VIII, D, 6 Light: Morphology of Eudendrium 337 



This species is named in honor of Dr. Lawrence E. Griffin, 

 chief of the department of zoology of the University of the Phil- 

 ippines, through whose kindness I was enabled to use the material 

 and at whose suggestion I took up the study of this interesting 

 form. Any value which this paper may have is due largely to 

 his help, and I wish to thank him here for his kindly aid and 

 advice. 



GENERAL MORPHOLOGY 



Hydrorhiza. — The hydrorhiza differs but little from the hy- 

 drocaulus in appearance. It is slightly thicker, with a heavier 

 perisarc and a more irregular appearance due to wrinkles in the 

 perisarc, to its very irregular branching, and to the presence of 

 particles of the substance of attachment which remain attached 

 to its sides and undersurface. Its perisarc, also, often has a 

 slightly more yellowish tinge than that of the hydrocaulus. The 

 main tubes are from 0.18 to 0.20 mm. in diameter and their 

 .branches from 0.18 to 0.19 mm. The main hydrorhizal tubes 

 are rooted by finger-like outgrowths which are swollen at their 

 distal end and characterized by a cap of unusually large ectoderm 

 cells. Branches are given off all along the hydrorhiza which may 

 either be hydrocauli or continuations of the hydrorhiza. The 

 hydrorhizal branches sometimes anastomose, forming a network 

 over the surface of attachment. This is best seen in those col- 

 onies growing on seaweed (fig. 2). These hydrorhizal branches 

 are simply the finger-like processes spoken of above, but here 

 found growing on the surface instead of rooted in it. They 

 differ from the hydrocaulal elements in that they originate as a 

 simple evagination of the ccenosarc, never assuming the appear- 

 ance or structure of a hydranth, whereas the hydrocaulus buds 

 arise as hydranth buds on the hydrorhiza and show hypostome, 

 tentacle zone, and sense ring before the pedicle is produced. 



It may be interesting to note here that in two instances hy- 

 drorhizal elements were found apparently developing hydranth 

 buds at the end of the tube. This was noticeable because of the 

 length of the tube as compared with the stage of development of 

 the bud, which contrasted strongly with conditions as found in 

 the hydrocaulus where young buds have little or no pedicel. In 

 both these instances the hydrorhizal elements extended some 

 distance beyond the surface of attachment, thus simulating the 

 habit of a hydrocaulus. 



Hydrocaidits. — A single hydrocaulus may bear from 1 to 10 

 or 12 hydranths, each on a separate pedicel. The length of these 

 pedicels (6 to 13 mm.) in proportion to the size of the hydranth 



