14 THE APODOUS HOLOTHURIANS 



the Molpadiidre as a distinct suborder, with the usual single family and six 

 genera. Their classification of the Paractinopodida is most confusing, for no sub- 

 ordinate groups are given save genera, of which 14 are recognized, two being 

 known only from fossil remains. Moreover, although several references are 

 made to 6'stergren's work, these only add to the confusion, for blunders are 

 made in attempting to indicate the subfamilies proposed by him, and an ob- 

 vious slip of the pen concerning the Chiridotinae makes matters still worse. 

 MacBride ( :06), ignoring Ludwig's important embryological work, divides the 

 holothurians into six orders, of which the last two are the MOLPADIIDA and 

 SYNAPTIDA, each containing a single family and six genera. Such a classifica- 

 tion is certainly no contribution to our knowledge of the group! Fisher ( :07) 

 recognizes the classifications used by Ludwig and by Ostergren, at the same 

 time calling attention to some points of nomenclature in which the latter seems 

 to be in error. He adds a new genus (Opheodesoma) which is practically sec- 

 tion B of Ostergren 's Euapta, and gives a very full account of the type species. 

 He also describes four other new forms of Synaptidae. 



In arranging any scheme of classification, the first point to be determined 

 is the relative value of characters upon which to base specific, generic, and 

 family differences. In the holothurians, the characters which best determine 

 family limits are the presence or absence of true retractor muscles and 

 respiratory trees, the number and form of the tentacles, and the presence or 

 absence of other ambulacral appendages. Since the ambulacral appendages, ex- 

 cept tentacles, vary greatly in number, size, and arrangement, even in a single 

 species, the characteristic features of the internal anatomy carry more weight. 

 The group of apodous holothurians therefore cannot be accepted as a natural 

 group, especially since the genera Himasthlephora and Gephyrothuria form an 

 obvious connecting link with the pedate forms. The absence of respiratory 

 trees is, however, a constant and remarkable character, and, combined as it is 

 with the absence of pedicels and papilla? and the presence of unbroken circular 

 muscles, marks out the Synaptidas as a well-defined family, aside from its 

 ordinal characters. The other apodous holothurians are less susceptible of 

 satisfactory definition, aside from the lack of pedicels, but the shape of the body 

 usually terminating in a more or less evident caudal portion, the peculiar, short 

 tentacles, combined with the presence of respiratory trees, are really quite char- 

 acteristic for the Molpadiidae. These two families may therefore be accepted 

 as including all known apodous holothurians (except the remarkable and unique 

 Pelagothuriida?), but there is no intention of implying any necessarily close 

 connection between them. Perrier's ( :03) reestablishment of the group Apodes 

 seems to be an attempt to base the classification on an obvious, external char- 

 acter, because it is convenient and easy to note, rather than an effort to reach 

 a natural arrangement. No new arguments are advanced in favor of the 

 Apodes, and the evidence is certainly against their standing as a subclass. 



