I04 NATURAL SCIENCE. February, 



a common median, free, internal plate with enormous muscles 

 radiating into the elongated lateral cavities of the mandibles ; and the 

 two other pairs of muscles, with the long tendons, also are very 

 characteristic (two small pairs of muscles issuing from the superior 

 margin of the mandibles are not visible). In Blatta the cavity of the 

 mandible turns backward, no median free plate exists, and of muscles 

 we find three : an enormous musculus adductor from the inner edge, a 

 musculus abductor from the outer edge, and a short oblique muscle from 

 the interior wall of the mandible. It may be remarked that Blatta, 

 which has been investigated by numerous authors, is without good 

 reason generally considered as being of low organisation ; the distance 

 from Blatta to such low forms as Japyx and Machilis is, in reality, 

 rather great, at least as great as between Blatta and Coleoptera. 



Lubbock makes mention of a second pair of maxillae in the 

 Thysanuran family CoUembola ; the limbs he has seen are the Jirst 

 pair of maxillae, as they originate in advance of the second well- 

 known pair on both sides of hypopharynx ; in Japyx this first pair 

 possesses a well -developed, three -jointed palp. Thus, in the 

 Thysanura we observe four pairs of mouth-limbs, homologous with 

 the four pairs in Isopoda and Amphipoda ; the conformity exists to 

 such a degree that in Machilis and the Malacostraca the two lobes on 

 the second maxilla originate from the second and the third joints, 

 while the lobes on the first maxilla in the lower Malacostraca are 

 produced from the Jirst and the third joints; finally, in the Amphipoda 

 the fourth pair, the maxillipeds, has one or two of the joints coalesced 

 in the median line, thus forming a kind of lower lip with palps — as 

 the fourth pair, the labium, in Machilis, etc. (These and numerous 

 other facts are mentioned in my preliminary note of 1893.) 



Not very far distant from the Thysanura we find Scolopendrella, 

 which forms a transition to the Chilopoda and the Diplopoda, the 

 distance between these three orders being much shorter than that 

 between, say, Phyllopoda, Isopoda, and Copepoda. 



The Crustacea, the Insecta, and the Myriopoda, are, in my 

 opinion, more closely related to each other than are the Arachnida to 

 any of them — not to mention the Pantopoda, with good reason now 

 generally accepted as a separate class. 



It is now, I think, generally admitted that the larval stages : 

 nauplius, zoea, larvae of insects, etc., have been very much over- 

 estimated in judging the relationships between the different classes of 

 Arthropoda and the relation of the whole series to lower animals. In 

 Korschelt and Heider's well-known "Lehrbuch" (vol. i., p. 904), we 

 find a similar view. 



The well-known characters for the four old Classes of Arthropoda : 

 the jointed legs (in all forms not much degraded by parasitism), the 

 arrangement and the quality of the muscles, the exuviation of the 

 skin, the eyes, etc., are, in my opinion, good evidence of their rather 

 close relationship to each other. 



