234 CTENOPHOR.IE. Part II. 



elongated cavity, and tlie vertical tubes extending along the base of the tentacidar 

 apparatus, constitute, indeed, most complicated pieces of niachineiy, in which hydro- 

 static jjower, elastic levers, and the contractions of the motory cells, give rise to 

 highly complicated combinations and most diversified phenomena. 



In the first place, the cavity itself from which each of the two tentacles issues 

 (PI. IP. Fig. 15yy\ Fkjs. 22 and 23/) is a wide, elongated, fusiform sac, the rounded 

 extremity of which is turned towards the actinal pole and bent obliquely sideways, 

 so that its flat base is turned towards the vertical axis, and its open extremity 

 towards the abactinal pole and sideways. In this cavity, to wliich the surrounding 

 water has free access through the opening y, the tentacle with its complicated 

 base is attached by a broad surface to the inner side of the sac. And though 

 the central chymiferous cavity communicates freely, through the interambulacral 

 tubes, with the base of the tentacular apparatus, there is no free passage from 

 one of the cavities into the other. The fluid which is injected into the tentacular 

 tubes runs back through the same channels into the main trunk, and the water 

 which fills the cavity of the tentacular apparatus empties through the same opening 

 by which it is introduced. In a state of dilatation, Avater penetrates from without 

 into the tentacular sac, and diluted ch^ane is injected from within into the tentacular 

 tubes ; and in a state of contraction, the chymiferous tubes are emptied at the 

 same time that the water is pressed out. During these alternate contractions and 

 dilatations, the tentacle itself may be coiled ujj in the cavity or drawn out at 

 full length, though in the most dilated state the threads generally hang out. There 

 seems to be also an antagonism, in a middle state of dilatation, between the filling 

 of the tentacular chymiferous tubes and the protrusion of the tentacles themselves. 

 But I was mistaken formerly, when supposing that the chymiferous tubes penetrate 

 into the Ijasal dilatation of the tentacles : ^ they only extend along their basal disk. 

 The filling of these tubes may, however, cause the whole tentacular apparatus to 

 protrude into the sac to which it is attached. 



Nearly two thirds of the length and Ijreadth of the proximate side of the 

 actinal or closed end {^Flg. 87 /" to j'^) of the tentacular socket is occupied ))y an 

 oblong disk (PI. IP Fkj. 15 and Fi,j. 87 ['> /)" f l'>"', T )' /' f), fi'om the mid-length 

 of Avhich the tentacle (^ j' / ) arises. The distal side {Fit/. 15 ';>) of the disk, or 

 that wdiich faces toward the periphery of the body, is convex, with a shallow 

 furrow {Fig. 15 ;•), extending from the base {g) of the tentacle to the actinal end {y") 

 of the disk; and the pi'oximate side [Fig. 87 j'"'), or that which faces toward the axis 

 of tlie body, is a plane, innnediatel}- l)encath whose surface and next to the edge 



' When compai-ing tin? platos of my paiier in sluinld \m: maili- lor this mistake, which is especialh' 

 the Memoirs of the American Academy, allowance nuticeablo in Firjs. 1 and 2 of PI. IV. 



