38 THE FKESH-WATER LOCHS OF SCOTLAND 
gearing proved too crude to deal with the dehcate seiches of Loch 
Earn ; and it was near the end of the time at my disposal before we 
were able to remodel it and the other Sarasin on the plan of the 
wasffon recorder which worked so well at St Fillans. 
Besides the limnographic apparatus, we had quite a battery of 
meteorological instruments — three microbarographs of the Dines-Shaw 
pattern, and a Dines pressure anemograph, which was installed near 
my house at Ardtrostan, and worked beautifully. One of the 
microbarographs was placed at Ardtrostan, one at the west end of 
Loch Earn, and one at Killin, at the west end of Loch Tay. At each 
of these places were also ordinary barographs, which were controlled 
by means of observations made twice a day with a standard baro- 
meter in charge of Messrs White and Watson. As the observations 
made with the triangle of microbarographs were found to possess a 
certain amount of independent meteorological interest, an account of 
them was published separately in a paper " On the Theory of the 
Leaking Microbarograph, and on some Observations made with a 
Triad of Dines-Shaw Instruments,"" Proc. Roy. Soc. Edin., vol, xxviii. 
p. 437, 1908. 
Various Causes of the Denivellation registered in a 
LlAINOGRAPH 
The ordinate of the limnogr^m taken at any particular station on 
a lake shows the height at different times of the lake-surface at that 
station above a certain arbitrarily chosen level. Retardation and 
damping due to the instrument being allowed for, the limnogram 
gives the total denivellation at the station due to all causes what- 
soever. 
The examples reproduced in fig. 14 are from Loch Earn, all 
taken by the waggon recorder near St Fillans, They give a good 
idea of the great variety in the form of the limnographic record, and 
of the complexity of the phenomena to be explained. The two upper 
curves are very smooth, and furnish excellent examples of what we 
call the configuration period of a dicrote seiche. The third curve is 
an example of the strongly marked embroidery which appears on the 
limnogram during stormy weather. The fourth curve is an example 
of a sieche in moderately calm w^eather broken by varying weather 
conditions. The fifth curve, except for the slight wind embroidery, 
gives an example of an almost pure sinusoidal curve ; it was taken 
near one of the points on the lake, which we shall presently define as 
binodes, and furnishes a test of the mathematical theory. 
Among the various causes which may affect the level of a lake, the 
following may be enumerated : — 
