1007.] on Seiches in the LaTces of Scotland. 665 



this tank ; but it seems Avell-nigh incredible, though unquestionably 

 true, that the same holds for two water particles at the two ends of 

 Lake Geneva, that is to sav, 45 miles apart. It was therefore not an 

 obvious remark, but a brilliant generalisation, which Forel made long 

 ago, when he asserted that the seiches of Lake Geneva were standing 

 oscillations, similar in nature to the one which Mr. Heath has just 

 started in the 12-foot tank. 



By stirring with a period of about 1\ sec. at a distance from the 

 end of our miniature lake = 0*21 of its length, we raise a standing 

 oscillation of another description, with two nodes each somewhat nearer 

 the end of the lake than a quarter of its length, and a ventral point 

 in the middle. At the ventral point the motion of the water is wholly 

 vertical, whereas at the two nodes it is wholly horizontal. This kind 

 of motion is called a pure hinodcd seiche. The orbits of the particles 

 at various parts of the liquid will be understood from the figure in 

 the lantern shde (Fig. 5). 



With equal ease we can stir up a trinodal seiche. 



It should be noticed that the uninodal water-surface for a parabolic 

 lake is always a plane, which oscillates about the nodal line between 

 the full drawn and the dotted positions in Fig. 5. For the same kind 

 of lake the binodal water surface is a parabola, which varies in 

 position and curvature between the dotted and full drawn positions. 



In the case of a lake of uniform depth, the corresponding surface 

 curves are sinusoids. 



Generation of Seiches hy Periodic Wind Impulses. — Mr. Heath Avill 

 now show us another method of generating seiches, which is nearer to 

 nature than the last. 



He has covered in half of the small trough,* and arranged to drive 

 an air-blast along the surface as before. So long as the blast is con- 

 tinuous we simply get surface progressive Avaves, as in a previous 

 experiment. There might possibly be a small seiche left on stopping 

 owing to the drift ciu-rent carrying an excess of water to the leeward 

 end of the lake, but it is imperceptible. Xow he repeats the experi- 

 ment with an intermittent blast, timed to a period of about 5 sec. by 

 the metronome. The result you see is a uninodal seiche. 



If we repeat the experiment, covering in the tank up to the binode 

 merely, and timing the puffs to J sec, we get, as you see, a binodal 

 seiche ; for the water rises and falls at both ends together, instead of 

 rising at one end and falling simultaneously at the other, as before. 



Hydrodynamical Theory. — And now a word or two about the 

 mathematical aspects, at least the more general of them, of this matter. 

 Wave motions of the kind shown are termed pure seiches, and they 

 may be uninodal, binodal, trinodal, . . ., i-nodal. 



* Length, 5 ft. ; breadth, 4 in. ; depth, 5 in. ; depth of water, about 3 in. 

 Like the large trough, it had a parabolic bottom. 



