238 Proceedings of the Royal Physical Society. 



we are led to infer that expulsion of the contents is not due to external 

 pressure acting on the capsule, but is caused by the walls shrinking to 

 their natural size after having been stretched. 



A resting nematocyst may therefore be regarded as possessing a state of 

 unstable equilibrium, the capsule being over-distended and the fluid 

 secretion in a state of tension. From this point of view, discharge is 

 initiated by the breaking down of the resistance which maintains this 

 unstable condition. 1 



It will be seen that this experiment with the model, as well as the 

 deductions drawn from it, is based upon the view that the immediate cause of 

 the discharge lies in the contraction produced by the surrounding cell. On 

 the view that discharge is due to osmotic pressure set up within the capsule, 

 the argument for the state of unstable equilibrium is less satisfactory. It 

 rests, in fact, only on the observation that the parts of the nematocyst shrink 

 in size when the fluid contents are being expelled. It may, however, be 

 pointed out, as perhaps lending some support to the suggestion, that a tense 

 condition in the nematocyst would help to explain the surprising rapidity 

 with which discharge is usually accomplished. 2 



The action of osmotic pressure in everting a flexible tube may be illustrated 

 by the following experiment. A glass tube, A (Fig. 3), referred to in the 

 description as the carrier, is securely fitted at one end into the stem of a 

 funnel, B, and at the other end into one of the tubulures of a glass receiver, 

 C. To the end of the carrier which projects into the receiver is fitted the 

 flexible tube, a piece of fresh gut serving the purpose. , The osmotic cells 

 employed consist of cylindrical parchment dialysers, E and E', each being 

 closed at both ends with rubber corks, and of about 800 cc. capacity. Two 

 of these are shown in the fig., but the number can be readily increased. The 

 dialysers are nearly filled with a saturated solution of common salt, and are 

 then connected, by means of an inverted Y-tube, with the lower orifice of the 

 receiver. By temporarily disconnecting the glass carrier, more salt solution 

 is then added until, having filled the dialysers, it passes up through the 

 connecting tube into the lower part of the receiver. Finally, the remainder 

 of the receiver, the carrier and the funnel are each in turn filled with tap- 

 water, which is used instead of salt solution in order that the flexible tube 

 may hang down freely. The experiment is started by immersing the dialysers 

 in jars of distilled water, F and F'. As soon as the salt solution begins to 

 absorb water, the parchment membranes become stretched and an upward flow 



1 Tlic mechanism upon wliicli the resistance depends is probably, in part at least, to be 

 found Hi the valve-like processes situated at the junction of the capsule and filament. 

 \ contributory factor is no doubt the tapering form of the filament. 



