ON THE SEICHES OF LOCH EARN. 491 



be said to have been given as yet. They are, however, of great interest, because there 

 is some reason to believe that in part at least they reflect in miniature the action of the 

 causes which produce the storm- waves of the ocean, our knowledge of which is still far 

 from complete, although they are of such vital importance to seafaring men. 



Inasmuch as our first object was to determine as accurately as possible the seiche 

 periods and the positions of the nodes of Loch Earn, the limited time at our command 

 was allotted and our apparatus disposed mainly for these two purposes ; and it was not 

 until near the end of our observations, after the extemporisation of the statolimnograph, 

 that much attention was given to the vibrations of the lake. We cannot, therefore, 

 pretend to offer much towards a final solution of the problem of the vibrations ; but 

 we may record a few observations which seem to enhance the interest of the question, 

 and may ultimately prove useful in its final solution. 



The embroidery caused by these vibrations, as may be seen by comparing the 

 figures of this memoir and figs. 5, 6, 7, 10, 11, 12, 16, 17, 18 of the previous memoir 

 of this series,"" varies considerably in form, and may be regular or irregular according 

 to circumstances. It must also be remembered, as was long ago pointed out by 

 Forel, that, owing to the damping effect of the well and access tube, each limnograph 

 reproduces more or less of these vibrations according to its adjustment. The stato- 

 limnograph, used with a wide access tube, owing to the very small inertia of its moving 

 parts, is best adapted for this purpose. Compare, for example, figs. 4 or 5 of Part II. 

 with fig. 1 1 of Part I. 



Although occasionally the embroidery continues regular for a considerable time, and 

 appears to have a perfectly definite period and constant or at least slowly varying range, 

 as a rule its configuration changes rapidly, and any regularity is transient. This makes 

 it very difficult to analyse it into harmonic components, even if analysis into a finite 

 number of such components were possible. 



In our observations the maximum range of the vibrations varied from to 21 mm. ; 

 an average value might be about 6 mm. At times the range of the vibrations [e.g. 

 fig. 19) exceeded the range of the seiche, so that the former quite obscured the latter. 



The periods observed showed much less variation. In the limnograms taken with 

 the Waggon recorder and Sarasin instruments, the period ran from l - 3 m to 2 m ; in the 

 statolimnograms, from '42 m to - 79 m . It must be remembered, however, that in the 

 latter the short-period embroidery obscures that of longer period ; and in the former 

 the vibrations of shortest period are damped out. For the ordinary limnograms the 

 average of the periods might be put at l*47 m . The period that actually occurred oftenest 

 in the cases we happened to examine was l'5 m . 



The embroidery was never observed unless there had been sufficient wind to cause 

 progressive surface waves ; and it subsided at once when these waves disappeared. The 

 observations of Halbfass, Endros, and others show that it is usually more marked when 

 the limnogram is taken at the leeward end of the lake : it may be very marked there 



* See Trans. R.S.E., voL xlv., pp. 366, 369, 370, 380, and 383. 



