294 



of the mostly quadrangularly rounded sections of the zocEcial cavities. Just as 

 the kenozocecia mentioned seen from the surface (fig. 2 a) are of very different 

 form and size, the same is the case with their transverse sections, and the inner 

 cavity in some of considerable size is in others almost slit-like. In two of them 

 (one on the left and one on the right side) there is only a distinct cavity in the 

 one half, whilst in the other the two walls lie close up against one another, and 

 this enables us to readily understand why these kenozocecia in many species, e. g. 

 in Ret. Beaniana and Ret. phoenicea, have no inner cavity even in the youngest 

 parts of the colony. On the kenozocecia arranged in layers we can distinguish 

 between three kinds of walls, terminal separating walls which separate the keno- 

 zocecia at different heights, lateral walls which separate the kenozocecia at the 

 same height and interjacent walls which separate the kenozocecia in different 

 layers. Whilst the last are provided with pore-canals, the two others are provided 

 with uniporous rosette plates (figs. 2 d and 2 e), which according to the extent of 

 the chambers may sometimes be placed in a zigzag row, sometimes in a single row. 



That we have here actually to do with layers which arise the one outside 

 of the other, is specially distinct in those cases whei-e for example a Spirorbis 

 is fixed on a Retepora colony, as the tube of the worm then becomes covered by 

 layers which grow up over it from the surface of the colony. This is seen for 

 example on PI. XXIII, fig. 3 a, which represents a section, obtained by grinding, 

 through a colony of Ret. cellulosa from Oran. We see here a Spirorbis tube cov- 

 ered by an expansion, which on the one side is two-layered and on the other 

 single, whilst on each side of the tube there is an elongated, triangular cavity, 

 which separates the covering layer from that on which the tube rests. On PI. X, 

 fig. 2 c is shown a part of a colony of Ret. mediterranea, which is much younger 

 than the fragment represented in fig. 2 b, b*it here also the frontal surface of the 

 zooecia is covered by kenozocecia. Fig. 3 b shows a section, obtained by grinding, 

 through the terminal part of a branch of Ret. Wallichiana and the kenozocecia 

 here also are provided with a distinct cavity. 



Time has not permitted me to enter further into the classification of this 

 family, and I have been obliged to restrict myself partly to describe a single 

 incrusting species, partly to mention a number of earlier described incrusting spe- 

 cies, which must be referred to this family. These are for example, Schizotheca 

 fissa Busk, Schiz. divisa Norman, Rhynchozoon^ bispinosum Johnst., Rhynch. longiro- 

 stre Hincks, Rhynch. profundum Mac Gill., Rhynch. crenulatum Waters, ^ Miicronella<^ 

 tubulosa Hincks, >Schizoporella^ armata Hincks, y Schiz. '^ lucida Hincks, * Schiz.* 

 scintillans Hincks etc. 



' As to the name Rhynchozoon see 25, p. 125 and 38 b, Index. 



