ed that at the moment of separation of any such particle it is
entirely surrounded by other particles of a _different_ kind with which it
is in close contact, and has not yet assumed those relations and conditions
which it has in its fully developed state, and which it can only assume by
association with other particles of its own kind. For, at the moment, its
elasticity is absent, and it is in the same relation to particles with
which it is in contact, and for which it has an affinity, as the particles
of oxygen and hydrogen are to each other on the surface of clean platina
(626. 627.).

659. The singular effects of retardation produced by very small quantities
of some gases, and not by large quantities of others (640. 645. 652.), if
dependent upon any relation of the added gas to the surface of the solid,
will then probably be found immediately connected with the curious
phenomena which are presented by different gases when passing through
narrow tubes at low pressures, which I observed many years ago[A]; and this
action of surfaces must, I think, influence the highly interesting
phenomena of the diffusion of gases, at least in the form in which it has
been experimented upon by Mr. Graham in 1829 and 1831[B], and also by Dr.
Mitchell of Philadelphia[C] in 1830. It seems very probable that if such a
substance as spongy platina were used, another law for the diffusion of
gases under the circumstances would come out than that obtained by the use
of plaster of Paris.

[A] Quarterly Journal of Science, 1819, vol. vii. p. 106.

[B] Quarterly Journal of Science, vol. xxviii. p. 74, and Edinburgh
Transactions, 1831.

[C] Journal of the Royal Institution for 1831, p. 101.

660. I intended to have followed this section by one on the secondary piles
of Ritter, and the peculiar properties of the poles of the pile, or of
metals through which electricity has passed, which have been observed by
Ritter, Van Marum, Yelin, De la Rive, Marianini, Berzelius, and others. It
appears to me that all these phenomena bear a satisfactory explanation on
known principles, connected with the investigation just terminated, and do
not require the assumption of any new state or new property. But as the
experiments advanced, especially those of Marianini, require very careful
repetition and examination, the necessity of pursuing the subject of
electro-chemical decomposition obliges me for a time to defer the
researches to which I have just