392 



PROFESSOR CHRYSTAL AND MR JAMES MURRAY 



and oiled. These conditions cannot be fulfilled when the instrument is set up on a 

 temporary structure, inadequately protected from the weather. The vertical rod is of 

 brass — it is square, and slides through two iron sockets slightly greater in diameter. 

 These cannot act as guides unless there is contact of one side ; and we thus have brass 

 sliding on iron, which is apt to get rusty and cause very serious friction. 



The extent to which the amplitude of the seiche was diminished on the trace in 

 consequence of friction, was strikingly exhibited when the trace of a seiche taken in 

 calm weather was compared with the trace of the same seiche during wind. Whenever 

 the wind rose the amplitude increased. That this was not a real change in the seiche, 

 but an improvement in the record resulting from reduced friction, was illustrated by 

 an experiment. When there was no wind its effect was imitated by tapping con- 



Fig. 22.— The M'Kinstry Waggon. 



tinuously on the framework. The effect was the same. The amplitude of the seiche- 

 trace increased during the tapping by as much as three times. The vibration from 

 either cause had the effect of shaking momentarily free the parts in contact where the 

 most serious friction occurred, and the trace gave the full amplitude, but in a series 

 of jerks. 



The portions of the limnograph, external to the recording box, are complex. 

 There are so many joints that even under the best conditions there is some 

 back-lash. It was resolved to do away with all this part of the apparatus 

 altogether ; and to render the instrument direct acting by taking a line over the 

 pulleys, and bringing the cylinder right under one pulley, so that the float would 

 hang free under the pulley. The triple steel wire used for the Pullar sounding 

 machines was used as the driving line. Grooves to accommodate the wire were 

 turned in the two pulleys. 



