208 Transactions of the 



firm hold on the vertical surfaces of trees and the under sides of 

 their branches. 



In flies we find an exactly similar arrangement, except that the 

 hairs are extremely minute and delicate, and that they have their 

 bulbous extremities flattened into disks. I have already drawn 

 attention to these structures, and to the manner in which the discoid 

 hairs are detached, in my work on the Fly. I may now repeat, 

 however, that although when all the hairs have the strain put upon 

 them equally they suffice to support the insect, each row may be 

 very readily removed separately. This is done by the manner in 

 which the tarsus is raised. The same seems to be the case in other 

 insects. 



In Dytiscus the pulvillus, or sucker as it is sometimes called, 

 consists of much larger and more complicated hairs. I think, how- 

 ever, there can be no doubt it is a mere modification of the usual 

 form of pulvillus, found on the feet of other beetles, flies, and a host 

 of insects. 



There are about 200 disk-bearing hairs on the anterior dilated 

 tarsus of Dytiscus marginatus, j^unctulatus, or circumjlexus, the 

 thi'ee common British species. Two of the disks on each tarsus 

 are of remarkable size, and differ in several points beside size from 

 the remainder. The largest of the large disks is situated on the 

 posterior portion of the proximal tarsal joint, close to the tibia, and 

 usually measures -iVth of an inch in diameter. The smaller is about 

 half this diameter, and is situated in front of the larger. The other 

 disks cover the remainder of the dilated portion of the tarsus, and 

 are not more than about yioth of an inch in diameter. The middle 

 pair of tarsi have theu' under side clothed with a dense pad of disk- 

 bearing hairs similar to the latter. 



If the integument of the upper surface of one of the anterior 

 dilated tarsi be carefully removed, the cavity of the tarsus will be 

 found to be occupied, in great part, by a large but delicate sac. The 

 main trachea with its sacculus and the tendon of the last tarsal 

 joint will be seen lying upon it, and these occupy all the remaining 

 space in the cavity of the tarsus. The sac will be found full of a 

 gelatinous viscid substance, the same as that which exudes from the 

 hairs. It is well suj^plied with tracheal vessels from the main 

 tracheal trunk, indicating that it is an active secreting gland. If 

 a portion of the upper part of the sac be removed, and its contents 

 cleared away with a camel's-hair pencil, the internal orifices of the 

 disk-bearing hairs of the pulvillus will be seen distinctly in the 

 centre of nipple-shaped projections of the integument which project 

 into the sac. Such a preparation, with a portion of the walls and 

 contents of the sac removed, is represented in Fig. 1. 



The largest disk is most easily examined ; and a vertical section 

 through its centre is the best mode of showing its structure. Such 



