356 On the Forms and States of Fluids [ 1 S3 1 . 



them into two series of ordinary advancing waves moving 

 directly across the course of the wind in opposite directions. 

 But as such series could not be caused by the wind exerted in 

 a manner similar to that by which ordinary waves are produced 

 (the direction being entirely opposed to such an idea), I think 

 the effect is due to the water acquiring an oscillatory condition 

 similar to those described, probably influenced in some way by 

 the elastic nature of the air itself (124) and analogous to the 

 vibration of the strings of the ^olian harp, or even to the 

 vibration of the columns of air in the organ-pipe and other 

 instruments with embouchures. 



These ridges were strong enough to arrange the sand beneath, 

 where ordinary waves had not been powerful enough to give 

 form to the surface. 



123. All the phenomena as yet described are such as take 

 place at the surfaces of those fluids in common language con- 

 sidered as inelastic, and in which the elasticity they possess 

 performs no necessary part ; nor is it possible that they could 

 be produced within their mass. But on extending the reasoning, 

 it does not seem at all improbable that analogous effects should 

 take place in gases and vapour, their elasticity supplying that 

 condition necessary for vibration which in liquids is found in 

 an abrupt termination of the mass by an unconfined surface. 



124. If this be so, then a plate vibrating in the atmosphere 

 may have the air immediately in contact with it separated into 

 numerous portions, forming two alternating sets like the heaps 

 described (95) ; the one denser, and the other rarer than the 

 ordinary atmosphere; these sets alternating with each other 

 by their alternate expansion and condensation with each vibra- 

 tion of the plate. 



125. With the hope of discovering some effect of this kind, 

 a flat circular tin plate had a raised edge of tin three quarters 

 of an inch high fixed on all round, and the plate was then 

 attached to a lath (69), a little lycopodium put on to it, and 

 vibrated powerfully, so that the powder should form a mere 

 cloud in the air, which, in consequence of the raised edge and 

 the equal velocity (70) of all parts of the plate, had no tendency 

 to collect. It was seen immediately that in place of a uniform 

 cloud a misty honeycomb appearance was produced, the whole 

 being in a quivering condition ; and on exerting the attention to 



