24 MELANDER. 



cient to insure a feeling of security no matter how its body may 

 be placed. In its natural habitat the insect is normally found 

 clinging to the under surface of its sheltering stone. But when 

 driven from its gallery, Enibia rebels if placed in an inverted 

 position and rights itself immediately by wriggling the tail and 

 clutching with the forefoot, which probably at the same moment 

 emits a strand of silk. 



The antennae of Enibia appear to be less sensitive to touch 

 than the body. The insect would frequently butt against some 

 obstruction notwithstanding the warning received through the 

 antennas. On one occasion the warning was insufficient to pre- 

 vent Embia from walking into a drop of honey, and after an en- 

 forced bath the insect seemed more solicitous of its forefeet than 

 of the rest of its body, carefully eating off the honey from their 

 plantar surfaces. When about to clean its antennae Embia rapidly 

 turned its head and thus brought one of these organs under the 

 body and then standing over it gradually drew it forward, at the 

 same time eating off the honey. This was the only time the 

 insect was observed to eat anything of a vegetable origin, and 

 then it was only with great reluctance. On several occasions 

 small pieces of a worm and a fly attracted its attention but after a 

 few nibbles it would run away. The abandoned web at the edge 

 of the jar seemed to be a more toothsome morsel, for Embia was 

 frequently observed nibbling at it. 



The spinning glands of Embia (Fig. 4) are unique and without 

 parallel. The silk is produced in the thickened anterior meta- 

 tarsi within chambers and conducted outside to the recurved 

 hairs at the edge of the plantar surface. The chambers of the 

 metatarsus which may be seen shining through the chitinous 

 integument, appear more or less regularly arranged in three 

 longitudinal layers, one next to the sole, a second in the middle, 

 and a third dorsally. In each series there are, roughly estimated, 

 about twenty-five chambers, placed about three abreast. Thus 

 in the whole joint the chambers number between seventy-five and 

 eighty. The cavities are nearly all of the same size about sixty 

 micra in diameter though the outer chambers are somewhat 

 smaller. Each chamber is more or less cuboidal and bounded 

 by a single layer of epithelium. This is for the most part quite 



