608 PROFESSOR CHRYSTAL 
by calculation, from the contours of the lake basin, or by using rough preliminary 
observations, if such are available; and then to deduce the uninodal and binodal 
periods from limnographic traces obtained at the binode and uninode respectively, 
where these seiches are very nearly pure.* 
§ 14. It will be obvious that a seiche, properly so called, differs essentially from 
an ocean tide. The origin of a seiche, and the absolute and relative magnitudes of 
the pure seiches of which it is composed, no doubt depend on external circumstances ; 
but the periods and the positions of the nodes of the component seiches depend 
merely on the configuration of the lake basin, and on the surface-level of the water 
at the time. Im a tide, on the other hand, the periods are dependent on external 
disturbing agencies, chiefly the sun and moon. In the language of physicists, a 
seiche is a free oscillation ; a tide a forced oscillation. 
It is by no means the case that the limnograph traces always indicate a seiche, 
properly so called. 
True and unmistakable seiches are generally seen in comparatively calm weather, 
or in the lull after storm. At the commencement of a seiche the trace is often 
irregular; and during storm, oscillations of great amplitude are found, the intervals 
between the successive maxima of which are not equal. This inequality might, it is 
true, be due to a jumble of different seiche components, only some of which survive 
in the subsequent steady oscillation. In the traces for Lake Erie, which does not 
show the ordinary seiche phenomena with remarkable clearness, but, on the other 
hand, is remarkable for oscillations of enormous amplitude, associated with the 
passage of tornadoes, I have noticed indications of the retardation and acceleration 
of phase which characterise the initial stages of the action of an intermittent external 
disturbing force, whose period differs from any seiche-period of the lake. The same — 
thing is, I think, visible in the trace of PLantamour’s limnograph during the cyclone 
of 20th February 1879, which is reproduced by Foret on p. 194 of his monograph on 
Léman. Seiches of this kind I propose, for distinction, to call forced seiches, and | 
propose to discuss them farther in a subsequent communication, dealing with the 
external causes of seiche phenomena. 
§ 15. In the following purely mathematical part of this memoir my purpose is 
twofold :—first, to establish generally, and also by means of special instances, the 
leading principles stated above; secondly, to furnish formule and methods which 
can be applied in the investigation & priori of the periods and nodes of a lake whose 
length is considerably greater that its breadth, and which does not present excessive 
or abrupt variations in the configuration of its basin. I believe that in many such 
cases a sufficiently close approximation can be obtained by finding the normal curve 
of the lake, and replacing this curve by a combination of parabolas, straight lines, or 
simple quartic curves. Thus, for example, as will be seen in a paper by Mr E. 
* This was written before I had access to the monograph of EnpRés, whose plan of observation for the Chiemsee 
is in many ways a model, 
