56 



species, although its tylostyles are considerably larger (PI. XII, fig. i a—t). If we follow Dendy 

 in taking B.M. 17 (PI. XI, fig. 1), B.M. 14 (PI. XI, fig. 6), B.M. 19, B.M. 36, B.M. 24 

 (PI. XI, fig. 24), B.M. 41 (PI. XI, fig. 5) as so many specimens of Spirastrella vagabunda Rdl. 

 and uniting with them Spirastrella congenera B.M. 20 (PI- XII, fig. 1), Spirastrella punctulata 

 B.M. 28 and Spirastrella solida B.M. 26 we have no reason to exclude other "species" on 

 account of the sizé of the tylostyli. 



Of course one might forward the opinion, that Dendy was wrong in uniting them and 

 that the specimens with slender tylostyli belong to one, those with stout ones to another species. 

 Against this may be said, that the more specimens one examines, the more difficult it becomes 

 to separate them, because one fincis innumerable transitions. In order to have an apercu I 

 divided the specimens at my disposal into three groups; I united those the tylostyles of which 

 were stout in another. The remainder forms a group the specimens of which either possess 

 both slender and stout spicules or tylostyles which are neither especially slender, nor very stout. 

 This group is, consequently, composed of transitional forms. Arranged in this way we come 

 to the followins' list. 



