114 THE VOYAGE OF H.M.S. CHALLENGER. 



the projection is really due to an increase in thickness of the dorsal body -wall itself. At 

 the same time, we seem to have here a more pronounced example than is elsewhere seen 

 of the tendency of the mouth to become removed from its terminal position. 



As noted above, Elasipoda are found in which the mouth is placed at the end of a 

 narrow neck-like part, which is bent downwards so as to form an angle with the ventral 

 surface; this is the case in Peniagone vitrea, Elpidia ivillemoesi, &c. In a very few 

 forms among the Elasipoda, as, for instance, in the genera Deima and Benthodytes 

 selenkiana, &c, the mouth and the tentacles are capable of being entirely retracted within 

 the body. All the individuals of the genus Deima 1 have had at my disposal are espe- 

 cially remarkable for the fact that the tentacles are always enclosed within the cavity, 

 which forms the most anterior part of the alimentary canal and communicates with the 

 exterior by an aperture which seems capable of being entirely closed (PL XLIII. fig. 3). 

 This aperture lies in the centre of a radially-wrinkled disk, and is surrounded by a single 

 crown of small papillae (PI. XLIII. fig. 2), the importance of which will be discussed further 

 on. I do not think it probable that Deima, in which any solid oral disk seems absent, 

 should not be able to extend the disk and its tentacles outside the body. It is unnecessary 

 to state that this capacity of retracting the mouth and tentacles within the body is not pecu- 

 liarly characteristic of a few forms in the order Elasipoda ; for numerous species among previ- 

 ously known Holothurioidea, and particularly the Dendrochirotas, also possess this power. 



In the Pedata and the Apoda the tentacles are generally arranged in a single row 

 round the mouth, and it is well known that only in a very few cases, viz., the genus 

 Phyllophorus, Gvube ,av.dSyna2)ta bifaria, Semper, exceptions are found where the tentacles 

 are disposed in two circles, an outer and an inner one. The tentacles, though usually of 

 equal size, are sometimes unequal, as in Echinocucumis, Sars, and sometimes, as is the 

 case in a great number of the Dendrochirotse, a couple of the ventral ones is considerably 

 smaller than the others. It is also well known that the genus Tliyonidium, Diib. and 

 Kor., is characterised by the possession of five pairs of large alternating with five pairs of 

 smaller tentacles, and that Orcula, Troschel, carries from ten to twenty tentacles, of 

 which five are always smaller than the rest. The tentacles vary highly in shape, and 

 are grouped by Semper in the following manner : — tentacula peltata in the Aspido- 

 chirotae ; arborescentia in the Dendrochirotae ; peltata and pinnata in the Molpadidae ; 

 digitata, pinnata, and peltato-digitata in the Synaptidas. In Haplodactyla, Grube, 

 belonging to the Molpadidae, the tentacles present their simplest conformation, and 

 deviate from the common type in being unbranched and without any enlarged terminal 

 part, thus plainly showing that the tentacles in general are to be regarded as simply 

 modified pedicels. The tentacles in the orders Apoda and Pedata varying in number 

 from ten to twenty-five, are in general a multiple of five, but notwithstanding this 

 exceptional forms are not infrequently met with, which possess twelve, thirteen, 

 eighteen, or nineteen tentacles. 



