40 THE FEESH-WATER LOCHS OF SCOTLAND 
given point. The general effect is, however, a fairly regular pattern 
of small progressive waves of apparently constant length, usually 
diversified by wave maxima at approximately equal intervals. This 
system persists for some time after the wind falls, and in this stage 
is usually spoken of as " swell."' 
5. Seiche Denivellations. — These are stationary oscillations of 
the whole lake, having nodes {i.e. places of no vertical motion), 
ventral points (^i.e. places of no horizontal motion), and periods 
depending only on the configuration of the lake-basin. 
The last three forms of denivellation — which for shortness we 
may call solitary zoave, .swells and seiche — all make themselves felt on 
the limnogram ; and it may be worth while to dwell for a little on 
the distinction between them, which is often imperfectly understood. 
Solitary Wave. — Suppose water to be poured to a depth of 
about three inches into a shallow rectangular trough, say 6 feet 
8 inches long and 8J inches broad. ^ If a vertical board nearly as 
broad as the trough be inserted at one end, and a moderately sudden 
sweep made towards the other end, a hump will be raised on the 
surface of the water, which travels along the trough without very 
rapid alteration of form, is reflected at the end, and travels backwards, 
and so on. Observation shows that the particles of water are 
affected by this wave only while the hump is immediately over them. 
If the trough were infinitely long, they would come to rest after the 
solitary wave had passed away. Each particle comes to rest in a 
position at, or at least near, its original one. It is the wave form, 
and not the constituent water, that really travels, as may be seen by 
watching a splash of red ink thrown on the top of the wave as it passes. 
Theory and observation agree in giving the formula V = J{gh) for the 
velocity of the highest point of the solitary wave, where g is the 
acceleration due to gravity, and h the depth. 
It is important to observe that a wave of this kind, travelling 
backwards and forwards in a lake of uniform breadth, depth (^), and 
length (Z), would produce a periodic disturbance at one end of the 
lake having a period ^l/J(gh). It should be noticed, however, that 
the curve on the limnogram would not in general be a sinusoid, like 
the lowest curve in fig. 14, but something more of this shape: 
— ^ ^ , where periods of positive denivellation 
alternate with periods of no disturbance. 
Progressive Waves generated by a Wind Current. — Next 
suppose water to be poured into a tank 12 feet long, 2f inches wide, 
fitted with a parabolic bottom, concavity upwards, till the depth at 
1 The dimensions given here and in what follows are taken from apparatus 
actually used in demonstrations during a lecture at the Royal Institution in 
London, much of which is reproduced in this article. 
