OF HIGH EEFEAN GIBILITY UPON GASEOUS MATTER. 
341 
as with air. With hydrogen the clouds appeared more delicate and lustrous ; and they 
sometimes fell immediately after their formation in nebulous festoons to the bottom of 
the tube. This doubtless is to he ascribed to the attenuation of the atmosphere in which 
they floated. In many cases, however, the particles remained suspended, and some of 
them continued to float even after the tube had been so far exhausted as to produce a 
tolerably good air-pump vacuum. 
An additional effect of considerable beauty and interest is obtained in the following 
way. Permitting the convergent beam to play for a time upon the mixture of air and 
nitrite-of-amyl vapour, or, better still, upon mixture of hydrogen and vapour, a coarse 
cloud is formed. Suspending the action of the lamp for a minute or so, a new distribu- 
tion of the vapour appears to occur ; for, on reigniting the lamp, along its convergent 
beam, and within the old cloud, a new cloud is precipitated. The tint of this new cloud 
is a delicate bluish white, and its texture is of exquisite fineness. This precipitation of 
one cloud within another may be obtained a dozen times in succession. Or, permit- 
ting a parallel beam to pass for a time through the coarser cloud, on pushing out the 
lens so as to concentrate the light, the fine cloud comes suddenly down upon the beam 
about its place of greatest concentration. This effect also may be obtained several times 
with the same charge of vapour. 
No phenomena of the kind thus far described have, I believe, been hitherto observed. 
The necessary conditions for their production are, first, that the light should decompose 
the vapour, and secondly, that one or more of the products of decomposition should 
either be a solid, or should possess a boiling-point so high as to ensure its precipitation 
when set free. 
For though chemical action might occur, and be even energetic, if the products of 
decomposition be vaporous and colourless they will remain unseen. In the case just 
considered, the nitrate of amyl is in all probability a product of the decomposition of the 
nitrite. The boiling-point of the latter is estimated at from 91° to 96° C., that of the 
former being 149° C. The J nitrite, therefore, can maintain itself as true vapour in a 
space where the nitrate, at the moment of its liberation, must fall as a cloud. 
§ VI. 
An exceedingly fine example of actinic action is furnished by the vapour of the iodide 
of allyl. The effect of light upon this substance was observed on the 7th of October, 
1868, hut I did not then know the meaning of the “thin cloud like a kind of smoke” 
which showed itself in the experimental tube. On satisfying myself regarding the 
deportment of nitrite of amyl, the iodide of allyl occurred to me, and on it experiments 
were immediately made. 
The decomposition of this vapour was slower than that of the nitrite of amyl. The 
slowness, moreover, augmented rapidly as the quantity of vapour was diminished. 
When only a few inches of the mixed air and vapour were in the experimental tube the 
action was very slow. The clouds were formed both in oxygen and in air. After the 
mdccclxx. 2 z 
