82 BULLETIN OF THE 



will be noticed that the flat faces which make up the sides of the animal below 

 the swimming-bells are not formed, as in A. elegans, Ijy the upper surfaces of the 

 covering scales, but by planes of the same at right angles to the upper and 

 lower surfaces of the bract. These faces are produced by the great thickness of 

 tlie scale. Its breadth is simply the thickness of the distal edge of the scale. 

 The upper surface of the lowest covering-scale is flat, and fits closely to the 

 lower surface of that bract which is immediately above it in the series, and so 

 on throughout tlie whole length of the polyp-stem to the lowest nectocalyx. 

 The thickened border of the bract does not present, when seen from the side, a 

 single continuous plane surface, but is made up of three or four slightly con- 

 cave furrows, separated Ijy ridges, which extend at right angles t(5 the upper 

 face of the scale, in the direction of the length of the stem. Both the swim- 

 ming-bells and the covering-scales are infested with Distomcc. The appendages 

 to the polyp-stem all arise from one side of the axis and hang dov.nward in such 

 a way that when the axis is extended longitudinally the free extremities of the 

 polypites slightly protrude beyond the covering-scales. The polypites are 

 more highly colored than those of Arjahna. No tasters were observed. The 

 tentacles resemble in character and origin those o^ A. elegans Y. Each ten- 

 tacular knob has a coiled sacculus, a well-developed involucrum within which 

 it can be drawn, and two lateral terminal filaments, one on each side of a 

 median vesicle. The distal extremities of the lateral filaments are slightly 

 enlarged, and colored with reddish pigment. 



Note. — It may be found, when older larvae of A. papillosum F. are studied, 

 that it is the same as A. Okenii Esch, 



Rhizophysa Eysenhardtii (?) Geg. 



A single specimen of E. Eysenhardtii (?) was taken in Castle Harbor. The 

 species is well marked, and can easily be distinguished from R. filiformis* by 

 the absence of tentacular knobs on the tentacular filaments. When first taken 

 from the water, the tentacles cling with the greatest pertinacity to whatever 

 foreign body they touch. R. filiformis is also said to grasp any adjacent object 

 in the same way ; but those which I have studied do not fasten the tentacles 

 with the same persistency as R. Eysenhardtii. 



* The anatomy of tlie above species of Bermuda Rhizophysa resembles closely 

 that of B. plawstoma Per. et Les., althoush in the figures of this species no side 

 branches to the tentacles are represented. It also agrees closely in form with a 

 species of Rhizojihysa described by Huxley, from the Indian Ocean. In the Ber- 

 muda species no sexual clusters were found at the base of the feeding polyps, as 

 mentioned by Huxley in his species. 



Several specimens of R. filiformis were found at Bermuda; one of these meas- 

 ured over three feet in length. In this specimen the sexual bells were very large, 

 and resemble very closely the sexual bodies of Physalia. 



