MYRIAPODA 117 arc connected by fine transverse commissures forming a so - called rope-ladder nerve -system. Visceral nerves are present, with an arrangement similar to that in Chsetopods rather than that found in Arthropoda. A dorsal and a ventral blood-vessel are present. The- dorsal vessel is, according to Gaffron, a true muscular heart provided with a pair of dorsally-placed valvular openings on each segment of the body, but without any trace of vessels in connexion with these. It lies as in other Tracheata in a pericardial sinus formed by a horizontally-stretched septum, and especially approaches in its general structure the corresponding organ of Diplopoda. The ovary in the female is closely similar to that of Lithobius in structure. The animal is viviparous, and the pair of long oviducts function as uterus. In the male the spermatozoa are long and filiform, exactly like those of Lithobius. They are actively mobile, and perform exactly the same movements as those of Lithobius, and, as in that genus, are formed into rod-like spermatophores. In its embryological development, which cannot here be followed, conclusions as to the affinities of Peripatus with the Tracheates receive strong confirmation. Early embryos of Peripatus bear a remarkable resemblance to those of scorpions, spiders, and Myria- pods, and the mode of formation of the procephalic lobes, nerve- system, and limbs is closely similar. The five-jointing of the limbs of the embryo of Peripatus is remarkably similar to that occurring in embryo spiders, such as Agclcna. In one important respect the embryonic history of Peripatus differs most remarkably from that of all other Tracheata, as far as is known. A large, open, slit-like blastopore is, as discovered by Balfour, formed in the very early embryo of P. capcnsis. This slit closes in its centre, and its ante rior extremity apparently remains permanently open as the mouth, whilst its posterior region persists as the anus. In the embryos of other Tracheates the only representative of this no doubt ancestral mode of development surviving is the so-called mesoblastic groove. The species of Peripatus live in moist places, in hollows in de cayed wood, and under stones and logs. The animals walk with a gait similar to that of caterpillars, with their bodies completely supported from the ground by their legs. When irritated they eject from their oral papillpe fine jets of the viscid slime secreted by their slime-glands. These fine jets form networks of fine threads, looking like a spider s web, which cling to the fingers with the tenacity of bird-lime. The New Zealand species is said to catch insects for food by means of these slime -jets. The animals roll themselves up like Julus when quiescent. They appear to be noc turnal in habits. The group EUTRACHEATA may be thus defined Tracheata of various form, usually with completelyjointed appendages ; never bearing diffuse stigmata indefinite in number, but with not more than two pairs on each somite ; ventral nerve-cords closely approximated, with well-marked infra-oesophageal ganglion, never united posteriorly dorsad of the anus ; no segmental organs present ; no blasto pore formed in the embryo; mouth and anus formed as a stomodaeum and a proctodteum respectively. The sub-class MYRIAPODA may be defined as follows Eutracheata with a head distinctly separate from the numerous closely similar posterior somites, with a pair of antennae and two pairs of jaws; with numerous similar jointed walking legs ; Malpighian tubes present. The Myriapoda, like Peripatus, approach the Annelids in having elongate bodies, either cylindrical or more or less flattened, composed of numerous similar joints or somites. They bear numerous pairs of walking legs on the somites posterior to the head, and in this particular differ most markedly from all insects except some of the most primitive forms, such as the Campodeidge, which approach them by the possession of rudimentary abdominal legs, and with which, as will be described, they are possibly connected to some extent by such a form as Scolopendrella. Some forms (Glomeridse) most remarkably resemble woodlice in shape, and hence Latreille s connexion, already referred to, of the Myriapoda with the Isopodous Crustacea. The head of Myriapoda is very like that of Insecta, and bears a single pair of antennae, the eyes when present, and two pairs of jaws. The first pair of jaws, the stout toothed mandibles, are, as in insects, always devoid of palps, but in nearly all cases they are jointed, a primitive condition not occurring in insects. The second pair, the maxillae, are f used _ together to form a sort of under-lip. In one aberrant family (Poly- zonidx) the mouth parts are formed into a tubular pricking and sucking apparatus. The body behind the head is composed of distinctly separated similar segments, usually numerous (in Pauropus nine only). There is no division of the body into thoracic and abdominal regions. The ventral nerve-cord stretches the entire length of the body, and is provided with a ganglion more or less distinct for each somite. The first three of its ganglia are fused to gether. Eyes are wanting in some forms, in others they are present on the sides of the head as simple ocelli or groups of ocelli. In one form only (Scutigera) are they compound and faceted. The structure of the ocelli differs remarkably from that of the simple eyes of spiders, in that the crystalline rods in each ocellus are so placed that their axes lie parallel to the plane of the equator of the simple lens, at right angles to the optic axis of the eye instead of in its direction. They are also not isolated by pigment. Grenadier hence concludes that it is impossible that any definite image can be formed on the rods, so as to convey any impression of it to the retinal cells with which they are in connexion. Hence these eyes must discrimi nate only variations in intensity of light. In Scutigera the internal structure of the eyes, though closely simulating that of the compound eyes of insects in many details, is in reality very peculiar and different from that in all other Arthropods. The digestive canal is simple and straight, except in Glomeridx, and ends in a terminal anus. At the commencement of that portion which acts as a rectum, and which in the embryo is formed from the proctodyeum, two or four long coiled urinary tubes (Malpighian tubes), homo logous with those of insects, open into it. There is a dorsal tubular heart, divided into a series of chambers correspond ing with the somites, and, as in insects, contained within a blood sinus (pericardium), formed by a horizontal septum stretched across the body-cavity, and provided with a series of pairs of wing-like muscles (alee cordis), by which the sinus- cavity is dilated. Respiration is effected by means of air- tubes or tracheae as in insects. In the Diplopoda, except ing the Glomeridse, in which they are ramified as in the Chilopoda, these are of essentially the same structure as in Peripatus, viz., tufts of unbranched tubes with feebly- developed spiral filaments springing from a common short sac-like chamber (fig. 5). Four of these sac-like chambers V ~"^" r c FIG. 5. Inner view of the sterna of a single segment of Jidtts londinensis, much trachea ; t, nne trachew given off from it ; ms, respiratory muscle attached to tracheal sac ; m, ventral body muscle. open on the ventral surface of the body by means of two pairs of stigmata on each somite. In the Chilopods wide tracheal tubes with well -developed spiral thickening of their lining membrane, springing directly from the stigmata, freely branching, and often anastomosing like those in in sects, are always present. In all cases single pairs of laterally-
