144 THE APODOUS HOLOTHUEIANS 



that layer, from the transverse muscles, irregularly scattered tubules made up of 

 muscle fibers arranged in the form of a hollow cylinder. While it is not impos- 

 sible that these are the last vestiges of the ambulacral vessels, such a view 

 seems very doubtful, as the tubules are never connected with the radial water- 

 vessels, but always with the transverse muscle. Gerould ('96) thinks that they 

 serve to support the transverse muscles and to unite more firmly the various 

 parts of the integument. Semper ('68) mistook them for rudimentary ambula- 

 cral vessels, and Danielssen and Koren ('82) and Sluiter ('80) have apparently 

 fallen into a similar error; yet it should be added that Gerould failed to find 

 similar tubules in the skin of either Molpodia antarctica or Ankyroderma jef- 

 freysii {M. oolifica jnv. ?) ; both of these species, however, have a very thin 

 skin. 



4. The radial longitudinal muscles are arranged in jiairs in most of the 

 Molpadiidae, but in Eupyrgus and Himasthlephora they form a sitigle band, as 

 in the Synaptid;e. (See Plates XII, fig. 27, and XIIT. fig. 4.) They lie on each 

 side of the radial water-vessel, but anteriorly the interradial edges of each pair 

 of muscles curve inward until they meet, while the adradial edges also merge 

 together just beneath or inside the radial vessel, and thus the two muscles form 

 a single hollow tube, which loses its lumen where it is attached to the radial 

 plate of the calcareous ring. This single muscle band may be of considerable 

 length and more or less laterally compressed ; the connection between the edge at- 

 tached to the body-wall and the edge extending into the body-cavity may be re- 

 duced to a mere sheet of connective tissue (see Perrier, :04fl and :05) or may even 

 be severed entirely. As the inner portion is connected with the calcareous ring, 

 we should have in the latter case what have been called retractor muscles. But 

 the formation of such retractors appears to be a very uncommon, if not an al- 

 together exceptional, event, possibly only occurring in certain individuals, per- 

 haps very old ones, and the presence or absence of such retractors cannot be 

 considered as having any value in taxonomy. Moreover, when strongly con- 

 tracted the ordinary longitudinal muscles may appear as though they were 

 special retractors, and in such cases the presence or absence of "retractor 

 muscles" becomes a matter of personal opinion of the investigator. Posteri- 

 orly the muscles of each radial pair become semicylindrical trunks which may 

 simply lie side by side or actually coalesce. They finally terminate in the con- 

 nective tissue about the cloacal opening. 



5. The inner epithelium is the innermost layer of the body-wall, and con- 

 sists of flat, polygonal cells, which are everywhere provided with cilia. 



BoDY-CAViTY. — The body-cavity of the Molpadiidae is very capacious and 

 shows no particular peculiarities. In those forms which have a caudal append- 

 age the body-cavity within that appendage is generally reduced, and is nearly 

 filled up by the strands of muscular and connective tissue connecting the ali- 



