60 REPORT ON THE PENNATULIDA. 



Virgithiria ynirahilis, from three quarters of an inch to three inches in 

 length, each fragment containing the portion of stem belonging to it, 

 and all five showing evident signs of having undergone partial digestion. 



The most interesting point still remains to be noticed. Of these 

 five fragments no fewer than three are tojyg, i.e. actual perfect upper 

 ends, a point the significance of which is at once evident when we 

 remember that of the speciniens of Virgularia mirahilis dredged either 

 off our own coast or elsewhere, only one single specimen — the one in 

 the Glasgow Museum — is known to have a perfect top. 



Mr. Darbishire's observation proves that fish do actually bite off 

 and swallow as food fragments of VinjuJaria ; also that they are able 

 to find specimens with perfect tops, for which tops they would appear 

 to have some special liking. It need hardly be pointed out that this 

 furnishes the strongest possible confirmation of the theory we had been 

 led to frame on purely independent grounds.* 



Two points still require explanation. Firstly, why, if the fish bite 

 off the tops and swallow them as food, do they not devour the whole 

 of the rachis as well ? Secondly, why do the fish eat the tops off 

 Virgularia and leave imtouched the allied genera, Pennatula and 

 Funiculiiia, which are found growing alongside it, and of which the 

 latter, at all events, would appear to be far more tempting as food, 

 owing to the much greater bulk of fleshy substance it affords, and the 

 much smaller thickness of its stem in the upper part. If it be 

 supposed that the calcai'eous matter of the stem is the real attraction 

 lo the fish, it is difficult to understand why Peituattila, with its 

 innumerable calcareous spicules, is allowed to escape. 



We shall return to both these points further on. 



.S. — Thf Polypes and Zooiih. — 



The general arrangement of the leaves is shown in Fig. 1 ; and the 

 leaves, together with the polypes of which they are formed, in Figs. 3, 

 4, and 5. In the Oban specimens each leaf is formed by the fusion of 

 seven to eight polypes, placed side by side, the number being constant 

 in all the leaves of any one specimen, but varying in different speci- 

 mens. The leaves are arranged strictly in pairs at the two ends of the 

 rachis, but about its middle often show slight irregularities, and may 

 even alternate with one another for some little distance. 



At the bottom of the rachis there is no trace of leaves or polypes, 

 but about an eighth of an inch higher up the leaves begin to appear as 

 small transverse ridges : they are at first very close together, and the 

 component polypes very sntiall ; but passing upwards the polypes gradually 

 get larger and the leaves wider apart. Having reached their maximum 

 size and distance from one another, the leaves preserve these for some 

 distance, and then, towards the top of the rachis, begin gradually to 

 get smaller and closer together. 



In the Glasgow specimen, which has the rachis perfect at both top 

 and bottom, we have been able to measure accurately the number of 



' Professor M'Intosh also mentions flndiiiR T'. mirahilis '' occasionally in tlie 

 stomach of the cod." 



