,0 PAINTING MATERIALS

weight. In no case, however, is the proportion of non-volatile products very large,
and undoubtedly the constituent to which a paint film owes its peculiar tenacity
and transparency is the addition product of the unsaturated glyceride with oxygen.
Chemically, there seems little doubt that the initial product is of the nature of
an organic peroxide; on the other hand, it seems that the peroxide is only a transi-
tional phase and does not exist to any degree in the final product. Many schemes
have been put forward in an effort to account for what takes place after the
initial formation of the peroxide. Although there is little or no experimental evi-
dence for the opinion, the most likely explanation is that rearrangement to a
hydroxyketone takes place, followed by subsequent polymerization:

-CH-CH- -> -C(OH)=C(OH)- -* -CHOH-CO-.
Studies on the mechanism of the formation of synthetic resins have recently
offered a new approach to the drying phenomenon in oils and natural resins.
Bradley maintains, from theoretical considerations, that the drying of oils is but
a typical manifestation of a more general phenomenon which consists of the
transformation of an organic substance from an essentially linear structure to the
so-called three-dimensional polymeric form (see Synthetic Resins). He has shown
that the oxygen convertibility of drying oils is closely related to their heat con-
vertibility and that convertibility, in general, is governed by the specific nature
and number of the reactive or functional groups of the oil molecules. The function
of the catalyst (i.e., lead, manganese, or cobalt linoleates — or resinates) is not
thoroughly understood either. It can be assumed that the metallic driers and the
initial peroxide product, in the absence of the driers, are both simple oxygen
carriers or catalysts for the reaction. It can also be assumed that the ' induction
period/ in the absence of a drier, represents the time taken to produce sufficient
* peroxide '-product in equilibrium with the rest of the system to act as the oxida-
tion catalyst. Long and his associates have found that little oxidation takes place
in a linseed oil film after it has set to a hard gel. The ultimate failure of these films
is not caused by continued oxidation; aging consists of a gradual transition of polar
liquid phase to solid phase of substantially the same ultimate analysis. Embrittle-
ment and failure of drying oil films is essentially a matter of reduction of the
percentage of liquid phase to low values.
Clewell has recently found that linseed oil surfaces, both dry and liquid, can
be studied by electron diffraction. He has followed polymerized oil through
different stages of drying and finds that the surface structure of a wet film is
completely amorphous. As the oil absorbs oxygen in the drying process, a gradual
orientation of carbon chain molecules, normal to the film surface, occurs. Com-
plete orientation does not exist until the film has completely dried. Alignment