MOTIONS OF INSECTS. 495 



side of their tarsi, which, though flaccid in old specimens, appear to be 

 inflated in the living animal or those that are recent.^ It is not improbable 

 that these vesicles, which are large and hairy, may act in some degree as 

 suckers, and assist it in climbing. 



The insects of the Orthoptera order are, many of them, remarkable for 

 two kinds of appendages connected wilh my present subject, being fur- 

 nished both with suckers and cushions. The former are concavo-convex 

 processes, varying in shape in different species, being sometimes orbicular, 

 sometimes ovate or oblong, and often wedge-shaped, which terminate the 

 tarsus between the claw, one on each foot. They are of a hard substance, 

 and seem capable of free motion. In some instances-, another minute 

 cavity is discoverable at the base of the concave part, similar to that in 

 Cimbex lutea.^ The latter, the foot-cushions, are usually convex appen- 

 dages, of an oblong form, and often, though not always, divided in the 

 middle by a very deep longitudinal furrow, attached to the under side of 

 the tarsal joints. Sir E. Home is of opinion that the object of these foot- 

 cushions is to take off the jar when the body of the animal is suddenly 

 brought from a state of motion to a state of rest.'* This may very likely 

 be one of their uses ; but there are several circumstances which militate 

 against its being the only one. By their elasticity they probably assist the 

 insects that have them in their leaps ; and when they climb they may in 

 some degree act as suckers, and prevent them from falling. But their use 

 will be best ascertained by a review of the principal genera of the order. 

 Of these the cock-roaches (Blatta), the spectres (Phasma), and the pray- 

 ing insects (Mantis), are distinguished by tarsi of five joints.^ The grass- 

 hoppers with setaceous antennae (^Acrida) have four tarsal joints. Those 

 with filiform antennce (Locusta and Acrydium), those with onsiform (Trux- 

 alis^), and the crickets (^GnjJlus), have only three. In Blatta, the varia- 

 tions with respect to the suckers and cushions (for many species are fur- 

 nished with both) are remarkable. The former in some (^Blatta gigajitea) 

 are altogether wanting ; in others (B. Petiveriand) they are mere rudi- 

 ments ; and in others (S. Madcrcc) they are more conspicuous, and resem- 

 ble those of the GryUida. The foot-cushions also in some are nearly 

 obsolete, and occupy the mere extremity of the four first tarsal joints (B. 

 orientalis, Americana, Capcnsis, &:c.). In B. Petiveriana there is none 

 upon the first joint ; but upon the extremity of the four last, not excepting 

 the claw-joint, there is a minute orbicular concave one, resembling a sucker. 

 In others (B. gigantea, he.) they extend the length of the four first joints, 

 and are very conspicuous. In some (B. Mouffeti K.^), which have no 



* Kirby in Linn. Trans, xi. 106. t. viii. f. 13. a. 



2 I observed this in the hind legs of a variety of Locusta mis:rntnrin. 



3 Philos. Trans. 1816, 325. t. xix. f. 5. ^4 ibjd. p. 395. 



* In a specimen in my cabinet of Blatta gi^rantea, the posterior and anterior tarsi of one 

 side have only four joints, while the intermediate one has five. On the other side the hind 

 leg is broken off, but the anterior and intermediate tarsi have both five joints. In another 

 specimen one posterior tarsus has four and the other five joints. 



« The name of this genus, properly spelled, is Trozallis, from the Greek Tpu)^a\\is, 

 Gryllus. 



'' This insect, which is remarkable for having the margin of its thorax reflexed, was long 

 since well figured in Mouffet's work (130./o'. infimn). It has not, however, been described 

 by any other author I have met wilh. It is common in Brazil. Some specimens are pallid, 

 while others are of a dark brown. It is to be observed that the Blattina are resolvable into 

 several genera. 



