i897. ARE THE ARTHROPODA A NATURAL GROUP? 109 



Peripatus, except in connection with the jaws. The antennae are 

 outgrowths from the primitively pre-oral region, and hence cannot be 

 compared with the antennae of Insecta or either pair of antennae in 

 Crustacea. The ciliation of the nephridia and ahmentary canal is 

 without parallel in the Arthropoda. The external cuticular skeleton 

 is weakly developed and the legs show no true jointing. Then, too, 

 the eyes are unlike those of any true arthropod, but are strikingly like 



those of the polychaete Autolytus, Fig. 3. 



^^1^^:^^^^^^ ^ It is further to be remarked that in 



»3-^A^:7 .rv- ^ 'Ctov'^. Autolytus there are two pairs of eyes, the 



second pair occupying the position of the 



problematical structures labelled d by 



Balfour {Quart. Journ. Micr. Set., xxiii. pi. 



xviii., fig. 22). The muscular nature of 



";;tv-<»n'»"". o^'iVo"'' ° the pharynx is not arthropodan but 



o^oo^v W^o ® resembles rather that of the Chaetopoda. 



® ®®©)|/Wg0^|^ On the other hand the features relied 



upon for associating Peripatus with the 



. F'^\-^' ,^^f°" °^ ^^^ (tracheate) Arthropoda are the following : 



of Annelid, Autolytus. e, epi- \ ^ j^ ° 



dermis ; g. ganglion layer ; /, —the presence of tracheae ; but we must 



lens; n. optic nerve; /.pigment remember that at least two other kinds of 



layer; r, rods. Compare with tracheae occur in the Arthropoda which 

 eye of Peripatus as figured by 1 j • 1 j 1 r i.u c 



J, .r ^ . r fr. c could m no way be derived irom those 01 

 Balfour, Quart. Journ. Micr. Set., ■' 



xxiii., pi. xviii., fig. 22, 1883. Peripatus, and further, that the resem- 

 blances between the tracheae of Peripatus 

 and those of Hexapoda are, aside from the physiological one, 

 extremely slight. In fact the evidence, so far as tracheae are 

 concerned, points to convergence rather than to community of descent. 

 In Peripatus a pair of nephridia is set apart exclusively for 

 reproduction purposes, a condition which recalls the structures in 

 both Hexapoda and Crustacea ; but we must not forget that similar 

 relations occur among certain annelids as well as in other groups. 

 Of more importance are the facts that the dorsal vessel, with its 

 several ostia, is enclosed in a pericardial chamber, while the circulation 

 is lacunar and the coelom is much more reduced than in most 

 annelids. To these might be added the claws terminating the legs 

 and the character of the jaws, apparently appendicular in nature. 

 We must, however, remember that so far as our knowledge goes, we 

 are doing no great violence to morphology when we compare the feet 

 of Peripatus with the appendages of a syllid worm, or when we 

 compare their terminal claws to chaetopod setae. In short, I may say 

 that there is a pretty large amount of doubt as to whether the 

 Malacopoda are Arthropoda. 



Now, leaving Peripatus and the nauplius out of the question, we 

 are met with the greater problem : — Is the group Arthropoda a valid 

 one ? Or is it rather an assemblage of convergent forms with no 

 common ancestor nearer than the annelids ? 



