58 THE ORIGIN OF 



Stage of the common frog, the animal grows and developes, while 

 it is not till the final stage that it becomes adapted for migrating 

 from one pond to another. 



In the birds we have an instance in which the operations of 

 feeding and dispersal are performed simultaneously, and conse- 

 quently no metamorphoses are required. 



The reason for the peculiar transformations of the marine 

 animals can be traced to the conditions under which they live. 

 In the ovum, the yolk is scanty by reason of the great number of 

 eggs produced ; and this, again, is necessitated by the extraordinary 

 risks to which the animals are liable, especially where the sea is 

 shallow. Another cause consists in the heavy armour with which 

 the adult Echinodermata are protected and which would render 

 migration impossible, or at any rate very difficult. Again, marine 

 animals are more easily dispersed by currents when they are small ; 

 consequently, the means of locomotion are eminently suitable for 

 an animal in its early incomplete stages. 



In the insects it is just the reverse. The mechanical difficult- 

 ies connected with the operation of flying can only be overcome 

 by an animal in its mature stage, and the same thing applies to 

 the amphibia, which for the purposes of dispersal require to be 

 capable of locomotion both on land and in water, a combination 

 which is naturally most easy of attainment in the mature state. 

 Thus, the operation of migration is naturally performed by the 

 adult animal and that of feeding by the larva, The result of this 

 division of labour is to cause the larva to degenerate. In propor- 

 tion as the adult becomes more and more highly developed, the 

 larva sinks lower and lower and loses its active organs, until we 

 come to larvEe which consist of little more than a soft, flabby sac, 

 furnished with jaws, or, going still further down in the scale, we 

 find that in the larva of the blow-fly even these jaws are replaced 

 by mere hooks. 



The type of the primitive insect, according to Fritz Miiller, was 

 very much like some of the present wingless orthoptera. Accord- 

 ing to Brauer and Lubbock, the nearest modern approach to the 

 ancestral type is represented by Campodea. The larvae of certain 

 carnivorous beetles are very similar to this type. Regarding the 

 wings of primitive insects, it is remarkable that many insects in 



