HICKSON— POLYTREMA 459 



may be explained as an expansion of the communicating passages which are seen in 

 Sporadotrema and Homotrema. 



In Homotrema (text-illustration C) there are apparently no interlocular spaces. In 

 a vertical section through one of the verruca? a cluster of chambers may be seen, communi- 

 cating with one another by large irregular passages. All these chambers, except those at 

 the surface, have solid walls which are not perforated by foramina. The chambers at the 

 surface exhibit a convex outer wall perforated by foramina. At the apex of each verruca 

 a few chambers, with incomplete outer walls, can always be seen. 



The foramina. When a specimen of Sporadotrema is decalcified and stained, one of 

 the most noteworthy features to be seen is a series of approximately parallel moniliform 

 tubes which break up into branches at the inner end before terminating in a perforated 

 membrane on the outer wall of a chamber and open by a single large aperture at the 

 surface (Plate 32, fig. 32). 



These tubes line-^je foramina, as can be seen in a stained section of hard and soft 

 parts together, and they are of a chitinous texture. Similar chitinous tubes have been 

 described and figured for Polytrema by Merkel and Mobius, but whereas in Sporadotrema 

 the tubes may be as much as 2 mm. in length, in Polytrema they are rarely more than 

 •02 mm. in length. In Polytrema the tubes are usually simple, but as Merkel has correctly 

 pointed out they sometimes divide at their inner ends. In Sporadotrema they are simple 

 only in the region of the terminal chambers; but, on the branches and stems they always 

 divide into a considerable number of smaller tubes which terminate in the chambers. In 

 the figures I have drawn only three or four of these secondary tubes — that is the number 

 that can be seen in a thin vertical section — but there must be actually nine or ten branches 

 springing from each of the main foraminal tubes. 



In many of the tubes there may be seen a few or in some cases several chitinous plates 

 stretching transversely across the tubes and these plates have all the appearance of the 

 tabulae of a tabulate coral except in texture. In the specimen of Sporadotrema mesenterieum 

 that I have examined these tabula? seem to be more pronounced than in the other species 

 and, as shown in Plate 32, fig. 25, the foraminal tubes in this species have the appearance 

 of being regularly tabulate. 



Two important questions naturally arise concerning these tabula?: (l) Are they 

 complete tabula?, that is to say, do they completely occlude the foramen ? (2) Are they 

 supported in any way by calcareous tabula? ? 



To the first of these questions it is difficult to give a definite answer. In nearly all 

 the good sections I have examined of decalcified sections and of ground sections of the hard 

 parts they seem to be perforated, but it is still quite possible that in some cases they do 

 completely close the passage. With such delicate structures as these are, it is always 

 difficult to determine the extent of the damage done either by the process of decalcification 

 or of grinding. 



To the second question the answer is that in Sp>oradotrema mesenterieum there are 

 certainly narrow projecting shelves of calcareous substance supporting the chitinous 

 tabula? (Plate 32, fig. 33), and it is probable that they also occur in some of the older 

 foramina in S. cylindricum as well. 



