DEVELOPMENT OF AN APODOUS HOLOTHURIAN 
507 
distinct one, markedly separate from all other animal types, but 
each of its component classes is equally distinct, truly annectant 
living forms being practically unknown. The facts from which the 
phylogeny of holothurians is to be deduced can be set forth more 
clearly and the hypotheses which their study has led me to adopt 
can be made more comprehensible if we first consider the inter- 
relationships of the orders of holothurians, and then discuss the 
relationship of the class to other echinoderms. 
The relationship of the orders of Holothurioidea to each other 
Ostergren, in a recent paper COT) has set forth quite fully his 
views on holothurian interrelationships, and as they differ in some 
important particulars from those held by Ludwig and others, they 
will serve admirably as a basis for this discussion. He recognizes 
five orders of holothurians (Dendrochirota, Aspidochirota, Mol- 
padonia, Elasipoda, Apoda) and I shall, for convenience, here use 
his names and accept these orders without further discussions 
other than to say that I am not sure the names are in all cases 
tenable. Ostergren bases his classification on his own extensive 
morphological and physiological studies (though of course giving 
due weight to the work of other investigators), and he lays partic- 
ular emphasis on the functions of organs and their relation to 
the habits of the animal. He describes his ancestral form • 
(Stammholothurie) as having a soft body-wall strengthened by 
scattered calcareous plates; creeping about by the contractions 
of the body musculature; and feeding on the organic matter in 
mud C'ernahrte sich von Schlamm It had "twenty (or ten?) 
short tentacles, without ampullae, five radial canals and a num- 
ber of scattered pedicels. The posterior part of the gut (cloaca) 
served as a respiratory organ, but there were probably no 'Svater- 
lungs" developed. From such an ancestor, Ostergren derives his 
five classes of holothurians, finding its nearest living represen- 
tatives among the Elasipoda ; such genera as Capheira, for exam- 
ple, differing only a little from this ''hj^pothetische Stammform." 
It is not necessary to criticize this theory in detail here, but there 
are three general criticisms which seem to me to seriously affect 
