23 



enlarged to much beyond the dimensions it has when ready for 

 flight, and is, indeed, larger than that which it had just before in the 

 pupa. 



The insect is so transparent at this stage that the process by 

 which the greater bulk is acquired may be observed. The abdomen 

 of the imago is for the most part a cylindrical tube of very delicate 

 thin walls, which, though they contain the whole anatomy of the 

 insect, appear to be little more than transparent membrane. These 

 walls are distended to their full capacity and the intersegmental 

 membranes fully stretched. The main contents of this hollow tube 

 are in its lower half watery fluid, in its upper half or so air, or at 

 least some colourless gas. 



It is by the addition of this gas to the ordinary contents of the 

 skin of the insect, by what I have called " inflation," that the 

 increased bulk is obtained. 



I have no doubt whatever that a precisely similar condition exists, 

 and might be easily observed in almost any other Tipi/la, and in all 

 or nearly all those Diptera that have exposed pupee, /. e. the 

 Orthorrhaphous families. 



In the Cyclorrhaphous Diptera it is usual for the larval skin to dry 

 up and form a cocoon for the pupa, and a lid is forced off this cocoon 

 when the fly emerges. 



In how many of these flies I do not know, but certainly in most 

 of the few that I do know, the pressure to produce this rupture is 

 obtained by a great temporary sac in front of the imaginal head, 

 which, after emergence, shrinks and practically disappears. This 

 has various names ; the frotifal vesicle seems a satisfactory one. I 

 ought to know, but I do not, how far this vesicle is filled with air 

 and how much with fluid, but observation of the flies immediately 

 after emergence will tell us that the abdomen is largely filled with 

 air, and the increased bulk of the insect, that enables it to press 

 from end to end of its pupal case and so force off the lid, is un- 

 questionably due to an increased total content of air. The frontal 

 vesicle probably contains fluid so forced into it, as well as tracheal 

 vessels. It probably has two uses, first to increase the length of the 

 insect, the body of the Muscidse and allied families not being easily 

 adapted for extension. The other use is to form an elastic pad by 

 which to convey the pressure, a pad of uniform structure, and not 

 like the front of the mature head consisting of antenm^ and other 

 delicate and sensitive parts unfitted for transmitting pressure. In 

 this respect its function is the same as the head-plate of many 

 Lepidopterous pupas {^Cerura, etc.), used to protect the head in 

 transmitting force for the rupture of the cocoon. It may be noted 

 that the frontal vesicle is a uniform structure, without any spines for 

 causing rupture, the old larval skin having a sort of suture, which 

 determines where rupture shall occur so soon as the tension is 

 adequate. 



In the Lepidoptera, inflation occurs at emergence from the pupa 



