1861.] 229 



mercury was 2J inches external diameter, and T \j-th of an inch wide 

 = about 0*675 square inch of surface ; the battery consisted of ten 

 Smee's elements 7 inches wide (4) immersed 1|- inch in the exciting 

 liquid, and the quantity of hydrogen evolved in the meter whilst the 

 ring produced a moderate and steady sound, was 0*47 cubic inch in 

 three minutes ; therefore, with the solution of cyanide of potassium 

 given (3), the quantity of electricity required to give a good sound 

 in a ring containing 1 square inch of mercury surface is equal to that 

 which will evolve about 0-232 cubic inch of hydrogen per minute ; 

 this quantity approaches the minimum required for the purpose. In 

 a series of four other experiments with a ring of mercury containing 

 just one square inch of surface, and yielding a steady sound, the fol- 

 lowing quantities of hydrogen were evolved in 3 minutes: A0'7, 

 B 0-69, C 071, and D 0-685 cubic inch ; that is, A 0-233, B 0-230, 

 C 0-236, and D 0-228 cubic inch per minute: average = 0*2317 

 cubic inch per minute. 



30. In all cases an increase in the quantity of electricity passing 

 (either by increase in the depth of immersion of the battery plates, 

 enlargement of the anode, diminution of the cathode, diminution of 

 conduction-resistance in the circuit, removal of films from the anode 

 by stirring, mixture of strata of exhausted electrolyte by stirring, 

 rise of temperature of the electrolyte, &c.) was attended by an in- 

 crease in the number and a diminution of the width of the crispa- 

 tions, until at length, by excessive quantity of electricity passing, 

 the movements became violent, and the crispations and sounds sud- 

 denly ceased. Too large or too small a quantity of electricity pro- 

 duced undulations (8) without sounds. 



3 1 . With a given electric current, a wide strip of mercury gave 

 wide crispations and a base sound, and a narrow strip gave narrow 

 crispations and a high sound, because in the latter case the quantity 

 of electricity passing into a given amount of surface was greater than 

 in the former. Diminution of the conductivity of the electrolyte by 

 dilution with water appeared to decrease more the loudness of the 

 sounds than to alter their note, probably by decreasing the specific 

 gravity of the solution, and thus diminishing the depth of the 

 crispations. 



32. In producing the sounds in a circular pool of mercury by 

 gradually immersing the battery plates, the first effects are furrows 



