BY L. HARRISON. 631 



Segmentation. — The relation of the dorsal shields to the segments 

 of the body has been variouslyinterpreted; and hardly any two cur- 

 rent textbooks of zoology will be found in agreement on the point. 

 Schmidt (1895) and Kenyon(1895) have, however, definitely estab- 

 lished the existence of a head and twelve segments; the first of 

 which, Schmidt, following Latzel(1880), calls the hind-head seg- 

 ment ; while Kenyon considers it to be the first trunk-segment. P. 

 amicus, owing to its large size, and the complete demarcation of its 

 segments ventrally, is a favourable form for the study of the seg- 

 mentation. I have, therefore, added a brief note on this subject. 



I agree withl Kenyon in dividing Pauropus into a head and 

 twelve trunk-segments. On the segmentation of the head, Kenyon 

 writes (1895, p. 82) — "'In the triangular head, so far as one may be 

 able to judge from the number of appendages, there are three seg- 

 ments, corresponding to the antennae, the mandibles, and the maxil- 

 lae. .. . But . . . there is other evidence of cephalic segmentation 

 and possibly there are more segments in the Chilognath head and 

 in that of the Pauropoda than the number of appendages would 

 indicate." Kenyon claims, for the Pauropoda, a close affinity w^ith 

 Polyxenus, in which Carpenter (1905, p. 478) has shown maxillu- 

 lae to be present. It is a matter of some doubt, therefore, w^hether 

 the mouth-parts of Pauropus consist of only tw^o pairs of appen- 

 dages. From an examination of the dorsal surface of the head of 

 P. amicus, I would suggest that at least five segments are present. 

 Behind the antennary segment, are four rows of clavate hairs. On 

 the trunk, each such row indicates a segment. In addition, there 

 is, in the young stages and in some adult specimens, a distinct 

 indication of a suture between the antennae and the anterior row of 

 hairs; and of a second between the second and third rows of hairs; 

 which w^ould give two dorsal plates, each bearing two rows of hairs, 

 strongly suggesting a comparison wdth the dorsal shields of the 

 trunk. 



In connection with the tnmk, the statement is usually made, that 

 eight segments fuse to form four double segments. It seems to me, 

 however, that all six dorsal shields have an equal value, each bear- 

 ing two rows of hairs, and each being in relation to two underlying 



