Disturbances due to rides and Wares. 153 



II. Experimental. 



A. Observations with Moored Electrodes. 



An opportunity occurred of making some observations 

 with electrodes laid outside the entrance to Dartmouth 

 Harbour. The electrodes found suitable for the purpose 

 were of silver coated electrolytically with a film of silver 

 chloride ; such electrodes were tolerably free from polar- 

 ization effects. The positions of the electrodes are shown 

 in PI. II. fio-. 3. Two selected electrodes were connected 

 directly to a Paul recording galvanometer ("temperature 

 recorder ") which acted as the voltmeter, and with this 

 simple apparatus records were made continuously for pro- 

 longed periods. These records are reproduced in PL II. 

 figs. 4-6. The recorder gives a dot about every two minutes 

 on a paper roll, and the curves are obtained by joining these 

 dots by hand. Owing to the intermittent character of the 

 record the curves do not accurately represent the variations 

 of short period such as those due to waves. The irregu- 

 larities, however, which are introduced by such variations 

 oive a fair idea of the magnitude of those fluctuations. 



PI. II. fig. 4 shows a record made during the period 

 9.ix.lS to 14.ix.18, taken with electrodes Mi and G 1 ap- 

 proximately 2000 yards apart. The record exhibits un- 

 doubtedly an E.M.F. varying periodically and having a 

 period equal to that of the tide. The fact that the E.M.F. 

 indicated does not reverse in sign has no significance, as a 

 pair of electrodes generally exhibits a slight electrolytic 

 E.M.F. which displaces the zero. The maximum change of 

 E.M.F. between trough and crest is about 37 millivolts, so 

 that the amplitude of the oscillation is 18*5 millivolts. It is 

 impossible to make an accurate comparison between this 

 figure and any calculated value, as there are so many un- 

 known factors to take account of. For example, the flood 

 tide runs practically parallel to the cable over most of its 

 length and the ebb tide crosses it nearly at right angles. 

 Again, the length of cable subject to tidal action is probably 

 only about 1000 yards. It may be pointed out, however, 

 that the observed E.M.F. is of the same order as the 

 maximum E.M.F. to be expected from a 1 knot tide flowing 

 between electrodes 1000 yards apart. 



The record shows very markedly how the amplitude of 

 the electromagnetic tide diminishes from spring tide to 

 neap. 



A remarkable feature of the record is the relation of the 

 times of hi<-h and low water to the curve, these times 



