The mesenterial Filaments of the Alcyonaria. 3 



six shorter filameuts ; and it now appears that their functions are as 

 different as their striicture. Before considering the minute structure of 

 the filaments I will briefly describe their arrangement, which so far as 

 known is uniform throughout the group. We may take Alcyonmm as 

 a type. 



Fig. 4 represents from the inner or dorsal side a half-grown bud, 

 in which the dorsal filaments are not yet fully developed, but which 

 will sufficiently well show the arrangement of the orgaus. The free 

 portion of the polyp — which is still very short — shows through the 

 canals. The eight septa radiate from the Oesophagus, oe, at nearly 

 equalintervals, hearing upon their edges the mesenterial filaments. Six 

 of these are short and thick, and in older polyps are always thrown 

 into transverse folds. These are the entodermic filaments. The other 

 two, d.f., borne on the dorsal septa, are extremely long, slender and 

 perfectly straight. They extend far hack towards the hinder end of 

 the stomach-cavity, one of them being usually longer than the other, as 

 in the figure. These are the ectodermic filaments, or as I shall some- 

 times prefer to cali them, the ectodermic ha nds. 



Connccting with the polyp ou ali sides we see the network of en- 

 todermic canals through which it communicates with older polyps on 

 each side, and along the middle dorsal line are the openings of similar 

 canals not represented in the figure. This line of openings corresponds 

 with a partition wall between two mature polyi)S, over and between 

 which the younger polyp lies, and which for the sake of clearness are 

 not represented in the figure. 



The relations of this canal -system to the polyp -cavities vary 

 widely among the Alcyonaria, and in these variations lies, as I believe, 

 one of the chief causes of the differences between the bud-development 

 of the various forms. 



Methods. 

 After testing many preservative and staiuing fluids, the following 

 methods were adopted as giving, upon the whole, the best results. 

 The animals were suddenly killed by momentary immersion in a mix- 

 ture of one part strong acetic acid and two parts of a concentrated so- 

 lution of corrosive sublimate in fresh water. After being quickly washed 

 they were transferred to a concentrated solution of sublimate in fresh 

 water and left two or three hours ; the internai cavities being injected 

 with the solution, where this was possible. They were then thoroughly 

 washed in running sea- water, then in distilled water, and finally 



