ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 
377 
received into a little capsule which may be sessile, but is more fre- 
quently continued into a long chitinous stalk. These fine stalks fuse 
to form a tendon, which is covered by hypodermis, and has an axial 
cavity more or less reduced, two details which suggest its origin as 
an integumentary invagination. 
Each fibre may be considered as a multinucleated cell, the sarco- 
lemma representing the cell-membrane. The semi-fluid homogeneous 
hyaline doubly refractive substance which fills the sarcolemma-tube, has 
a nutritive function in relation to the longitudinal and radiating fila- 
ments which are imbedded in it. The longitudinal filaments run 
parallel to the long axis, and are of course the essential contractile ele- 
ments ; the radiating filaments follow the surfaces of van Gehuchten’s 
reseau transversal , and are attached to the sarcolemma, acting as anta- 
gonists to the pressure of the internal substance, and supporting the 
longitudinal filaments. The author believes that they also transmit the 
nervous stimulus to the longitudinal filaments, and restore these to their 
position after contraction. 
Association of Mites and Ants.* — M. Charles Janet discusses the 
association of Discopoma comata Berlese and Lasius mixtus Nylander. 
Too little, he says, is known in regard to these interesting associations ; 
the most connected observations are those of Michael, who showed that 
Lselaps cuneifer , found in the nests of Cam^onotus herculeanus, ate the 
corpses of the ants and other insects. The Uropod which Janet observed 
occurs sparsely in the galleries, and in great numbers on the lame of 
males and queens, and especially on the abdomen of adult workers. An 
ant may bear from one to six mites. 
When Janet placed a mite in the nest, the ants attacked it with fury* 
though they seem to resign themselves to those they carry about. The 
attack may end fatally ; but very frequently the ant’s mandibles slip on 
the resistant and flexible carapace of the mite, which is then projected to 
a distance of 3-4 cm. 
The mites do not eat either the living larvae or the corpses of the 
ants ; they use their chelicerse to make minute punctures in the articular 
membranes of their host, and are true external parasites, absorbing the 
blood. 
Prothoracic Gland of Dicranura Vinula.f — Mr. O. H. Latter 
gives an account of investigations into the function, structure, and 
homologies of prothoracic glands. Experiments with Dicranura vinula 
have led him to conclude that in that species there exists a special 
relation between the undischarged silk and the formic acid secreted by 
the larva, and that the formic acid is utilised not only for defensive 
purposes during larval life, but also for rendering the cocoon extremely 
tenacious, hard, and waterproof. Similar experiments with the silk- 
glands of other species gave quite different results, and it is probable 
that the chemical composition of the silk in various species is far from 
constant. After a resume of the work of other investigators, from De 
Geer, who in 1745-6 accurately described the main features of the organ 
in question, the author gives a detailed description of the structure of the 
* Comptes Eeudus, cxxiv. (1897) pp. 102-5 (4 figs.). 
t Trans. Entom. Soc. London, 1897, pt. ii. pp. llo-23 (1 pi.). 
