PEOOFS OF ORGAXISLXG MLXl) 





and Lepidoptera, he believes them to be the most nnmenjus in 

 species of all the orders of insects. 1 will now endeavour to 

 state in the fewest words possible tbe general results of his 

 studies, as well as those of the students of the other orders 

 mentioned, which are all in substantial agreement. 



In those insects which have the least comi)lete metamorpho- 

 sis — the cockroaches — the young emerge from the egg with 

 the same general form as the adult, but with rudimentarv 

 wings, the perfect wrings being acquired after a succession of 

 moults. These seem to be the oldest of all insects, fossilised 

 remains of a similar type being found in the Silurian forma- 

 tion. Locusts and Hemiptera are a little more advanced, and 

 are less ancient geologically. Between these and the four 

 orders with complete metamorphosis there is a great gap, which 

 is not yet bridged over by fossil forms. But from a minute 

 study of the development of the egg, which has been examined 

 almost hour by hour from the time of its fertilisation, the 

 conclusion has been reached, that the great difference we now 

 see between the larva and imago (or perfect insect) has been 

 brought about by a double process, simultaneously going on, 

 of progression and retrogression. Starting from a form some- 

 what resembling the cockroach, but even lower in the scale of 

 organisation, the earlier stages of life have become more sim- 

 plified, and more adapted (in the case of Lepidoptera) for 

 converting living tissues of plants into animal protoplasm, thus 

 laying up a store of matter and energy for the development 

 of the perfect insect ; wdiile the latter form has become so fully 

 developed as to be almost independent of food-supply, by being 

 ready to carry out the functions of reproduction within a few 

 days or even hours of its emergence from the pupa case. 



At first this retrogression of the first stage of growth towards 

 a simple feeding machine took place at the period of the suc- 

 cessive moults, but it being more advantageous to hav(^ the 

 larva stage wholly in the form best adaptcMl for the storing up 

 of living protoplasm, the retrogressive variations became stop 

 by step earlier, and at length occurred within the egg. At 



