IN THE ANTENNAE OF INSECTS. 151 



The diameter of the simple sacs is about -^^so inch, that of the largest from 2^00 to 

 ,/oo inch ; the length of the longest sac ^^o inch. 



In the Grasshoppers {Tetrix) (Tab. XXX. fig. 11) I find a structure very similar to 

 that of the simple sacs of Libellula, and formed in the same manner ; but there is a distinct 

 papilla (of which there is some trace in Libellula) in the centre of the floor of the sac 

 (probably where the nerve is connected with it), and from this papilla a small chain of 

 minute bodies extends to the centre of the membrane wliich closes the apertm-e. This 

 connexion is well seen in profile (fig. 11, c 1), and, when viewed from above, the point of 

 contact with the external membrane is seen in the centre, as is indicated in fig. 11, B. 



The diameter of those represented in fig. 15, is about ttoo inch. There are about 

 twenty-two joints in this antenna, the lower of which possess about twelve sacs; the 

 number increases towards the apex, tiU about from the fourteenth to the last there 

 are 100 on each, so that we may estimate the total number on each antenna to be at 

 least 1000. 



In Gryllus, Gryllotalpa, and Blatta, I have found these structures in each joint of 

 the filiform antenna, generally from two to four in number, and commonly occupying 

 the same position and aspect of the joint. I have drawn those of Gryllus domesticws at 

 Tab. XXX. fig. 12. 



In the Coleoptera, so far as I have examined, similar organs are to be met with. 



In the attenuated antenna of the Longicorns, they are placed near the upper end of 

 each joint, where they form groups of eight to ten, one group on each side. 



In Silpha clavicornis they are scattered mostly on one side of the thickened joints, the 

 membrane by which the aperture is closed projecting hemispherically. 



In Meloloiitha vulgaris the whole surface of the lamella is thickly covered by organs 

 which are apparently of similar structure; these are figured in Tab. XXX. fig. 12. 

 They consist of an irregularly-round cup-like depression, the centre of wliich rises up 

 nearly to the external surface of the antenna. From this projection a tube runs inwards 

 quite through the internal layer of the wall of the antenna, whUe the external end 

 which forms the central projection is closed by a delicate membrane. 



The diameter of the cup is about ttW» while that of the central projection is ytoo inch. 

 The external layer of the cuticle is marked by hexagonal lines so disposed that each of 

 the cups occupies one of the areas considerably enlarged. 



The nerve entering the antenna is large, and at the base of each lamella sends into it a 

 branch, which dividing into four or five smaller branches, spreads out over the whole 

 internal surface. 



The number of these organs on each antenna is as follows. The area of one side of 

 each lamella is about 3^- inch by -^ inch ; each square of -^ inch has on each side 45 of 

 these structures ; the square would therefore hold 2025. As there are four of these 

 areas in the entii-e side of the lamella, the nimiber would be 8180. The sum for both 

 sides will be 16,360, and as there are seven lamellae on each antenna, the total number 

 of these organs on the whole antenna will be 114,520. A deduction of 2000 or 3000 

 may be made for irregularities, difference of size, &c. ; still, making the most liberal 

 deduction, we may say there are at least 100,000 on each antenna. 



