CLADOCAEPUS DOLICHOTHECA. 51 



marked character. The parts of the rachis to which the backs of the hydro- 

 thecfB are appUed are divided, by imperfect septa (septal ridges), into 

 numerous very distinct chambers, while a few similar ridges also project 

 into the cavity of the intervening portion. 



Where the stem ceases to give off pinna?, it becomes divided into three 

 or four internodes by very oblique joints, so as to assume, for some way 

 down, the appearance of being twisted, and then continues towards the 

 hydrorhiza as a simple continuous tube (Fig. 5). Along nearly the whole of 

 its course from the termination of the pinnate portion to the base, the stem 

 carries a longitudinal series of tubular nematophores, which are situated 

 at short and equal intervals from one another, and give to this part of the 

 hydrosoma a close resemblance to certain forms of graptolites.* 



The phylactogonia, or protective appendages of the reproductive cap- 

 sules, resemble in form the antlers of a stag. Their branches are set with 

 large tubular nematophores. They arch over the front of the stem, their 

 branches crossing one another from opposite sides, and forming a cage-like 

 roof over the gonangia. They occiu- only on some of the pinn;v, which are 

 situated close to the distal end of the stem, one springing from each pinna 

 close to its origin. Though the pinnaa which carry them retain their normal 

 form, they are all more or less shortened, most of them supporting only 

 a single hydrotheca. 



It is difficult to form any well-founded opinion as to the exact homology 

 of these appendages. The nature of the changes which have resulted in the 

 formation of a corbula in certain species of Aglaophenia might lead us to 

 suspect that in Cladocarpus dolichotheca the phylactogonium represents the 

 mesial nematophore of the proximal hydrotheca of its supporting pinna. 

 The fact, however, that this nematophore is at the same time present in its 

 normal state renders such an explanation untenable. The phylactogonium 

 probably represents, in a greatly modified condition, the mesial nematophore 

 of a hydrotheca, which had itself been totally suppressed. 



The sex of the gonophores could not be determined in the specimen. 



* I have elsewhere (Gymnoblastie Hydroiils, p. 17!!) endeavored to show the probability ihat the den- 

 ticles of graptolites represent the nematophores of the Plnmularidaa, the hydrotheeie beinj^ entirely sup- 

 pressed ; and I have attempted to support this view on both anatomical and embryological grounds. As 

 the nematophores of the PluraularidEe are filled with sarcode capable of a rich development of pseudopodia, 

 the graptolites would by this comparison be brought into close relation with the Rhizopoda. They would 

 thus represent an ancestral form in which the affinities looked on one side to the Hydroida, and on the 

 other to the Rhizopoda. No hydranths were developed in them, for the hydroid characters had not yet 

 gained that ascendency over the rhizopodal which we see in the existing Plunudarida!, which, according to 

 this hypothesis, have iidierited their nematophores from the extinct graptolites. 



