Anatomy of Gorgonid^. 133 



ules, capillaries, etc., and then the endothelial lining of one of 

 the primary water-vascular canals. This lining of the primary 

 canal is seen to be continuous with that of the calicle of the 

 polyp on the right. Hence it appears that the endodermal 

 investment of the axis cylinder and the endodermal lining of 

 the primary canals are homologous but not identical. It seems 

 that the calicular lining of certain polyps extends downward 

 to form the endodermal investment of the axis cylinder, and 

 the calicular lining of other polyps is continuous with the 

 endodermal lining of the primary canals. 



The fibrous portion of the axis (fig. 5, e) is not seen near 

 the growing tip of the axis (fig. 7) but is gradually inter- 

 calated between the core and the endothelial investment as 

 the branch grows older. It forms the hard "woody" outside 

 part of the axis, preponderating over the core as the axis 

 grows larger, until it forms nearly the whole of the stolon by 

 which the zoanthodeme is attached or rooted. 



In fig. 8, PI. Ill, we have a greatly magnified view of a 

 cross section of a portion of a branch which may make still 

 more evident the relation of the parts near the tip of a 

 branch: a, is the peripheral edge of the horny core; b, is a 

 portion of the investing endodermal layer of the axis. The 

 individual cells are seen to be somewhat columnar in shape 

 and are considerably larger than those constituting the lining 

 of the water-vascular canals; c, is a portion of the mesoder- 

 mal basal substance which is seen to be between the invest- 

 ment of the axis, and the cells, d, forming the endothelial lining 

 of the water-vascular canals. These latter cells are quadrate 

 with large nuclei and are considerably smaller than those con- 

 stituting the investment of the axis. 



The extreme tip of the axis is an apparently homogeneous 

 translucent mass of gelatinous consistency. It is embraced, 

 as it were, by the layer of endothelial cells which seem to 

 me to be in all probability the active agents in the secretion of 

 the axis. When a superficial view of these cells can be ob- 

 tained, they are seen to present a very close resemblance to 



