1908] Robertson. — Incrusiing Bryozoa. 271 



This earlier stage of colonial formation has not been hitherto 

 described and the appearance of the zooecia in the two stages of 

 growth is so different that it is difficnlt to believe them stages of 

 one and the same species unless the transition steps are seen. 

 Many colonies showing these transitions were obtained at La 

 Jolla during July of 1905. At that time of year the larvae are 

 settling, and new, fresh colonies are abundant. The youthful 

 condition with branched spines (fig. 23) continues for about ten 

 or twenty rows when the adult stage (fig. 22) with elongated 

 zooecia and unbranched spines is assumed.^ 



Memhranipora villosa is found at various localities from 

 Puget Sound to San Diego, and is probably very abundant in 

 the belt of kelp some distance out from shore on our southern 

 coast. 



This species is especially interesting from the fact that it is 

 one of the few members of the genus INIembranipora whose meta- 

 morphosis from a Cijphonautes larva has been observed. Cypho- 

 nautes was originally described by Ehrenberg ('33) as a rotifer, 

 and was re-described by Johannes Muller ( '54) as a larva of an 

 annelid. Later, in 1857, Semper declared his belief in its mollus- 

 can affinity, and it was not until Schneider in 1869 watched its 

 metamorphosis, that the true relationship of this interesting or- 

 ganism became known. Schneider found it extremely difficult to 

 keep Cyplionmdcs under artificial conditions. Indeed, he found 

 it impossible to do so unless the larvae were matured at the time of 

 capture and just about to settle. It was by a rare piece of good 

 fortune that the relation between the Cyphonautcs common on 

 the Pacific coast and its sessile form was discovered. On a kelp 

 brought into the laboratory at La Jolla during the summer of '05, 

 were found multitudes of minutes bodies covering every portion 



1 Since writing the above, I have had the opportunity to examine 

 figures of d'Orbigny's Flustra Isabelliana, and consider the resemblance be- 

 tween it and the aclult stage of M. villosa Hincks as most striking. There 

 is a possibility that the two species are identical, in which case d'Orbigny's 

 name should liave the preference. Flustra Isabelliana is evidently a Mem- 

 branipora and like M. villosa, has spines at the anterior margin and has its 

 front wall ornamented by two rows of minute spinules (petites pilosities). 

 In view of the full description of the stages of growth of M. villosa here 

 given, and until more knowledge can be obtained of the development of F. 

 Isabelliana, or until direct comparison of M. villosa with the South American 

 species can be made, it is not considered best to do more at this time than 

 to call attention to the similarity and possible identity of the two species. 



