viii. D. B Light: Morphology of Eudendrium 351 



cases division of a cell into four. Their average diameter was somewhat 

 over 9 /^^. 



Those in E. gnffini are spherical in outline, 6 to 6.5 ix/x in 

 diameter, and have a very distinct nucleus and pyrenoid. The 

 nucleus, which is roughly spherical, or oval, stains very deeply 

 in haematoxylin, safranin, haemacalcium, Bismarck brown, etc., 

 and seems to be a homogeneous mass, not showing the granular 

 composition found by Mangan ('09) in the nuclei of the zooxan- 

 theliae of Millepora. The granules in the cytoplasm are nu- 

 merous and irregular, seeming to form a sort of network. This 

 shows very clearly in sections stained with Delafield's haema- 

 toxylin and counsterstained with eosin. In sections stained with 

 eosin alone these granules are strikingly shown, together with 

 the pyrenoid and its surrounding clear area. As this stain does 

 not bring out the nuclei, it is useful in distinguishing between 

 them. The pyrenoid, with its surrounding clear space, has a 

 very characteristic luminous appearance. It is spherical, and 

 stains much less deeply than the nucleus. The clear area about 

 the pyrenoid is very definite in its boundaries. The sharply 

 spherical shape, the lightly staining pyrenoid with its luminous 

 clear area, and the deeply staining nucleus enable one to distin- 

 guish between the algae and the cell nuclei which they often 

 resemble closely to the untrained eye. 



Under the low power of the microscope a cleared or stained 

 hjndranth of E. griffini appears covered with minute dots and 

 the tentacles appear marked with transverse rows of closely 

 packed dots. This is due to the zooxanthellae which inhabit the 

 endoderm cells, and are easily seen through the thin ectoderm. 

 The endoderm cells of the entire colony — coenosarc, hydranth, 

 tentacles, and gonophores — contain these algse in numbers vary- 

 ing with the region, the number in a region seeming to depend 

 on the amount of light there available. In general, they are 

 found at the bases of the cells, forming a layer just inside the 

 mesogloea. In the gonophores they are scattered through the 

 endoderm of the spadix, and in the female are found in the e^g 

 cell and in the male at the edges of the sperm chambers. If 

 living in the ectoderm, they are few and scattered, and I have 

 been unable to identify any of the few spherical bodies of the 

 ectoderm cells as zooxanthellae. They are least numerous in 

 the coenosarc, but here there are often as many as three or 

 more to a cell, and most numerous in the tentacles where they 

 form a closely packed layer just within the peripheral walls of 

 the large disk-shaped endoderm cells. In the walls of the gas- 



