THE STRUCTURE AND AFFINITIES OF HERDMANIA CLAVIFORMIS. 247 



The epicardiac tubes are situated ventrad from the abdominal viscera, and are 

 strongly compressed dorso-ventrally. One of them, the right, is considerably larger 

 and thinner-walled than the other (PI. XVIII, Fig. 8, e'er, dx., e'er. s.). In many 

 places the wall of the smaller tube attains a considerable thickness, the cells becom- 

 ing cuboid and easily stainable. More frequently than otherwise in the regions where 

 the wall is thickened the lumen of the tube is more or less completely filled with a 

 fine granular or flocculent substance which does not stain readily and which con- 

 tains no cells. There is little doubt that this is a secretion from the walls of the tube. 

 In some places not everywhere this substance grades into the cytoplasm of the 

 cells so gradually as to leave hardly a question that here the cell substance is becom- 

 ing transformed into the secretion. Even where the secretory process is going on, 

 the portion of the cell on the outside of the nucleus from the lumen of the tube is 

 much more finely grained than the secretion; in fact it generally appears almost 

 perfectly clear and structureless (PI. XIX, Fig. 12). I have never seen any of this 

 secreted substance in the larger tube. In a few places, however, the wall of this 

 tube is also found to be thickened somewhat. This is particularly true along the 

 edge adjacent to the other tube. 



The relation of the tubes to the pericardium is as follows: The larger thin- 

 walled one lies ventrad to the pericardium and is in close contact with it. Whether an 

 actual fusion of the two walls exists or not in the adult I have not determined, but 

 in the embryo zooid the pericardium arises from this tube. The smaller tube is situ- 

 ated, at its posterior extremity, on the dorsal side of the pericardium; so that here 

 the two tubes are on opposite sides of the organ as they are in Fragaroides (Maurice, 

 '88). It soon passes, however, on its way forward, to the left side of the pericardium 

 and so reaches the position it occupies for the most of its course alongside the other 

 tube. The two tubes, distinct from each other, can be followed with great ease 

 forward into the posterior end of the thorax. A little before reaching this they 

 diverge from each other and pass one to each side of the oesophagus. Whether their 

 original connection with the branchial sac is ever retained in the adult is somewhat 

 difficult to ascertain. It is certain, however, that such is not usually the case. This 

 I have determined with certainty on several zooids. That the connection was with 

 the branchial sac and not with the peribranchial is proved by the position occupied 

 by the detached ends. The branchial sac projects backward slightly in two broad 

 shallow pockets on each side of the oesophagus at its posterior extremity. On the 

 outer sides of each of these rest the posterior tips of the peribranchial sac, and on 

 their inner sides the free ends of the epicardiac tubes. 



