20 Proceedings of the Royal Irish Academy. 



(c) Retraction. — That the polyps are capable of great retraction is a fact 

 which is of the utmost importance in speciSc determination. The manner 

 in which this is accomplished is very simple. The tentacles are first 

 infolded, the eight lobes of the verruca close over them, and then the whole 

 is withdrawn into the coenenchyma. The lougitudiual section (fig. 53) of 

 Scirpearia flagellum shows the attachment of the strong I'etractor muscles 

 which accomplish this ; and fig. 10 of Juncella juncea shows the polyps 

 completely embedded in the coenenchyma. 



It would be difficult to imagine that such extreme differences as that 

 given in figs. 49 and 51 of S. Jlugellum could occur in one species, were it 

 not for the fact that as great difierences actually occur in one individual 

 colony, e.g. figs. 77, 78, 79, and 80 of Scirpearia furcata. 



This has been discussed in detail under the different species, so that it is 

 necessary here to refer only to the actual existence of such a phenomenon. 



[d) Bistrihidio^i. — The distribution of the polyps has been used as a basis 

 for generic diagnosis, so that it is essential to study this character in detail 

 and see to what extent the various distinctions can be said to obtain. Wright 

 and Studer (L.), p. Ixv, in defining their new genus Scirpearella, make the 

 following statement : — " The polyps are arranged in rows or spii'als, retractile 

 with more or less prominent verrucae," thus separating it from Scirpearia, 

 which they describe as having " the polyps seated in two longitudinal rows 

 on each side of the stem." 



Hickson, in discussing these, says : — " The genera Scirpearia and Scii-- 

 pearella, however, appear to me to be still good genera. The arrangement of 

 prominent verrucae in two rows in the former genus and in a spiral manner 

 in the latter, combined with other characters, renders them relatively easy of 

 identification." 



Let us for the present disregard the question of spicules, and consider the 

 group as a whole with regard to this character. 



It must be borne in mind that since the publication of the work of these 

 authors, the species Juncella trilincata Thomson and Henderson has been 

 established ; and the present memoir- contains another new species, namely, 

 Scirpearia quadrilineata. The result of this is that unless the distinction 

 drawn between these two genera is modified, these two species would 

 necessitate the establishing of two new genera to include them. 



The first problem before us then is : — What is the factor underlying t/ie 

 distribution of the verrucae ? The answer to this question — namely, the number 

 of main longitudinal canals — has already been discussed. 



The only e.xception to this rule is what may be termed " the low verruca- 



