212 Mr Harmer, On Gyclostomatous Polyzoa. [Oct. 26, 



several distinct types. Thus the ovicell of Crisia is practically 

 a zocecium which has become pear-shaped. Its morphological 

 value as a modified zooecium is apparent at the first glance owing 

 to the fact that it replaces an ordinary zooecium in the alternate, 

 biserial arrangement characteristic of this genus. There is no 

 sudden constriction between the proximal and the distal part of 

 the ovicell ; and this condition is probably due to the fact that as 

 the fertile polypide is never more than a mere bud, the proximal 

 part of the ovicell at no period functions as an ordinary zocecium. 



In other Cyclostomes which I have examined, there is a sudden 

 enlargement of the ovicell which corresponds with the point at 

 which it becomes, so to speak, disengaged from the series of zocecia. 

 This enlargement no doubt corresponds with the time at which 

 the embryophore begins to increase rapidly in size. 



In the Tubuliporidae, the ovicell in other respects resembles 

 that of Crisia, except that it projects less beyond the ordinary 

 zocecia. In the simpler cases, as in Diastopora suborbicularis, it 

 is still more or less pear-shaped, but in other species of the same 

 genus and in Idmonea serpens it enlarges so as to grow round the 

 neighbouring zooecia. The lobes of the ovicell then unite with 

 one another on the distal side of the zooecium round which they 

 have grown, and the zooecium is left as a column passing freely 

 through the cavity of the ovicell. 



The ovicells of Hornera probably form a third type ; while 

 those of Lichenopora are the most complicated which I have so far 

 examined. Here the ovicell no longer develops as a continuous 

 cavity ; but, although at its commencement it is merely an ordinary 

 zooecium, its extent is afterwards increased by the fusion with it 

 of the " alveoli " which are developed at the growing margin of the 

 colony. 



I hope to have the opportunity of making a more extended 

 study of the ovicells in this group of Polyzoa ; but in the mean- 

 time I wish to emphasize my opinion that it will probably not be 

 found impossible to give diagnoses of recent Cyclostomatous genera 

 with much less precision than those which are given for the genera 

 belonging to other groups of Invertebrates. 



Certain considerations with regard to the general architecture 

 of the Cyclostome colony have suggested themselves to me during 

 my recent observations 1 . 



The colony in Lichenopora is a modified funnel, whose mouth 



1 The varieties of form assumed by the colony, in various genera and species of 

 Cyclostomata, strikingly resemble those which have been described by Mr H. M. 

 Bernard in the Coral Turbinaria ("Notes, Morphological and Systematic, on the 

 Madreporarian Genus Turbinaria," Ann. Mag. Nat. Hist. (6) xv. 1895, p. 499; 

 "Catalogue of the Madreporarian Corals in the British Museum" (4to.) Vol. n. 1896). 

 It is not unlikely that the growth of a colony of Polyzoa may be regulated by laws 

 which resemble those that govern the growth of a colony of Zoantharia. 



