OH ELECTRICAL DISCHARGES THROUGH RAREFIED GASES. 
169 
coil is at work, a certain amount of sensitiveness will usually be visible. (See Plate 16, 
fig. 6.) But if the coil be stopped and the current allowed to How from the condenser 
through the tube without disturbance from the entrance of the coil-discharges, the 
discharge will be found to have lost all its sensitiveness. The explanation of this we 
believe to be that so long as the coil was at work its pulsations produced sudden 
periodic elevations of tension in the condenser, which caused corresponding discharges 
through the tube. It is probable that in these cases the periodic discharges are only 
superposed upon the continuous discharges which we see upon stopping the coil, and 
which are caused by the mean tension in the condenser, so that it would be more 
strictly accurate to say that the discharge is subject to intermittent or periodic 
variations of intensity than to say that it is itself intermittent. But, as all the 
phenomena agree with the supposition that the discharge is partly sensitive and partly 
non-sensitive, and that the sensitive portion is wholly caused by the pulsations 
produced by the coil, it is better to look upon it as a superposition of two discharges, 
and to regard the sensitive one as intermittent. 
It is not, however, universally the case that the discharge from the condenser is 
non-sensitive when the coil is not acting. When the tension in the condenser becomes 
low, it is very usual to find sensitiveness in the discharge. This phenomenon, how¬ 
ever, does not in any way make against the view that sensitiveness arises from inter- 
mittence; for the very last stages of the discharge from the condenser are intermittent 
even to the eye, and it is not strange therefore that the stage immediately preceding 
should also be intermittent, though with a much more rapid period. That such is the 
true explanation of the phenomenon can also be shown directly, either by means of a 
rapidly revolving mirror, or by interposing a telephone in the circuit, as will be more 
fully described hereafter. 
Again, certain tubes appear to render the discharge from a continuous source sen¬ 
sitive without the necessity of artificially producing intermittence in the current. 
Such cases will, however, be found to offer no difficulties. They are merely instances 
of tubes which at higher tensions of the condenser give the phenomenon already 
described as being so common when the tension is low. No doubt, if the tension in 
the condenser, or perhaps, more strictly speaking, if the quantity of electricity 
discharged (strength of current) were still further increased, the sensitiveness could 
in all cases be made to vanish.* 
Again, a sensitive discharge may be produced by connecting one terminal of 
* It is worthy of remai’k, that the correlative proposition has been found to be in general true of the 
tubes examined by the authors of this paper for the purposes of this investigation. Though the greater 
number of them exhibit no trace of sensitiveness when the current is produced by the full power of a 
large twelve-plate Holtz machine, yet they can be made to show sensitiveness (without the introduction of 
an air-spark) by throwing some of the plates out of work, so as to reduce the quantity of electricity in the 
current. 
MDCCCLXXIX. 
Z 
