ii.] PREHISTORIC CATERPILLARS. 61 



have seen, were only very gradually assumed. In the 

 third stage, the line has disappeared, leaving the white 

 spots. In the fourth, the caterpillars have become very 

 variable, but are generally much darker than before, and 

 have a number of white dots under the spots. In the 

 fifth stage, there is a second row of white spots under 

 the first. The caterpillars not being good to eat, there 

 is, as has been already pointed out, no need for, or 

 attempt at, concealment. Now if we compare the 

 mature caterpillars of other species of the genus, we 

 shall find that they represent phases in the development 

 of D. euphorbia. D. liippophae, for instance, even when 

 full grown, is a plain green, with only a trace of the 

 line, and corresponds, therefore, with a very early stage 

 of D. euphorbia ; D. zygophylli, of South Eussia, has the 

 line, and represents the second stage of Z>. euphorbia ; 

 Deilephila livornica has the line and the row of spots, 

 and represents therefore the third stage ; lastly, D. 

 vespertileo and D. galii have progressed further, and lost 

 the longitudinal line, but they never acquire the second 

 row of spots which characterises the last stage of D. 

 euphorbia. 



Thus, then, the individual life of certain caterpillars 

 gives us a clue to the history of the species in past ages. 



For such inquiries as this, the larvae of Lepidoptera 

 are particularly suitable, because they live an exposed 

 life ; because the different species, even of the same genus, 

 often feed on different plants, and are therefore exposed to 

 different conditions : and last, not least, because we know 

 more about the larvae of the Lepidoptera than about those 

 of any other insects. The larvae of ants all live in the 

 dark ; they are fed by the perfect ants, and being 



