72 



Intelligence and Miscellaneous Articles. 



Thomson, and by Kolacek, and expressed by the formula 



The velocity of propagation v is thus a function of two forces which 

 are independent of each other — g (gravity) and T (surface-tension). 

 On the other hand, considered as a function of X, it has a minimum, 

 namely 23*11 centhn. for the wave-length X= 1*709 centim. ; so 

 that from this there results a natural division of waves into two 

 classes — ripples and large waves. Although observations on ripples 

 have been made by Thomson, Matthiesen, Lord Rayleigh, Arendts, 

 and others, these observations are wanting in the requisite accuracy, 

 and accordingly the author has carried out systematic observations 

 by various methods which control each other. In the first method, 

 tuning-forks were used to produce the waves ; in the second, the 

 simultaneous vibration of the armatures of two electromagnets ; 

 and finally in the third, which was made on a large scale in the 

 Rostock filtering-basins, small plates which were fixed to rods at 

 the level of the water. Stationary waves were always used, as 

 these are more favourable for observation and for calculation. 

 Reference must be made to the original for the observation of the 

 numbers of vibrations and the wave-lengths in the individual cases, 

 and for the manner in which the numerous difficulties were over- 

 come. As the results of the various methods are essentially the 

 same, it will be sufficient here to give a summary of that of the first. 

 Five different tuning-forks gave the following results : — 



n. 



X, in centim. 



A\. 



v, calc. ir 



centim. 



Mean of 

 one ob- 

 servation 



0-029 



Error of 

 the result. 



1 



0004 



Calc. 



Obs. 



Thomson. 



Obs. 

 23-26 



8-415 



2-947 



2-764 



0-183 



24-80 



16-215 



1-435 



1-321 



0-114 



23-27 



21-42 



0-009 



0*001 



36677 



0*739 



0*675 



0-064 



27-10 



24*76 



0-020 



0-009 



37140 



0*732 



0-670 



0-062 



27-19 



24-88 



0011 



0-008 



64*635 



0493 



0-450 



0043 



31-87 



29 09 



0005 



o-ooo 



The calculation of Thomson's results was made by putting 



981 

 67m 



■n 2 V 3tt 2 T 



tan 



a/ tan I = tanp, X= v / 3tt' 2 T tan 2/3 



and from this X, and therefore v, was calculated. It will be seen 

 that the experiment agrees sufficiently well with the theory 

 as regards the position of the minimum of v, and of the general 

 course of the values of X and v ; on the other hand, all the observed 

 values are considerably smaller than the calculated ones. In order 

 to follow this more minutely, measurements were made at various 



