ZOOLOGY AND BOTANY, MICBOSCOPY, ETC. 393 



proportion of these forms, is shown not to hold for a number of 

 these very species, and hence is probably untrue of all. Again, the 

 "dendroid," "retiform," and "vesicular" modifications of the canal- 

 system described by Hackel in various Leucones are in reality not 

 present there. Polejaeff agrees with Yosmaer in regarding the radial 

 tube of the Sycones as simply a form of flagellated chamber, and not, 

 like Hackel, a "person" equivalent to an individual Ascon ; the 

 Ohjnthus is simply a common form, out of which may be developed, 

 by different processes, either a Sycon or an Ascon, the mesoderm in 

 the former being more developed than in the latter, and thus giving 

 capacity for a more differentiated canal-system. From their minute 

 characters, in combination with Von Lendenfeld's observations on 

 Aplysinidce, he is led to ascribe the function of receiving food to both 

 the ecto- and entodermal pavement-cells of the canals. 



The mutual relations of Leucones and Sycones are elucidated by 

 the discovery of a Sycon (Amphoriscus elongatus) in which the radial 

 tubes, instead of opening directly and singly into the cloaca, debouch 

 by groups of three, four, or more, into secondary chambers which in 

 turn open into the common cloaca ; the secondary chamber has only 

 to be exaggerated to form the excretory canal which leads from the 

 flagellated chambers of a Leucon ; the skeleton of the radial tube 

 of this and other primitive Sycones is non-articulated, i. e. does not 

 form a succession of septa in the parenchyma surrounding the tube, 

 and thus affords another point of connection with the similarly circum- 

 stanced Leucones. 



The canal system, as thus elucidated, is taken as the basis of the 

 modified classification, which runs thus : — 



Class Calcaeea. Order 1. Homocoela. Family 1. Asconidm. 



„ 2. Heteroccela. „ 2. Syconidce. 



„ 3. Leuconidce. 



„ 4. TeicJionidce. 



(The last family is Carter's, whose reasons for establishing it are 

 adopted.) The genera of the class are entirely reconsidered on the 

 basis of allowing to all the elements of the organization a share in the 

 systematic distinction, and the law of priority in the nomenclature, set 

 aside by Hackel, is reasserted. The genera, as revised, are — 



Fam. Asconidce : Leucosolenia, provisionally adopted as the only 

 genus. 



Fam. Syconidce : Sycon, Grantia, TJte, Heteropegma n. gen., Ampho- 

 riscus, Anamixilla n. gen. The chief distinctions employed are the 

 articulation or non-articulation of the tubar skeleton, the mutual 

 independence or not of the tubes, and the form of the spicules. Hetero- 

 pegma differs from Grantia in having a cortex composed of spicules of 

 a different size from those of the parenchyma. Anamixilla has no 

 special tubar skeleton. 



Fam. Leuconidce : Leucilla, Leuconia, Leucetta, Pericharax n. gen. 

 They are based on the form of the ciliated chambers, the arrangement 

 of the spicules, the presence or absence of a cortex, or (^Pericharax) 



