OYICELLS OF SOME LICIIKNOPOR^. 



281 



recent writer, Mr. Th. Marsson*, revives Discocavea and Defrancia 

 for LicJienopora, though his Defrancia contains species which, 

 according to what I have said above, cannot be placed with Licheno- 

 pora, and his Discocavea is entirely based upon the radii being 

 uniserial ; but this cannot be looked upon as a generic character, 

 and even in specific determination must not be used with too much 

 confidence, as there are some species where the rays are in parts 

 uniserial, in others biserial. 



In some species there are spines arising from various parts of 

 the surface of the zoarium, but these do not appear to furnish 

 constant characters ; and, further, I have on one or two occasions 

 called attention to the great variation to which the shape of the 

 peristome is subject in different parts of the same colony ; on 

 the other hand, which side of the peristome is prolonged seems 

 to be of considerable diagnostic importance. 



The internal "rays" in the zocecial tubes and caneelli should 

 be more studied, and I should consider the denticulation of recent 

 Heteropora cervicornis, d'Orb. (Journ. R. Micr. Soc. vol. ii. p. 392, 

 pi. XV. figs. 9-11), Liclienopora hullata, MacGr., L. echinata, 

 MacGr., and L.pristis, MacG., as the equivalents of the rays with 

 knobs in Entalophora intricaria, B., Lichenopora radiata, Aud., 

 L. reticulata, MacG., L. Soldsworthii, B., &c. 



Further, the position of the connecting-pores in the interior 

 walls of the zocecia should be examined. 



Having seen how many of the characters used for diagnosis 

 are variable, it is quite clear that numerous species made on ac- 

 count of a difference in some unimportant character can never be 

 recognized again, making a better acquaintance with this genus 

 very desirable ; audit seems that a knowledge of the ovicells may 

 often assist us in more fully grasping the amount of variation 

 caused by the conditions under which the colony grew, and may 

 show the amount of variation in diff'erent parts of a colony. AYe 

 are as yet unable to say how far the ovicells may be trusted for 

 specific determination ; but there is no doubt that they are for this 

 purpose of value. Also as some species have been described with- 

 out ovicells and some with, in cases where tlie author has not 

 appreciated whether they were present or not, exact studies in both 

 conditions are required. In so recent a book as Busk's ' Catalogue 

 of the British Museum Polyzoa,' pt. iii., where 14 species are 

 described, the ovicells are never mentioned, nor are they in the 

 * Bryozna der weissen Schreibkreide der Tnpel Riigen, 1887. 



