396 THE SEICHES OF LOCH EARN. 



The Syphon. — Two sizes of red rubber tubing were employed — one of ^-inch inside 

 diameter in fine weather, when we wished to observe as many of the seiches as possible ; 

 one of j- inch diameter, when we wished to eliminate the smaller seiches or the surface 

 swell. The syphon was filled in the loch ; — the larger size of tube could be stopped bv 

 inserting the finger till the end was moved into the pail — the smaller one by a lead- 

 pencil. After the syphon was in place, it was usual to fill the pail considerably above 

 the surface of the loch in order to test whether the syphon were working. 



Method of using Index. — After the apparatus was set up, and it had been ascer- 

 tained that the syphon was working properly, a little time was allowed for the water 

 in the well to adjust itself to the level of the lake before beginning the observation. 

 Readings were taken every half-minute. A large, fairly pure seiche could be accurately 

 enough plotted from readings taken at intervals of one minute. This interval was 

 found to be much more trying to the observer than a shorter one, and there was even 

 a greater liability to error ; the longer interval was too sustained a strain on the 

 attention. Readings were sometimes taken at shorter intervals (every quarter-minute) 

 when it was desired to measure the smaller seiches. 



At first all the readings were noted down in figures, and the curves were afterwards 

 plotted, and in wet weather this continued to be the method adopted. In dry weather 

 the curves were plotted directly on to the squared paper, a dot being made for each 

 reading. A great saving of labour was thus effected, and there was the further 

 advantage that one could readily see, after a few minutes' work, whether there was 

 enough seiche to make it worth while to continue. 



When well-marked short seiches occurred along with larger ones, the larger seiches 

 could not be accurately traced by the index readings. The half-minute readings might 

 be sometimes at the maximum and sometimes at the minimum of the shorter seiches, 

 and so the curve was rendered irregular. If it was desired to read both the long and 

 the short seiches, an attempt was made to watch for the turning-points of the shorter 

 seiche, whether they occurred at the half-minute readings or between them ; but as 

 the eye had to be momentarily removed from the index, in order to plot each point as 

 read, the maxima and minima were apt to be missed. 



All need to attempt to read very small seiches was removed by Professor Chrystal's 

 adaptation of the statoscope as a limnograph (see Professor Chrystal's paper, p. 367). 



It then became the custom to cut out the smaller seiches by using narrower syphons 

 (^ inch). When this was done a certain amount of lag was introduced, which had to 

 be measured and corrections made before the curve could be compared with that of the 

 Sarasin limnograph. 



In the carrying out of these observations and experiments, Mr Archibald Fraser, 

 who was engaged from first to last, was of invaluable assistance. He offered many 

 useful suggestions for the improvement of the instruments, and made many excellent 

 observations with the index limnographs. Mr Fraser is shown in fig. 23 taking an 

 observation. 



