270 Transactions of the 



internal surface of its inferior membrane. The small disks are" 

 cup-shaped when at rest, and are not placed at right angles to the 

 pedicle, but obliquely to it. When in use they are flattened and 

 brought at right angles to the pedicle. This opens a communica- 

 tion between the cavity of the disk and that of the pedicle, and so 

 with the secreting sac. 



If we compare the structure of the smaller disks with that of 

 an ordinary insect's hair, we shall find they differ only in form. 

 Each consists of an external structureless layer, an internal cellular 

 one, and a cavity. The gland connected with the base of the hairs 

 of the pulvillus is also represented in other hairs. Several cater- 

 pillars have a hollow poison gland at the base of certain hairs, 

 which is very similar to the tarsal sac of perfect insects. Both are 

 probably a mere modification of the internal cellular layer of the 

 hair, which is itself only a modification of the internal cellular layer 

 of the integument. For hairs in insects are mere processes of the in- 

 tegument itself. In the larger disks we have the columella added to 

 these structures ; but this is clearly only a modification of the cellular 

 layer, which becomes fibrous in all the larger hairs of insects. The 

 difi'erentiation of the cellular layer of a hair into a fibrous and cel- 

 lular layer, separated by a large space, is a very remarkable modifi- 

 cation of the ordinary structure : it is, however, in my opinion 

 nothing more. The structure of the gland in the tarsus is so deli- 

 cate, and its contents coagulate so readily, that I have been unable 

 to make out whether there is a single locculus for each disk, or 

 whether the cavity of the gland is single. In either case I look upon 

 it as a mere modification of the internal cellular layer of the integu- 

 ment, and identical with the poison glands connected with the 

 hairs of caterpillars already alluded to. 



With regard to the functions of these hairs, Mr. Blackwall, in 

 a paper "On the Pulvilli of Insects," as long ago as 181(3, showed 

 so completely that most insects chmb vertical surfaces or walk on 

 inverted ones by the aid of a viscid fluid exuding from the hairs on 

 the under side of the tarsus, that I feel some apology is needed in 

 reviving the subject. The opinion is, however, very prevalent that 

 the phenomenon is due to atmospheric pressure acting on vacuum 

 suckers. This has been repeated so often by good observers in the 

 case of the fly, and more especially in the case of Dytiscus, that I 

 shall venture to give the evidence we possess upon the subject with 

 regard to the adhesion of the latter. For so far as I know, no one 

 has hitherto denied that this insect adheres by means of atmospheric 

 pressure. 



I will now repeat before you an experiment upon Dytiscus first 

 suggested by Mr. Blackwall, and made by him upon the fly, with 

 certain slight modifications made necessary by the nature of the 

 insect. I have here a male Dytiscus rendered insensible by the 



