26 Mr. W. 8. Jevons on the Cirrous form of Cloud. 
We shall thus have portions of the upper stratum continually 
sinking into the lower, and corresponding portions of the lower 
rising through the upper; and this movement, as the experiment 
demonstrates, takes place by an interfiltration of minute, thread- 
like streams. 
[It is evident that the difference of temperature of the strata 
in this experiment is not a material point, being simply a means 
employed to enable us to lay one stratum upon another of a 
slightly greater density when of the same temperature, so that 
we may afterwards observe the mixing process and change of 
place in the most gradual manner possible. ] 
Exp. 2. Let the first experiment be now repeated in exactly 
the same manner, with the exception of adding the sugar to the 
lowest stratum instead of to the highest, as before. 
The appearances will now be totally different : but little cloud 
at all will be seen to form, even after a considerable length of 
time ; and whatever may happen to be caused by accidental dis- 
turbance will lie in a uniform or streaked flat sheet at the surface 
where it is produced, until it finally subsides to the bottom by 
its own density. 
These two experiments exhibit a most striking contrast ; and 
the only difference of conditions being in the inversion of the 
light and dense fluids, we are at once led to the conclusion, that 
different portions of liquids may, from the effects of very slight 
differences of specific gravity alone, be caused to mix and pass 
into each other in the form of minute streamlets, which, if ren- 
dered visible, as by the formation of a precipitate along their 
sides, present exact resemblances in form to the fibres of cirrous 
cloud. 
Now as gases are subject to the same laws of equilibrium and 
pressure as liquids, excepting only as far as they are modified 
by the property of elasticity, it is probable that strata of the 
atmosphere, which, being at perfect freedom to assume the den- 
sity due to the superincumbent pressure, will be further unaffected 
by their compressibility, must act otherwise; just like strata of 
very rare liquids, for instance, will, under the same circumstances, 
lie in quiet horizontal strata, or will displace each other violently 
or gradually, as the case may be. On this account it appears to 
me certain, that masses of air, in gradually displacing each other 
and mixing by reason of slight differences of density, will exhibit 
the same phenomena as we have seen to take place in liquids. 
It may perhaps be objected, that liquid cohesion has some 
hand in producing the thread-like appearance produced m the 
first experiment, and that my argument fails, since cohesion is 
non-existent in gaseous bodies. Now though liquid cohesion 
might tend to keep the particles of cach stratum together, and 
