On ike AnaUwijj of on Arenaceous Pohjzoon. 3 



of the polypide and can be examined from the suilace, wlien 

 it ])resent,s the ap[)eai-ance shown in Fig. 6. 



Tlie greater part of my oljservations were, howevei, made 

 upon sj)irit-preserved material. Small ]30rtions of the zoarium, 

 comprising several of the sandy masses, were stained in toto 

 in borax carmine, according to the usual method. After 

 dehydrating and dealing, the material was transferred to a 

 slide and teased up with needles in a drop of balsam. The 

 teased preparation was examined with a low power of the 

 microscope, and it was possible with care to pick out the 

 individual separated ])olypides on the point of a fine needle 

 and mount them by themselves on fresh slides. 



When a separate })olypide was thus mounted it was 

 possible, by carefully jnishing the cover-glass with a needle, 

 to roll it over into vari(jus positions, as might be required. 



The Ccencecium. 



The Coencecium is dichotomously branched, and the 

 branches come off in seveial planes (Figs. 2, 3, 4.) It consists 

 primarily of a slender chitinous tulje. At fairly regular 

 intervals, usually at each angle of the branching system, this 

 tube breaks up suddenly into a number of very delicate 

 tubular zooecia, which are invested in a common sandy 

 mass. 



Thus the whole coencecium is divisible into what may be 

 termed, for the sake of convenience, nodes and internodes. 

 The nodes are dense ao-oreorations of grains of sand, 

 enveloping and firmly held togetlier by the chitinous 

 zooecia (Figs. 1, 2, 3, 4, 5, N.) The internodes ai'e longer or 

 shorter, slender chitinous tubes, connecting the nodes 

 together. 



I have already mentioned that the coentecium, as a whole, 

 is dichotomously branched. This results from the ftict that 

 whereas only a single tubular internode enters the lower 

 surface of each ai'enaceous node, there are usually two such 

 tubes originating from its upper surface. The branching, 

 howevei", is not always a perfectly regular dichotomy; 

 sometimes more than two internodes come out from the 

 upper surface of a node. Thus in Cryptozoon concrefuni 

 I have observed in one instance no less than six short tubes 

 coming out from the surface of a terminal node, in addition 

 to the internode on which it was supported. As the node 



B 2 



