38 ECHINODERMATA. 



panded to their utmost dimensions, are observed sweeping successively 

 through the water. One of them then curving its extremity inwards, 

 the central orifice dilates to receive it, wherein, after its size diminish- 

 ing, it is absorbed almost to the root. While thus occupied, one of the 

 antennular organs likewise curves into the orifice, and both are evidently 

 engaged below. Next, the arborescent tentaculum stiU reduced in size, 

 returns to expand anew, and the antennula follows it. Now, another 

 tentaculum contracts and curves to penetrate the orifice, left vacant by 

 the former, wherein it is also speedily followed by the second antennula. 

 Both are employed within, and their functions being discharged, both 

 return to resume their original figure and dimensions. 



Before indulging any conjectures on the real object to be accom- 

 plished by such remarkalile evolutions, a general view of the parts en- 

 gaged in them may be taken, by w^liich means more skilful physiologists 

 will be aided in determining their purpose. 



^^liat I here denominate the cijlinder or case, which bearing the ten- 

 tacula is extruded beyond the body, is in itself a compound organization 

 of considerable complexity. 



When the tentacula are parti}' contracted during quiescence of the 

 animal, the origin of this cylinder as the whole issues from the body ap- 

 pears, Plate VIII,, fig. 2, c. If the case or cylinder be entirely detached 

 from the animal, several bony or shelly receptacles are seen around its cir- 

 cumference, into which the tentacula, b, are retracted, and the whole 

 then withdrawn within the body, which is closed over it. 



The lower part of this apparatus is united to the intestine, and here 

 also appears the transparent vessel or Ampulla Poliana. 



On narrower inspection the channels seem partly osseous, and partly 

 cartilaginous. They are white circumferential sheaths receiving the in- 

 active tentacula. 



The central orifice is susceptible of gi-eat dilatation ; it communicates 

 through the oesophagus with the intestine, towards one side of which, 

 connected with the cyhnder, is the transparent vessel, Plate XI., fig. 7, 

 a. The eight arborescent tentacula issue from the higher circumference 

 of the case or cylinder at regular intervals, unless at one point, where 



