387 



point of fusion of the tracheal trunks as they grew inwards, 

 towards each other, in the embryo. 



The longitudinal vessels open to the exterior by four 

 pairs of spiracles; one on the third segment, a second on the 

 fifth, the next on the sixth, and the last on the seventh seg- 

 ment, each connected by a rather short stigmatic trunk to 

 the great longitudinal vessels of the larva. The tracheal 

 vessels in the region of the fourth, and the eighth to the 

 eleventh segments, are provided in each segment with a pair 

 of small rudimentary trunks, which are related to certain 

 spiracles which do not develop till the next larval moult; 

 there are, therefore, nine pairs of spiracles functioning either 

 throughout larval life, or only in the later larval instars. 



Besides these nine spiracles there is a pair of tenth 

 rudimentary stigmatic trunks, situated in the twelfth seg- 

 ment. They do not open on to the surface till at the time 

 of the last larval moult, and become the posterior spiracles 

 of the abdomen of the imago. Of the ten potential stigmatic 

 trunks only nine, therefore, function at some time or other 

 during larval life. It is interesting to notice that the larva 

 of the honey-bee develops the full set of ten spiracles. 



In each segment the tracheal trunks give off a number 

 of thinner branching vessels, which on account of their 

 different structure I shall speak of as tracheoles. They are 

 clearly seen (figs. 1, 2) in living larvae as fine silvery lines 

 ramifying among the organs of the larva ; generally there are 

 two or three pairs in each segment which pass vertically, 

 as well as, especially in the more posterior segments, several 

 pairs which run dorsally, but are usually more difficult to see. 

 The anterior transverse vessel supplies the head by means 

 of two groups of tracheoles, which run forwards, but are not, 

 at this early stage, very strongly developed. From the pos- 

 terior transverse vessel several small branches are given off 

 to the anal segments. 



Structurally there is a very pronounced difference between 

 the great longitudinal and transverse tracheal trunks and the 

 stigmatic trunks, on the one hand, and the smaller branching 

 tracheoles on the other. 



In very young larvae the true tracheae are tubes with 

 an epithelium of rather thick cubical clear cells, which have 

 already secreted the "spiral" intima; before the end of the 

 first instar these cells have become slightly granular, and the 

 general growth of the larva is accompanied by a gradual 

 flattening out of these cells. The stigmatic trunks are similar 

 in structure (fig. 93) ; the spiracles are small cup-like struc- 

 tures, lined by an intima devoid of "spirals." Their intima 

 is shed, and a new one reformed at each moult. 



