BEPORT ON THE PENNATULIDA. 223 



slightly, attaining its maximum diameter at about the point marked c in 

 Fig. 2 ; above this point it diminishes in size, but very gradually, 

 remaining of considerable thickness throughout the length of the 

 rachis, and ending at its top in an abruptly truncated extremity. 



In the Oban specimens the diameter of the stem at its vi^idest part 

 varies from 0-026in. to 0-050in. ; at its upper end, which, it must be 

 remembered, is imperfect in all the specimens, from 0"016in. to O-OS'Jin. 

 The average taper from the widest part of the stem upwards is ■002in. 

 per inch length of stem. 



In the Glasgow specimen of Virfiuhina mirabilin, in which the top is 

 perfect, the upper end of the stem projects above the top of the fleshy 

 rachis for a length about equal to its own diameter ; and a similar 

 condition has been noticed by Herklots, Koren and Danielssen, and 

 others, in perfect specimens of allied species of Virgularia. The most 

 obvious explanation of this feature is that the fleshy ccenosarc has, 

 owing to the action of the spirit in which the specimens are preserved, 

 contracted slightly and so left the end of the steni bare ; but there 

 appears to be some doubt as to whether this is the true one. Koren 

 and Danielssen speak on this j^oint as follows : — " Herklots and several 

 others have presumed that the reason of the axis being bare at the 

 upper end is to be sought for in a contraction of the sarcosoma under 

 the influence of the preserving liquid : this is, however, not the case ; 

 on the contrary, we are convinced that it is a natural state, and not 

 produced by any contraction of the ccenosarc. As well in this species 

 (Virduhiria ajfiiiis) as in many other genera and species, all the speci- 

 mens exhibited during life the same bare axis, and likewise the 

 saroosoma connate with (attached by growth to) the axis at the place 

 where the axis begins to be bare. In one specimen we even saw 

 several serpuhr attached to the bare part." * This last statement is 

 certainly strong evidence in favour of the view advocated by the 

 Swedish naturalists, for the specimen in question was brought up 

 living, and the ^evpuhr certainly could not have attached themselves to 

 the stem unless it had been already bare while in the water. 



The present seems a suitable place to discuss further that curious 

 mutilation of the specimens which we have seen to be so constant, nay 

 almost universal, a feature of museum specimens of Virgulnria niirabilis, 

 and which applies also, though apparently in rather less degree, to 

 other species of the genus as well. 



The facts on which all authorities are agreed are the following : — 



1. — The great majority of specimens of I'irpitlarin inirabilis as brought 

 to the surface by dredging are broken short at both ends. 



2. — The fracture at the upper end occurs at very variable situations, 

 but that at the lower end occurs very commonly at the junction of 

 stalk and raciiis, and nearly always within a short distance of this 

 point. 



* Sars, Koren and Dauielsseu ; " Fauna Littoralis Norvegise," Part 3, 1877, 

 p. 91. note, 



