ON THE SENSITIVE STATE OF VACUUM DISCHARGES. 
625 
shadows, we might expect that it would also give rise to virtual shadows. And such 
is found to be the case, although the effects are decidedly more feeble than in the case 
of positive relief. They are, however, generally present, and under favourable cir¬ 
cumstances are clear and distinct (Plate 29, fig. 31). They have not formed the 
subject of any special examination as yet, but so far as can be determined by inspection 
they do not materially differ from those due to positive relief in any other respects 
than would naturally flow from the inferior intensity of the action to which they 
are due. 
Thus we see that positive special effects are clearly distinguished from relief effects 
by the attraction of the positive luminosity and the appearance of phosphorescence on 
the tinfoil. It is true that they both produce virtual shadows and phosphorescence 
on the glass opposite to the tinfoil, but these are given only in a feebler degree, and as 
secondary phenomena, by the positive special. It is worthy of remark that in the 
latter case we have the phosphorescence both on and opposite to the tinfoil, a pheno¬ 
menon of which no instance will be found in the experiments on interference which 
have hitherto been described, and which at first sight appears to contradict the laws 
of interference of molecular streams already established (see page 609). It is probable 
that the solution of the difficulty is, that the arrival of the molecules which causes the 
one is not synchronous with the discharge which causes the other, but the authors of 
the present paper have been unable to come to any definite conclusions as to the way 
in which this occurs. 
XXV.— On the relief and special effects in tubes of high exhaustion with a negative 
air-spark. 
We have already noticed that the main peculiarity of discharges in high as com¬ 
pared with low vacua is the prominence of the special characteristic of negative 
discharge, viz. : molecular streams, and the comparative insignificance of the special 
characteristic of positive discharge, viz.: positive luminosity. So much is this the case 
that while in the tubes of low exhaust we found it necessary to rely chiefly on the 
effects of positive discharges, as they alone rendered themselves plainly visible, we are, 
in the case of tubes of high exhaust, compelled to rely chiefly on the effects of negative 
discharges to guide us in our investigations. 
It follows naturally from these considerations that negative-relief is a comparatively 
uninteresting subject of investigation, for all the impulsive discharges produced 
thereby being discharges of positive luminosity, are extremely feeble in their lumi¬ 
nosity, if not practically invisible. It is in fact exactly similar to positive special, 
except that the unaffected state of the discharge is different, the luminosity in the case 
of a negative air-spark being in the form of a diffused haze, and not in the form of a 
thin pencil-like column. And even this difference contributes to make the phenomena 
of negative-relief still less impressive and striking. 
But, as we have already said, it is not only the negative-relief that has its counter- 
