MUSEUM OF COMPARATIVE ZOOLOGY. 245 



nections between the cavities of the hydrothecae and those of the stem and also 

 between the different stems through their lateral communications, while others 

 were carefully cleaned of their contents and stained with cosine, and of still 

 others numerous cross-sections wore also made, I arrived at the following results. 

 The basal portion of the main stem and of most of the branches are polysipho- 

 nic, and these cauline tubes decrease in number towards the distal extremities 

 until they are but two, then one, and finally they are all discontinued ; where 

 two exist, they are found to occupy opposite sides of the branch, one on the 

 upper and one on the lower surface, the hydrothecae imbedded in and stretch- 

 ing out from the sides of the branches ; where one only is left, it usually occu- 

 pies the upper surface of the branch. These cauline tubes are provided with 

 a number of perforations, regularly arranged, by means of which they have 

 direct communication with the contents of each hydrotheca. Where but one 

 tube exists, the perforations may readily be seen in the floor of the tube or 

 stem, one near the base of each hydrotheca and towards the opposite side of 

 the stem from which the hydrotheca near whose base it lies is imbedded ; 

 they are therefore arranged alternately on opposite sides of the tube. See 

 Plate III. Figs. 16, 20. 



Where there are two tubes of the stem, an upper and a lower, tliere is an 

 orifice in the lower tube directly opposite each one in the upper tube, and 

 through these apertures there extend processes of the coenosarc which com- 

 municate with each other and with the neighboring hydrotheca). By this 

 means direct and abundant communication is established between all portions 

 of the cauline stems and the hydrotheca). By the aid of the cross-sections I 

 find that the walls of the cauline stems, where there are two of them, never 

 adjoin one another, while the walls of the opposite hydrotheca) do touch 

 throughout. See Plate III., Figs. 17 b, c, d. 



Thus far the structure described does not disagree with anything given in 

 Professor AUman's description, but careful caniera-lucida sketches of front 

 views of portions of the extremities of the branches where the cauline tubes 

 are discontinued do not indicate any floor to the hydrothecae, but show them 

 to be fully oi)en below without a floor. Still more distinctly and conclusively is 

 this shown in a camera-lucida drawing of a lateral view. See Plate III., Fig. 18. 



Figs. 19 and 20 on Plate III. are from the type specimens, showing the 

 same structure. In Fig. 19, Plate III., the uppermost hydrotheca belongs to 

 the right side of the branch, the median and lower ones to the left side. In 

 Fig. 18, Plate III., the hydrothecae of one side only are represented. 



Sertularia complexa sp. nov. 



Plate IV. Figs. 26-38 b. 



Trophosome. — Hydrocaulus attaining a height of about three quarters of an 

 inch, unbranched, divided by transverse nodes into rather short internodes ; 

 hydrothecae not springing from opposite sides, but imbedded in one side of the 

 stem, deep, the lower halves of each pair resting against each other, the distal 



