150 Proceedings of Royal Society of Edinburgh. [sess. 
While the determination of the various periods of oscillation 
was a comparatively simple matter, that of the exact position of 
the nodes was very troublesome. 
Even when the system is oscillating in as perfect a manner as 
can he attained by means of the apparatus employed, it would 
seem that the nodal line does not assume a fixed position, hut 
oscillates through a small range in the direction of the seiche 
motion. Besides, the slightest variation in the rhythm of the 
movement causes the apparent position to vary considerably. 
What requires to he determined is, therefore, the mean of these 
positions when the oscillations are most perfect. There are two 
properties of which advantage may he taken in trying to determine 
the actual position. At a node the horizontal motion is a 
maximum and the vertical motion a minimum. It at first seemed 
that the place where the horizontal displacement was greatest 
would be easily determined ; for it seemed a practicable course to 
measure the range of the oscillating particles at or near the 
supposed position of the node, and hence find the required 
maximum position. For this purpose a very thin plate of mica 
was hung by means of a bifilar suspension in the water, where it 
swung to and fro with the surrounding liquid. This object was 
viewed through a small telescope, and the magnitude of the 
oscillations read on a scale fixed in an appropriate position. 
After repeated trials, however, it was found that there was a 
fkJ.ISC 
