186 THE CANADIAN NATURALIST. [June 



in the adjoining plates. Each one of these plates is, therefore; 

 connected by these canals with the four plates in contact with it. 

 The canals are excavated in the substance of the plates, and com- 

 municate with the central canal of the transverse tubes. The 

 canals are not always perfectly circular, but are often flattened or 

 irregularly circular. The endorhin varies greatly in the extent to 

 which it is developed. In some specimens the plates are well- 

 defined and rhomboidal, with perfectly circular pores at the angles. 

 In others the plates are not at all defined, the ectorhin being one 

 continuous integument without sutures, but always with the full 

 complement of pores. The latter in such specimens are not all 

 circular, but are variously shaped orifices, sometimes with rough 

 edges. There are also specimens in which the endorhin consists 

 of only a thin film capping, as it were, the tubes and inclosing the 

 canals, the pores being proportionally larger than they are in those 

 with well- developed plates. The end of each tube, in these speci- 

 mens, forms an irregular, rounded tubercle instead of a rhomboidal 

 plate. 



The tubular skeleton above alluded to consists of numerous 

 small, straight, rarely curved, cylindrical tubes or hollow spicula, 

 placed parallel to each other and at right angles to the plane of the 

 body-wall of which they form the greater portion. They connect, 

 and at the same time keep asunder, the ectorhin and the endorhin. 

 One of these tubes springs from the centre of each plate of the 

 ectorhin : it is, at its base, or next to the ectorhin, very slender, 

 but enlarges so as to attain its full thickness at about one fourth of 

 its length, and then remains at the same diameter throughout 

 until it reaches the endorhin, by a single plate of which its inner 

 extremity is, as it were, capped. The outer extremity of each tube 

 has four small slender stolons, one proceeding to each of the four 

 angles of that particular plate of the ectorhin from the centre 

 of which it (the tube) springs. It there seems to form a connec- 

 tion with the stolons of the three adjacent plates whose angles meet 

 at that point. The stolons are so arranged that one of them al- 

 ways points inwards towards the nucleus, and another on the oppo- 

 site side of the tube outwards or upwards. It is proposed to call 

 these the radial stolons ; they form continuous lines radiating in 

 all directions away from the nucleus. The other two stolons of 

 each tube project at right angles to the direction of the radial 

 stolons ; they form circles round the nucleus, and may therefore 



