INSECT TRANSFORMATIONS. 59 



the embryo stage show traces of three pairs of wings as well as 

 three pairs of legs, and it therefore seems probable that the front 

 pair were not retained because they were too far away from the 

 insect's centre of gravity. The breathing trumpet of the pupa of 

 the gnat is probably a modified form of these appendages. In 

 certain fossil insects these " prothoracic wings are well developed," 

 and according to Broginart this is notably the case in certain 

 palaeozoic insects from the Coal Measures. 



The pupal stage of insects is unique, and has no parallel in 

 any other division of the animal kingdom. It has been suggested 

 that the necessity for this stage originated in the changes which 

 the mouth parts undergo. This explanation is insufficient, because 

 many insects which pass through a quiescent pupal stage do not 

 change their food in passing from the larval to the imago stage. 

 Sir John Lubbock has suggested that the pupal stage is a conse- 

 quence of the immature condition of the larva at the time of its 

 birth ; and Brauer has pointed out in confirmation of this theory 

 that insects which have large eggs are not generally quiescent as 

 pupae. But there are exceptions to this rule ; moreover, the pupa 

 is too remote from the larval stage for the conditions at its time of 

 birth to have much influence on it then. Again, in many cases, 

 the larva in the egg positively retrogrades, so that at the time of 

 hatching it has degenerated from its form in the earlier stages of 

 the egg. 



A more hkely explanation is that pupation is a consequence of 

 this very degeneration of the larva. As the imago developes, the 

 larva degenerates, and extremes thus recede until a resting stage 

 is required for the regeneration of the insect. Thus, the blow-fly 

 is a highly-developed insect, but its larva is little more than a 

 digestive sac, furnished with hooks, which, as above mentioned, 

 take the place of proper jaws. 



The divergence between the larva and imago state is in many 

 insects doubtless partly due to their honey-sucking habits. Of all 

 insects, the bees have been the most influenced in this respect by 

 flowers ; lepidoptera come next ; and many diptera are also honey- 

 suckers. Not only have insects played a prominent part in the 

 development of flowers, but the juices of flowers have formed an 

 important article of food to insects, the more so as honey can be 



