ARCHIANNEL1DA. 457 



separated lateral nerve cords placed close to, though not in, the ectoderm. The 

 ventral cords are sometimes ganglionated segmentally. 



The body-cavity is occupied (or traversed) by a network of connective tissue, 

 in which the genital cells appear to arise. The spaces of this network are 

 specially enlarged round the internal ends of the nephridia. There are no 

 septa, no dorsal and ventral mesentery, and no vascular system. 



The nephridia are in five pairs, of which the posterior in the male communicate 

 by a ciliated funnel with the cavity of the testis. With the exception of this 

 fifth pair, the nephridia open on the sides of the ventral surface : they are 

 simple tubes which internally possess a large flame-shaped cilium, or a row of 

 small cilia which give the appearance of a flame ; it is uncertain whether they 

 open internally or not ; indeed it is uncertain whether we ought to speak of a 

 body-cavity at all in these animals. The part of the nephridia next the external 

 opening is ciliated, and no nuclei are distinguishable in connection with their 

 walls. 



The animals are dioecious. The nephridia of the fifth pair are connected with 

 the testis internally and join together to open at the hind end of the body into 

 a vestibule, which opens externally in the middle ventral line ; they are dilated 

 near the opening, and constitute a pair of vesiculae seminales. There is a median 

 penis projecting into the vestibule. Spermatozoa are introduced into the female 

 by means of the penis which perforates the skin at any point of the surface, and 

 so introduces spermatozoa into the body-cavity. The ovaries like the testes 

 appear to arise from cords of the parenchyma, and the eggs when ripe probably 

 pass into a special median ventral passage in this parenchyma, which opens to 

 the exterior by a median pore at the base of the caudal appendage. 



There appears to be a thin layer of circular muscles, and a pair of longitudinal 

 muscular bands just external to the lateral nerve cords. 



Saccocirrus in many points of its anatomy closely resembles Polijgordius and 

 Protodrihis, near which it may be provisionally placed. The principal point of 

 difference is the presence of setae while the resemblances are numerous : the 

 ciliated pits, the uniformly segmented body, the oblique septa, the dorsal and 

 ventral muscles, the position of the ventral cords in the ectoderm. It clearly 

 resembles the two Archiannelidan genera far more closely than does DinopJuhis 

 (see below), which is without coelomic septa, oblique septa, or vascular system, 

 and has a body-cavity extensively occupied by a reticulum, and nephridia of 

 doubtful relations. Moreover, the structure of the male and female generative 

 organs of Dinophihis is quite different from that of Archiannelids, with the 

 doubtful exception of the female Protodrihis, which needs reinvestigation on 

 this point. 



It must be remembered that the sub-oesophageal muscular organ, upon which 

 so much stress has been laid as indicating affinity between Dinophihis and 

 Archiannelida, though present in Protodrilus, is altogether absent from 

 Polygordius . 



The union of Histriodrilus with the Archiannelids is more fully justified, 

 but before finally deciding the matter we require more knowledge of the 

 generative organs and nephridia, and a re-examination of the vascular system. 



In the present state of our knowledge the Archiannelida must 

 be regarded as having very much the same relation to the Chaeto- 

 poda as Chiton has to the other Gastropoda ; while, carrying on the 

 same comparison, Dinophilus and Histriodrilus occupy, relatively to 



