SOUTH AFRICAN TERMITES. 57 



it is not, the furcate trachea is present but differently disposed, 

 fig. 6'2b (PI. VI). 



The second illustration turnishod by the wing of this 

 species is to be found in the field of the cubitus. This, in 

 the final wing, frequently displays more or less interrupted, 

 erratic, or independent ridges between the cubitus offsets, 

 some being ill-defined, others plainly marked. These ridges 

 are due to the pendulum-like swinging of the offsets of 

 the cubital and to the formation of stronger or weaker 

 thickenings about them, in their earlier positions. In short,, 

 a thickening once formed tends to persist, and so gives rise 

 either to a ridge or a rib. Figs. 63a, b are diagrams designed 

 to illustrate this point. Fig. 63a represents the first position 

 of two cubital offsets. In fig. 63b these are seen to have 

 moved forward, leavino- the thickenino- behind. The deserted 

 thickenings become ridges in the final wings, and those 

 occupied by the tracheae become ribs. 



As the final wing develops it forms into longitudinal rolls, 

 and into these is gathered all the tissue. The rolls soon 

 become contorted and vermiform ; they appear coiled up and 

 apparently inextricabh^ mixed, fig. 64. In this condition the 

 Aving is unsheathed. 



However, complicated as the arrangement looks, it is not 

 difficult to see that each roll is the counterpart of a rib that 

 will form in the final wing. In other words, the tissue has 

 centralised along the courses of the thickenings. Inasmuch 

 as the trachea) also grow one would expect to find them to 

 be the cores of those thickenings that formed latest about 

 them. This is so in part, but not wholly, and the condition 

 seems to be due to the tracheae not attaining to the same 

 length as do the rolls, and, therefore, not following- all the 

 plications. 



Well before the final ecdysis, the imaginal wing becomes 

 surrounded by a layer of air Avhich sepai-ates it from its 

 sheath . 



After the moult, the adult is wliite and soft-bodied. 

 The wings remain tightly packed, except for the expan- 



