514 
MESSRS. W. DE LA RUE AND H. W. MULLER ON THE 
diametrically opposite, a stratum was touched at an 
opposite end of its diameter 
each wire ; the readings were :— 
With -J- shunt. 
Without shunt. 
The current was from 
To left 3 — 
30- 
1 to 2 
„ 8- 
79- 
JJ 
The wires were now reversed, a 
to 1 and b to 2, and the following observed : — 
With i shunt. 
Without shunt. 
The current was from 
To tlie right 12 + 
119 + 
1 to 2 
2 + 
I") ?) ‘“I 
20 + 
>> 
To the left 8 — 
79- 
2 to 1 
O 
>> >> c> — 
30- 
5) 
0 
0 
It is evident, therefore, that there is a current sometimes in one direction, some¬ 
times in the contrary, across a diameter of the tube, as if the motion of the molecules 
conveying the discharge was of an epicycloidal character. By way of comparison a 
tin-plate, 14^ inches (36'8 centims.) long and 5^ inches (14 centims.) wide, with 
circular ends, was made ; a connecting clamp was soldered to each end and two 
others across a central diameter. The direction of the current was ascertained when 
the poles of the battery were connected one to each end of the plate, and the opposite 
wires of the galvanometer also one to each end. Then the two clamps across a central 
diameter were connected alternately with the opposite ends of the galvanometer, the 
terminals remaining connected with the ends of the plate, but no deflections occurred 
even without a shunt. 
One of the main objects of the experiments recorded in this paper was to endeavour 
to discover some of the obscure secrets of the electric discharge through gases. 
Although many more will undoubtedly have to be made before a theory can be formed 
which will account even for a few of the complex phenomena presented by this dis¬ 
charge, yet every onward step, however small, brings us nearer to the goal, and we 
shall consider ourselves well rewarded if we have advanced the frontier in the very 
smallest degree. We believe that what we have done has secured some advance, and 
rendered evident that electricity is always conveyed in gases by ponderable matter 
and not by the so-called ether; moreover, that there are eddies in the discharge in a 
vacuum tube, and that it is possible that these eddies may be connected with the 
production of strata whose form would depend on the kind of eddy originally pro¬ 
duced. This recalls the phenomena figured in Part III., Plate 8, fig. 34 ;* in this is 
seen an eddy of strata through small holes in a tube confining the main discharge. 
We have much pleasure in thanking Professor Stokes for his many valuable 
suggestions, and for his kind advice generally, during the course of this investi¬ 
gation. To our chief assistant, Mr. James Beam, we are indebted for able and 
zealous co-operation, and we wish to place on record the excellent services of our 
junior assistant, Mr. Ernest Davis. Mr. H. Pveynolds has, as on former occasions, 
taken the photographs with his usual skill. 
Phil. Trans, for 1880, Vol. 171. 
