INTRODUCTION. xix 



merely the mechanical result of the impact of the ciliary 

 current on the closed tip. If so, the current runs upward 

 on the whole inner surface of the walls, and then returns 

 down the centre. And this, I am persuaded, is the case. 



That the tentacles are perforated at the tip is, however, 

 certain : but it is closed or opened at the Avill of the animal, 

 the outer annular layer of fibres acting as a sphincter. 

 Nothing is more common than to see a fully expanded indi- 

 vidual of T. crassicornis, when suddenly alarmed, eject 

 slender streams of water from the tips of its tentacles ; and 

 I have seen an instance in which, the animal being but just 

 covered with water, the jets were projected to a height of 

 three inches above the surface. In S. bellis, after macera- 

 tion, the slightest pressure on these organs causes the 

 pigment to ooze out at the tip. In many that I so treated, 

 not one allowed it to escape at the side ; nor in any case 

 was there the least appearance of resistance, suddenly 

 yielding as if by a rupture ; nor did the aperture in any 

 case enlarge, nor was it in any case otherwise than at the 

 precise extremity. From which circumstances I infer a 

 natural foramen there ; and think that it exists in all 

 species, except those (as Corynactis and Caryopliyllia) 

 which have a globose appendage at the extremity of the 

 tentacle. 



The circulation of the nutrient fluid is aided by a curious 

 apparatus of foramina, of which I have met with no 

 description. It is difficult to find them in dissection, for 

 they appear to close with contraction ; but in bellis, on 

 making a transverse section just below the disk, I have 

 found a small round aperture in each primary and secon- 

 dary septum, through which I could thrust a probe without 

 laceration. It is daring life, however, that, under certain 

 favourable circumstances (for they cannot at all times be 

 detected), they must be studied. In dianihus, when very 

 much distended, I have seen the principal septa perforated 

 with a large circular foramen in the midst of their broadest 

 part, resembling iron girders supporting a floor, excavated 

 for lightness (Plate XI. fig. 1, b). In Antliea cereus they 

 are conspicuous ; * but I have been unable to detect them 

 in T. crassicornis or in Corynactis. 



' The most satisfactory observations I have made on these perforations 

 were on a specimen of Anthca cereus, var. sulphur ea. Being very much 

 expanded, and distended to translucency, the base adherent to the side of 

 a glass tank, the column greatly exceeding the base, the •window opposite, 



I 2 



