TEMPERATURE OBSERVATIONS IN LOCH EARN. 



767 



fig. 14 it is seen that there is no definite maximum in the curve below that corre- 

 sponding to the 10-hour binodal seiche. This is not conclusive evidence against the 

 existence of plurinodal seiches of appreciable amplitude, for it is probable that seiches 

 of high nodality are more easily disturbed by surface winds than the uninodal and 

 binodal seiches. If, then, the phase of the seiche be altered at various times during* 

 the period discussed in the periodogram, it is evident that the periodogram method 

 will tend to obscure the periodicity rather than make it evident. 



























/ 



2 



3 



* 



5 



6 



7 



8 



• 



9 



10 



II 



12 



Period in hours. 

 Fig. 19. — Periodogram Analysis for Short Periods. 



In order to see whether evidence of plurinodal seiches could be found, a shorter 

 period from 16 h. on 25th August to 14 h. on 27th August, which was a time of great 

 calm, was chosen for examination. Readings of the 11° isothermal at Station 2 

 (fig. ll) were taken at shorter intervals of time than formerly, viz. 20 minutes, 

 and to this series of readings the method of periodogram analysis was applied as 

 described above. The curve of periods is shown in fig. 19. It could not be expected 

 that there should be strongly marked maxima, but there is distinct evidence of the 

 trinodal and quadrinodal seiches, the periods of which may be taken to be about 

 %\ hours and 6^ hours respectively. 



