476 PROFESSOR CHRYSTAL 



Let us suppose that the shower begins when the uninodal seiche culminates, and 



that it lasts for half the uninodal period. Then, if dkj denote the alteration in the 



amplitude of the uninodal seiche at the end of the lake, we get, from formula (46) of 



Part V., 



8^=12 + 1^, 



where § q is due to the impact, and frTj to the static effect of the precipitated water. 



To take an extreme case,* let us put <r= "3/60 = '1/30, i> = 700. Then, taking 

 Tj = 1 5 x 60 as a round number, we get q = *024, r = 1/30. Hence 



dk x = -036 + 22-5 = 23 mm., say. 



The result would therefore be a seiche having a range of 46 mm. It will be noticed 

 that the effect arising from the impact, viz. '036, is negligible. 



The conclusion thus arrived at bears out the inference of Endros t regarding the 

 effect of a rainfall of 7 mm. during 20 m upon the 43 m seiche of the Chiemsee. Such a 

 fall on one half of a parabolic lake having a 40 m period would generate a uninodal seiche 

 having a range of 10 '5 mm. 



We have little doubt that in some of the cases, to be cited presently, the precipitation 

 played an important part ; but the observations of Shaw and Dines on the effect of 

 passing rain-clouds in raising the barometric pressure tend to place difficulties in the 

 way of separating the effect of precipitation from the barometric pressure proper. It 

 would appear that the pressure to which the lake reacts so delicately is equal to the 

 pressure before the rain has fallen, that is, while it is still in the cloud in the form of 

 vapour ; but the matter requires and deserves further investigation. 



5. Effect of Squalls. — On 11th August, 8 h to 9 h , a prolonged depression on the 

 microbarogram is associated with a prolonged elevation on the limnogram. The release 

 of this denivellation caused a considerable uninodal seiche (see fig. 7). 



The embroidery on the two limnograms is interesting. It has the same period, 

 T= ri5 m , at the binode and at the Picnic Point. 



The limnogram taken at the binode on 21st August between ll h and 17 h gives a 

 good illustration of the effect of the passage of well-marked disturbances, whose whole 

 duration (including a positive and a negative phase) was not very different from the 

 period of the uninodal seiche. The binodal limnogram in fig. 8 is a photograph from a 

 very rough tracing which had to be made, because at the moment the Sarasin pen was 

 out of order, and was replaced by a pencil which gave only a very faint trace. 



It will be observed that there are two well-marked increases of range, one about 

 12 h 15 m , the other about 14 h 30 m . These are connected with two sharp V-shaped 

 disturbances on the microbarograms, the first of which is preceded by a couple of 

 undulations whose period is not very different from 15 m . 



The velocities of propagation of the two disturbances in the direction of the axis of 



* See Hann, Lehrbuch der Meteorologie (1906), pp. 270, 275. 

 t Seeschwankunyen beobachtet am Chiemsee (1903), p. 103. 



