STAR-FISHES. 153 



form is in both cases that of a star of tive or more rays. In the star-fishes there is no 

 such well-marked distinction between the disc and the arms as there is in an ojihiu- 

 roid, for the stomach, and the ovaries or spermarics, run into the arms. Along 

 the underside of each arm runs a deep ambulacral furrow, from the depths of which 

 project the ambulacral feet, which are provided at their ends with suckers, by means 

 of which the animal moves. At the base of each sucker-foot within the arm is a 

 vesicle or ampulla, connected with the radial or ambulacral canal of the water system. 

 This arrangement seems very different from the solidly j)lated arm of a serpent-star, 

 with its enclosed row of vertebral ossicles, and its tentacle-like feet un])rovided with 

 sucking discs or with vesicles at their base, yet it is demonstrable that the halves of 

 tile vertebra.' of an ophiuran are to be considered identical in their nature with the 

 cak'areous ambulacral jjieces which form the deeper ]wrts of the sides of the ambula- 

 cral furrows of a star-fish. The development of the interambulacral areas varies 

 greatly, so that while some star-fishes, as Brisinc/a, closely approach the serpent-stars 

 in the distinctness of the arms, others have the angles between the arms more or less 

 filled up by the development of the interambulacra, which in many forms are so exten- 

 sive that the creature is nearly or quite a jientagon, and the existence of arms can only 

 be traced by the lines of the ambulacral furrows on the oral aspect. 



Star-fishes have a very comjilex skeleton, varying greatly in the different groui)s, 

 but always consisting of ])lates or thick rods composed of a dense calcareous net-work. 

 The sides of the ambulacral furrows are bounded by two rows of regtdarly-placed and 

 similar ambulacral ossicles, which abut against each other like the two sides of a gable, 

 while at their outer ends they abut against a row of short, thick, adandjulacral or 

 interambulacral ossicles, which form the borders of the furrow. These are the 

 constant parts of the skeleton, but the sides of the arms are in many cases enclosed by 

 plates of considerable size, known as marginal plates. The net-work of rods which 

 forms the support of the upper ])art of the arms and disc is very variously develojjed 

 in the different families and genera. The ])ores through which the sucker-feet ])ass 

 are each formed by two ambulacral plates, one half of the pore consisting of a notch 

 upon the outer side of one jilate, while the other half is on the inner side of the next 

 plate. Around the mouth the ambulacral and interambulacral ])lates are modi- 

 lied so as to form a jtentagon or polygon. The pieces of this calcari'ous ring that cor- 

 respond to the interambulacral plates of the arms are furnished with strong ]>apilla' or 

 spines which form a sort of imperfect dentary a]>paratus. The siiines which jiroject 

 from the surface, both above and below, and which usually form regular 

 rows on each side of the ambulacral furrows, are more or less movably 

 united with the skeleton, but are without the regular joints found in 

 the sea-urchins. Pedicellariie are jiresent. Above, that is to say, in- 

 ternal to the dentary jiajiillre, the gullet opens into a wide stomach, 

 the oral or cardiac part of which is produced into sacs that sometimes 

 extend into the cavity of the arm to which they corresjiond. On the 

 aboral side of these sacs, the alimentary canal again narrows somewhat, 

 and widens again into a pyloric sac, the angles of which are again ]iro- 

 longed into tubes which run alon"' the aboral side of the ravs- The fi«- i''-'- — ^^i"."? 

 upper or jiyloric part of the stomach is attached by a mesenteric mem- iio,ii,eii.uia at 

 brane to the aboral wall of the body, while the cardiac sacs are similarly 

 connected with the ossicles or calcareous plates around the mouth. Beyond the jiyloric 

 sac is a short tubular intestine, which encls in a more or less minute anal pore, apparently 



