one species is accidentally deposited in the wrong layer of 

 the leaf, death of the larva ensues owing to its inability to digest the 

 proteins of that layer. Thus each species, it is said, has enzymes which 

 are specialized in the narrowest degree for digestion not only for the 

 proteins of a single plant but for those of a particular part of that plant 

 (all others being unsuitable). Sucking forms, like Aphis, explore 

 diff^erent regions of the plant tissue and it may perhaps be inferred that 

 they have a wider range of enzymes than the leaf-miners. 



Most interesting of all is the relation of phytophagous insects to 

 cellulose, which is incapable of digestion by any vertebrate. Only a 

 few wood-boring beetle larvae [Cerambycidae) have been shown to 

 possess an enzyme which digests cellulose. The great majority of 

 insects do not possess a cellulase and as all plant cell contents are 

 contained within cellulose envelopes, it is clear that digestion can 

 only follow when either protoplasm is released by mechanical injury 

 of the cell-wall or the enzymes are able to penetrate the cell-wall and 

 act upon the contained protoplasm. In lepidopterous caterpillars, 

 which digest vegetable protoplasm with much greater success than 

 do mammals, the latter explanation has been shown to be true. 



The insects which live on wood (excluding the Cei^ambycidae) can 

 be divided into two classes: (i) those, like bark beetles, which feed 

 on fungi, growing in their tunnels, and (2) those which harbour sym- 

 biotic organisms in special parts of their alimentary canal. In the 

 latter class maybe mentioned the wood-boring larvae of certain crane- 

 flies and of death-watch beetles (e.g. Xestobium). In these cases the 

 supposed symbiotic organism is the yeast, Saccharomyces. How it 

 assists in the assimilation of wood is not known. On the other hand, 

 those of the termites which eat wood in normal life always contain 

 the flagellates belonging to Trichonympha and other genera (p. 68) 

 living free in the intestine. The absolute dependence of certain 

 termites on the flagellate is shown by the fact that when the flagellate 

 fauna is removed (which can be done without harming the termite 

 by heating to 40°) the termites will starve although they continue to 



