8 RATIONALE OP 



view. I may add that the explanation given does not render 

 necessary the supposition of two distinct cun-ents flowing in opposite 

 directions and crossing each other, one flowing in unbroken continuity 

 from without inwards, and the other flowing in unbroken continuity 

 from within outwards, either in the same or in different channels. 

 More probable is it that the transference is going on continuously, 

 simultaneously in both directions — a process which may perhaps be 

 made more intelligible by an explanation of what takes place in the 

 decomposition of water by galvanic influence. If a plate of zinc and 

 a plate of copper connected by a wire be kept apart and introduced 

 so into a vessel of water, the water will be decomposed, one of its 

 elements, oxygen, will be disengaged in connection with one of these 

 plates, and hydrogen, the other constituent element, will be dis- 

 engaged at the other ; but there is no appearance of decomposition 

 taking place at any one or more points, and of the elements flowing 

 thence to their respective issues, — and what is actually seen, may be 

 explained by supposing that in immediate contact with one plate, or 

 with each, a particle of water is decomposed by the liberation of one 

 of the constituents which escapes, and the other is transferred from 

 particle to particle, only becoming visible when it escapes in contact 

 with the other plate. 



The process may be illustrated mechanically by means of a six- 

 inch parallel rule, thus : let the one scale represent oxygen and the 

 other hydrogen ; let perpendicular lines divide them into six equal 

 spaces, which conjointly may represent particles of water composed 

 of oxygen and hydrogen ; let these be numbered, the first couple 

 each marked 1, the second couple 2, and so on ; let the two scales of 

 the ruler next be drawn in opposite dii-ections, and if the spaces 

 marked on them be equal to the distance which each projects beyond 

 the other, you will have again spaces appropriately representing 

 particles of water ; but you will have oxygen No. 1 above hydrogen 

 No. 2, oxygen No. 2 above hydrogen No. 3, and so on ; hydrogen No. 1 

 and oxygen No. 6 representing what seems to take place in the 

 decomposition of water by that simple galvanic apparatus. 



As in the decomposition of water by galvanism thus illusti'ated, 

 the escape of the hydrogen of water in contact with one plate, and 

 that of the oxygen of water in contact with the other plate, seem to 

 result from a transference of hydrogen from particle to particle, and 

 simultaneously with this a transference in the opposite direction of 

 oxygen from particle to particle, through the whole line of particles 

 of water intervening between the two plates, — so in the exosmosic 



