SEICHES AND OTHER OSCILLATIONS 
65 
Sometimes a violent disturbance on the microbarogram is accompanied 
by a moderate or slight disturbance on the limnogram, and occasion- 
ally the disturbance on the limnogram is much greater than might 
at first sight be expected from the microbaric disturbance. The 
mathematical theory (and indeed common sense apart from recondite 
theory) indicates the reason for this. If an increase of pressure 
operates on one half of a symmetric parabolic lake during half the 
period of the uninodal seiche while the water in that half is falling, 
it will evidently work the whole time towards increasing the amplitude 
of the seiche. Also, if there were to be increase of pressure for half- 
periods alternately on the two sides of the uninode, always tending 
to drive the water in the direction in which it w^as going, it is obvious 
that a very small increase might end by producing a very large seiche. 
As a matter of fact, a considerable rise of seiche is occasionally found 
when the microbarogram is comparatively smooth, but in such cases 
a closer examination usually shows a faint undulation with a period 
not very different from that of the seiche which is generated. 
On the other hand, if an increase of pressure is supposed to act 
on one half of the parabolic lake during the whole period of the 
uninodal seiche, or if it is distributed equally on both sides of the 
uninode, it is easy to see that the final result in altering the range of 
the seiche will be nil, however long the increase of pressure may act. 
Absence of microbaric disturbance is accompanied by absence of 
seiche disturbance ; that is to say, either there is no seiche at all, or 
an existing seiche continues unaltered. Under these circumstances 
the limnograms from Loch Earn are of great beauty. As an example, 
mention may be made of a record taken by the .converted Sarasin 
near the binode from 23rd to 27th August. This shows a regular 
uninodal seiche with an average range of 6 to 7 mm., which continued 
for over eighty-nine hours. During all that time the microbarogram 
shows only very slight disturbance — faint undulations, occasionally 
periodic. The range of the seiche is not absolutely constant, but 
sometimes rises and sometimes falls gradually, the minimum being, 
say, 4*5 mm. and the maximum 8 mm. (corresponding to 7 mm. and 13*6 
mm. at St Fillans). There is nowhere any sudden change of phase. 
Examination of the limnograms shows that seiches may be 
generated " suddenly i.e. attain their full range in one or two oscilla- 
tions, or may be generated " gradually,""' i.e. the full range may be 
attained only after a considerable number of oscillations. 
Among the causes that might generate seiches suddenly the follow- 
ing may be considered : — 
1. The sudden release of a static denivellation of the whole lake- 
surface, due to the progression of the general system of the atmo- 
spheric isobars. 
5 
