240 BULLETIN OF THE 



organs (ocelli) are rudimentary and destitute of pigment (Colossendeis 

 colossea, C. macerrima, Scceorhynclms) or entirely absent [Colossendeis 

 angusta). On the other hand, in Pallenopsis the ocelli are relatively of 

 unusually gi-eat size. All the other species are known to occur also in 

 shallower water, and the ocelli ai'e of the ordinary form. 



SccBorhynchus and Colossendeis are of especial interest as showing 

 clearly from anatomical evidence the complete independence of the 

 accessory legs and first pair of ambulatory legs, which has been already 

 proved by Dohru from embryological data. The accessory legs have 

 been something of a stumbling-block in the way of those who would 

 trace the Arachnid affinities of the Pycnogonida by a direct homology of 

 their appendages. In order to reduce the Pycnogouid appendages to a 

 convenient number for such homologizing, the accessory legs have by 

 certain writers been assumed ofF-hand to be simply branches of the first 

 pair of ambulatoi-y legs, with which they are usually closely united. 

 Dohrn showed that in the early stages there was every reason to 

 believe that the two appendages were innervated by entirely distinct 

 ganglia, aud therefore belonged to different segments of the body. And 

 in some adult forms the first ventral ganglion, which sujjplies nerves to 

 the palpi, accessory legs, and first ambulatory legs, is divided into an 

 anterior part supplying the two former appendages, and a posterior part 

 sending nerves to the latter pair of appendages. In Nymphon, and 

 perhaps in some other forms, these two portions are quite separate as 

 two independent ganglia, although remaining in close proximity. In 

 Scceorhynchus they are separated by a considerable interval, and con- 

 nected by slender commissures. These two ganglia are nearly as large 

 as the other ventral ganglia, so that there seems to be one more than 

 the usual number. Moreover, the accessory legs are separated by a 

 wide interval from the ambulatory legs, and are articulated to promi- 

 nent lateral processes from the body, scarcely distinguishable, except in 

 size, from those to which the ambulatory legs are attaclied. It is clear 

 from this case that this pair of appendages has nothing to do with the 

 ambulatory legs, but really belongs to another segment. In Colossendeis 

 the accessory legs have undergone still another and very remarkable 

 change of position. They have moved forwards and become so closely 

 united to the pialpi that the two appear precisely like the outer and in- 

 ner rami of a single appendage. As in the case of Savorhynchus, the gan- 

 glion from which they derive their nerves is entirely distinct from that of 

 the first ambulatory legs, the two being connected by long commissures. 



In all cases, however, the palpi and accessory legs are innervated by 



