s rit il er du 20 Anil l\)il M 



cou 1(1 take piare in any part of the antenna, one 

 would tinci, (specially in youn«? specimens, shorter 

 joints in bctui^cn the olhers. As far as I know, no 

 siK-h joints have ever been observed. On the contrary 

 and more especially in yoiin«; specimens the joints 

 follow one another from the base to the tip in a per- 

 fectly rejrular manner, becoming gradually longer 

 and Nvithout any smaller joints in between. The only 

 |)oinl at which short, relatively wide joints, some- 

 what like a pile of rings, are found is immediately 

 after the third joint, from the 4th. to the 8th. The 

 shortness of these joints as well as their greater 

 breadth, equal to that of the third, lead one to sup- 

 pose that they represent newly formed elements- 

 While the older joints have had time to lengthen, at 

 the tinic of oliservation the neuer ones retain their 

 original dimensions. 



Moreover the length of the third joint is not 

 constant, but varies in different specimens and at 

 different ages. For example, I have observed that 

 the third joint of a new ly hatched specimen is 6 times 

 as long as the fourth, while in the antenna of an 

 adult it is only twice as long. This must be so if the 

 tip of the third joint acts as a centre of growth. It 

 should also be observed that marks of division (cir- 

 cular notches) can be seen frequently in the 

 neighbourhood of the tliird j(Mnt and nowhere else. 



These observations added to the facts ascertained 

 in the case of Termites lead us to conclude that the 

 increase in the number of joints is the result of suc- 

 cessive divisions of the third joint or, to be more 

 evaet, of its tip (fig. 1). 



