/ 



390 



and the co-operation of these must produce a very efficient 

 respiratory system for so sluggish a larva. 



So far as I am aware, no respiratory vessels similar to 

 those here described have been observed in other insects. The 

 complexity of shape of these great branching cells (fig. 76), 

 indeed, finds no parallel, except amongst the nerve cells of 

 higher animals. In some ways, indeed, they closely resemble 

 these; and their method of extension is very similar to that 

 observed by Ross Harrison in his well-known work on the 

 growth of embryonic nerve fibres in plasma media; while as 

 an example of a Trophospongial cell, in the sense in which 

 Holmgren employs it, they are quite unrivalled. 



In the majority of insects the smaller air tubes are true 

 multicellular tracheae, the terminal portion of which alone is 

 devoid of spirals and is evidently intracellular; it would seem, 

 then, that the great unicellular tracheoles above described are 

 homologous with the terminal portion of the tracheoles of 

 other insects. Indeed, Perez (1910, p. 191) gives an account 

 of the dfevelopment in Calliphora of the terminations of the 

 tracheoles among the muscles of flight, which is not unlike 

 the process by which the large tracheoles of Xasonia are 

 developed. From this it would seem to follow that the 

 dorsolateral air-sacs of the adult IV a son ia, as well as some of 

 the great head and abdominal vessels which develop during 

 pupal life (see below) are homologous with the general system 

 of smaller tracheal vessels occurring in other insects. 



The main change which the tracheal system undergoes 

 in the first instar is a slight increase in the complexity of 

 the tracheoles; towards the end of this period those stigmatic 

 trunks which have not yet opened on to the surface (the 

 fourth, and the eighth to the eleventh) grow outwards, and 

 at the next moult begin to function. 



The tracheal system has now attained to its mature 

 * condition, and during the rest of larval life is characterized 

 mainly by a considerable increase in size and the extent of 

 its ramification, as the larva itself grows. Especially marked 

 is this tracheal proliferation in the head region, where the 

 brain is developing. The increase in complexity of the 

 respiratory system is shown by comparing figs. 1 and 2, 

 and is due entirely, so far as I could observe, to an increase 

 in the size and complexity of the great branching cells, not 

 to a formation of new ones. 



Tlie extensive branching of the tracheoles makes it im- 

 possible to measure the size of these, but that the increase in 

 bulk is very large is unquestionable. The nuclei of the tracheoles 

 show only a slight increase in size. Thus, while tlie nucleus 

 of the tracheoloblast measured 21fx by 8/x at the end of the 



# 



