4 IMr. A. W. Waters on 



Idmonccp arc represented in tlic Bartonian either by the 



same s])ecies or allies, and ovieells of nearly all are now known 



in the Meiliteriaiiean. It will thns evidently be a slow and 



gradnal process checking the importance of ovicell characters. 



1 * have repeated more tlian once my opinion that 



characters of great value in one gronp or family are almost 



useless in the next, and all attempts at fixing certain 



characters of A.l importance, others secondary, and so on, 



may lead to no result, and we have to see what cliaracters 



occur together in various groups. Lcvinscn has expressed 



the same idea in other words. Although ii is dangerous to 



sav a certain character ought to have the first place, another 



the second, of course it is not meant that physiological 



reasons must be ignored — however, is not the value of a 



character quite as much a question of when a separation 



based on it took place ? 



Cunu and liassler t say, "We repeatedly have to remark 

 tliat tlie zoarial form is of no value for generic classification." 

 I certainly cannot go as far as this, and it brings us up against 

 a most important point that requires settlement. It has been 

 established that, in the classification of the Cheilostomata, 

 zocecial characters are more important than zoarial, but in the 

 C'vclostomata the classification has been entirely based on 

 zoarial characters. It is, however, often difhcult to decide 

 what is zooecial, what zoarial. Tn the Cheilostomata, as a 

 rule, each zocccium is only in groMth-conuection with its 

 immediate neighbours, and there may be exactly similar 

 zooccia adnatc or erect, or })laced back to back — as, for 

 exami)le, in Stcganojxjrella and several other genera, — and it 

 took some time before the last generation could agree to 

 their being placed together. Zoarial characters should, 

 however, be stated in both of the suborders. 



In the Cyclostomata the young zooecia grow under the old 

 zooecia, and may commence a considerable distance from the 

 end of the zocccium, and the way they are grouped together 

 seems to depend almost as much on the zooecia as on the 

 zoaria. In a section o{ Entulophora and several other genera 

 the small early Z(joc('ia are seen in the centre, so that they are 

 not definitely under the older zooecia as in Idmonea, but the 

 princi|)le is the same. Smitt % has j)assed a fine hair through 

 one of the basal cells of Idmonea atlanllca for the distance 



« " Bry. fromj^anzibar," I'roc. Zool. Soc. 1913, p. 460, note. 



+ Loc. cit. i>. 7oi*. 



t Krit. Ibit, otSkaud. llafs. Biy. p. 441 (18G(i). 



