REPORT ON THE PENNATULIDA. 225 



completely in tlie sea Viottom, this pai-t of the explanation seems to us 

 entirely satisfactory. 



Concerning the fracture of the upper end, however, the case is 

 different. The cause here must be an altogether different and inde- 

 pendent one. It is almost inconceivable that any influence at the time 

 of capture could invariably break off the tops of the specimens. Neither 

 the dredge, nor the i-ope,nor the tangles, could, so far as we can see, possibly 

 effect this fracture : their tendency would always be, as we have just 

 shown, to break the stem at its point of emergence from the ground. 

 We are, therefore, driven to the conclusion that the upper fracture is 

 not effected at the time of capture, but that Virgularia, while living 

 undisturbed at the bottom of the sea, has already lost its top. This is 

 confirmed by an observation of Darwin,* who describes the Vircjularia 

 {Stijlatula Dancinii of Kijlliker) seen by him living on the shores of 

 Patagonia as truncated at the upper end. 



Having thus narrowed our problem and defined its limits more 

 precisely, we have now to determine, if possible, what are the causes 

 which, acting normally during the life of a Virgularia, and quite 

 independently of any influence exerted by man, lead to the almost 

 invariable truncation of its upper end. 



The first explanation that suggested itself to us was, that in the 

 ordinary course of growth the top, after attaining its full development, 

 dies, withers up, and drops off, and in this way causes the truncation. 

 This is at first sight an attractive theory, and accords well with the fact 

 that the leaves at the bottom of the stalk are always small and 

 immature, and gradually increase in size and development as we pass 

 upwards ; i.e., that the development of leaves appears to proceed from 

 below upwards. 



However, closer examination reveals fatal objections to this view. 

 In the first place the actual upper ends of the specimens as dredged, 

 show no sign whatever of disease, or of being about to perish. On 

 the contrary, in all the specimens examined the rachis is perfectly 

 healthy right up to the top. Secondly, the truncation does not occur 

 always at or about the same spot in different specimens, but at 

 various points of their length. In some {cf. Fig. 1) it occurs above 

 the largest leaves, in others some way below them, and in others again 

 about the position of the largest leaves ; i.e., the widest part of the 

 rachis. This variability is certainly not what we should expect were 

 the truncation due to death from natural causes. Thirdly, even 

 though it were true that the polypes after living a certain time 

 died and withered away at the top of the rachis, this would not 

 account for the stem Iteiiip iiwariabh/ broken off at the junction of liviiirj 

 and dead polypes. This stem contains, as we have seen, as much 

 as 85 per cent, of mineral matter, and it could hardly be maintained 

 that the death of the polypes encrusting it would so affect the stem as 

 to cause it to continually break off at the exact boundary line between 



' Darwin : '• Naturalisfs Voyapre Ronncl tho World," 1860, p. 99. 



