INVERTEBKATA, CRYPTOGAMIA, MICROSCOPY, ETC. 711 



The digestive tube in this group is more like that of the Insecta 

 (Hexapotla), thau to that of any other Arthropod ; it consists of a 

 buccal, a median, and a terminal portion, and is provided with a 

 pair of salivary glands, and with one or two pairs of Malpighian 

 tubes ; the buccal intestine may be short and broad (^Jidiis), or very 

 long and capacious (Cryptops), or very long and excessively narrow 

 (Geophilus) ; the epithelium maybe absent from its walls, and the 

 chitinous cuticle is in a few cases (Litliohius) provided with short 

 chitinous j^rojections which are directed backwards ; in Cryi)tops there 

 is a false gizzard intercalated between the fore and the median intes- 

 tine, which is very muscular, and is provided with a number of setee 

 or chitinous spinules. The terminal portion of the intestinal tract is 

 ordinarily short, but is elongated in the Chilognatha (herbivorous 

 forms), and is longest in Glomeris ; the following table will indicate 

 the variations in its disposition, concerning which it was stated by 

 the late and lamented Gervais, that it was only coiled iu Glomeris and 

 its allies : — 



ri „ ii . -1 1 f Coils occupyius: a large I o/ncn?. 

 Greatlj coiLd | of the cavity °. fph'-°^\f- . 



r...^^ •!„ i / Coils occupying a small J ^yP*<^P'- . 

 Crreatiy coiled < extent 1 ^^"^'ntanum. 



^ " j Geophilus. 



Simple terminal curvature Julus. 



No convolution or cui'vature Lithobius. 



This terminal portion is always richly supplied by trachefe. 



The salivary glands may be racemose (Lithobius, Cryptops) or 

 tubular (Julus, Glomeris) ; in Geophilus the excretory canal (efferent 

 duct) is very long, but in Julus and Glomeris it is short. The Mal- 

 pighian tubes, which are inserted into the point at which the median 

 passes into the terminal intestine, are ordinarily two in number, but 

 Julus has fom- ; they may be of the same diameter along their whole 

 length, or be dilated into reservoirs (Lithobius, Geophilus). By trans- 

 parent light they are seen to be of a yellow colour, and their secreting 

 cells are much smaller than in most insects. 



Examined physiologically, we find that the Myriapoda may be 

 carnivorous or herbivorous ; the former, or Chilopoda, live on flies, 

 earthworms, spiders, larvae, and so on, which they seize between their 

 forcep-like foot-jaws, and poison ; the effect of the poison of Litho- 

 bius on the house-fly is reported to be as rapid as the bite of the sj^ider ; 

 the victim, so soon as it is killed, is devoured rapidly, everything 

 being swallowed, either in large (Lithobius) or in minuter (Geophilus) 

 mouthfuls. The herbivorous forms (Chilognatha), live on decom- 

 posing vegetables (Julus), decayed wood (Polyxenus),OT moss (^Glomeris), 

 which they gnaw ; and in swallowing their food they, like the Chilo- 

 poda, take in some, though not so much, earth or sand. In Cryptops 

 the gizzard already mentioned delays the food, which is, in it, sub- 

 jected to the action of a digesting fluid of neutral reaction. In most 

 cases digestion proper is efi'ccted in the median portion of the intestine, 

 where the food is acted upon by a yellowish or brownish liquid 

 secreted from its walls ; this is ordinarily neutral, but in Lithobius, 



