5i 



There is also no real difference in the length of the calicles, though Klunzing er says 

 there is. Further the ribs on the surface do not give any support. In the typical D. nigre 

 I found them as well developed as in the typical D, micranthus. Also in Ortmann's specimen 

 of D. nigrescens is "das Cönenchym stellenweise doch etwas gestreift". So I think there is no 

 real difference between the species of Dana and that of Ehrenberg. 



From Stat. 277 I have numerous young specimens, some consisting of only a few calicles, 

 but showing the typical method of branching. 



*2. Dcndrophyllia japonica nov. spec. (Plate VII, fig. 6). 

 Japan. 1 ex. 



Corallum arborescent, with thick branches in all directions. Height of the colony 9 cm ; 

 the mainstem, just above the enlarged base, 2 cm. thick. The branches are short, ± 5 cm.. 

 The corallites on the branches are arranged in two rows. On some branches they are very 

 crowded and long, giving the branch a fan-shape; this is seen in some branches of D. micran- 

 thus too. The alternate bending of the corallites forward and backward is to be seen only on 

 one side of a few branches and even than not very clearly; so it seems that the corallites 

 are more irregularly arranged. On the mainstem secondary calicles have developed on the front- 

 and backside. The corallites usually protrude a little (3 a 4 mm.); some are ± 1 cm. long. 

 The ribs on the stem are formed by rows of granulations, with nearly no pores between them. 

 Towards the end of the branches the number of perforations increases and the ribs are more 

 prominent. On the corallites the ribs are low, sometimes with one row of granulations, but 

 generally the granulations are irregularly distributed. The intercostal furrows are here shallow 

 and much perforated. 



The calicles are rather wide especially those on the stem and somewhat elliptical in 

 shape (9 X l 1 mm.). On the branches they are a little smaller (7X8 mm.). The depth of the 

 calicles is 4 to 5 mm.. The septa all project above the edge of the calicle, those' of the first 

 and second cycle being somewhat thicker than the others. There are four cycles of septa, a 

 fifth cycle may be developed in some systems, and there are even indications 

 of a sixth cycle. The septa of the two first cycles are larger than the others. 

 The septa of the third cycle, being short, do not reach the columella. The 

 septa next to those of the first order are long and reach the columella. The 

 arrangement of the septa is shown in the accompanying diagram (fig. 3). The 

 edge of the septa of the first and second cycle is entire or with very small 

 teeth, the sides are covered with a few irregularly scattered granulations. The septa of the 

 higher cycles are jaggedly dentate on the edge. There is a well developed columella, :z 5 mm. 

 long, and ± 1 mm. broad, consisting of a spongy network. 



3. Dcndrophyllia minuscula Bourne. (Plate VIII, fig. 30). 



Deridrophyllia minuscula Bourne 1905. 



Stat. 49a. Sapeh-strait. Depth 69 meter. 2 ex. 

 Stat. 51. Molo-strait. Depth 69 — 91 meter. 2 ex. 



103 



1 pj 



