2:î0 



greatly (fig. ß). In some cells the nucleus is still in its primitive position, 

 in contact with the blepharoplast; it is no longer pear-shaped, however, 

 but has a spherical form. In other cells the nucleus is halfway between 

 the apex and the base of the cell, and in others again it has taken up 

 its final position at the base of the cell. Hence in Clathrinidae the 

 assumption by the nucleus of an a[)ical position as a preliminary to di- 

 vision, is simply a return to the jiosition which it occupied in the larva, 

 and which may be reasonably regarded, I venture to think, as the }>hylo- 

 genetically primitive position. 



In spite, therefore, of the criticisms of Hammer and Dendy, my 

 belief in the systematic and phylogenetic value of the structure of the 

 collar-cells in calcareous sponges remains unshaken; with, however, 

 the correction, that the point to be taken into account is not merely the 

 exact position of the nucleus in the cell, as I stated formerly, but the 

 relation of the flagellum to the nucleus, which is the factor determining 

 the situation of the nucleus in the collar-cell. The differences between 

 the two types of collar-cells in Calcarea show an interesting jìarallelism 

 with the two species oî 2Iai<tfgiiia described by Goldschmidt^'. In il/. 

 setosa the flagellum arises directly from the nucleus, as in Leitcosolenia 

 and S/jcoìì; in M. vitrea ^ however, the flagellum arises quite indepen- 

 dently of the nucleus, which may be situated at any point of the cell. 

 It may be supposed that there was a remote period in the history of the 

 world when the Oly/iflnis-ïonn represented the highest point reached 

 in the evolution of the calcareous sponges, and was a genus with at least 

 two species, one of which, in addition to other peculiarities, had the 

 nucleus connected with the tlagellum, like Mastigiiia setosa^ while in 

 til»' other species the flagellum arose quite distinct from the flagellum, 

 as in d/. vitrea. Such a character, in the case of the sponge, can have 

 but little importance in relation to the conditions of life or the struggle 

 for existence, and hence has persisted unchanged to the present day, 

 and is one of the characters which, in my opinion indicates the deepest 

 phylogenetic divergence in the pedigree of the calcareous sponges. 



In conclusion I will give a brief diagnosis of the two families of 

 the Homocoela, in correction of that given by me in 1900. 



Class Calcarea. Sponges Avith the skeleton composed of calcite, in 

 the form of spicules either monaxon, triradiate, or quadriradiate in form. 



Grade 1. Homocoela. Calcarea with the gastral layer of collar- 

 cells continuous, not forming separate flagellated chambers. 



Fajnily 1. Clathrinidae. Oscular tubes generally short, arising 



3 Lebeiiss^osclnchte der Mastigamöbeii. Arch, rrotistoiikuiule. Suppl. 1. 1907. 

 p. 88-168. Tat". V— IX. 20 ïext-tiçrg. 



