518 



action. The original hypothesis was formulated as follows: "If we 

 suppose that a cilium is a hollow curved extension of the cell, occupied 

 by hyaloplasm, and invested by a delicate elastic membrane, then it 

 must follow that if there be a rhythmic flowing of hyaloplasm from 

 the body of the cell, into and out of the cilium, an alternate extension 

 and flexion of that process would thereby be brought about. . . . The 

 same result might be got, supposing the cilium to be a straight and 

 not a curved extension of the cell, if the enveloping membrane were 

 thicker (or otherwise less extensible) along one side than along the 

 other. This assumption would enable one better to account for the 

 spiral direction of the movement of certain cilia: for this form of 

 movement would be produced if the line of lessened extensibility in 

 them were to pass in a corkscrew fashion along the cilium in place 

 of straight along one side, as might be assumed for ordinary cilia." 



It will be seen that two separate hypotheses are here enunciated, 

 both depending upon the movement of fluid and the principle of 

 hydraulic pressure, but differing from one another in the fact that the 

 one theory assumes a pre-existent curved form of the cilium , the 

 other a pre-existent structural modification. Of the two, the first, 

 which supposes the cilium to work on the principle of the Bourdon 

 pressure gauge, is the simpler. I felt, however, constrained to adopt 

 the second because I could not at the time see how the spiral move- 

 ment of many cilia and flagella would be accounted for by the second 

 hypothesis. It did not occur to me that variations of pressure within 

 a pre-forraed hollow spiral would produce movements comparable with 

 those of the spirally moving flagella. The fact, however, that simple 

 modifications of form of the cilia, without any structural modification 

 such as is involved in the assumption of a fibre or line of lessend 

 extensibility along one side or in a spiral, is adequate, on the hydraulic 

 theory, to account for all the movements of cilia can be demonstrated 

 by models constructed in the following manner: 



Pairs of tapered slips of thin india-rubber sheeting are laid one 

 upon the other, straight or spirally (as shown in the figures) over a 

 glass cylinder, to which they are temporarily fixed. Their adjacent 

 edges are stuck together with liquid india-rubber, such as is used for 

 mending bicycle tires, leaving open the base of the flattened, tapered 

 tube which is thus formed. When the adhesion of the edges is secure, 

 the tube is removed from the cylinder, and if it is now suspended 

 in fluid, it will be found to preserve the form imparted to it by the 

 cylinder. If the strips had been laid transversely over the cylinder 

 (Fig. 1) the flattened tube will have a simply curved form: if laid 



