horizontal and parallel to the bedding. Hence they increase
the stratified appearance of natural exposures of chalk.

It will be seen from the details given above that concretions may be
calcareous, siliceous, argillaceous and phosphatic, and they may consist
of carbonate or sulphide of iron. In the red clay of the deep sea bottom
concretionary masses rich in manganese dioxide are being formed, and are
sometimes brought up by the dredge. In clays large crystals of gypsum,
having the shape of an arrow-head, are occasionally found in some
numbers. They bear a considerable resemblance to some concretions, e.g.
crystalline marcasite and pyrite nodules. These examples will indicate
the great variety of substances which may give rise to concretionary
structures.

Some concretions are amorphous, e.g. phosphatic nodules; others are
cryptocrystalline, e.g. flint and chert; others finely crystalline, e.g.
pyrites, sphaerosiderite; others consist of large crystals, e.g. gypsum,
barytes, pyrites and marcasite. From this it is clear that the formation
of concretions is not closely dependent on any single inorganic
substance, or on any type of crystalline structure. Concretions seem to
arise from the tendency of chemical compounds to be slowly dissolved by
interstitial water, either while the deposit is unconsolidated or at a
later period. Certain nuclei, present in the rock, then determine
reprecipitation of these solutions, and the deposit once begun goes on
till either the supply of material for growth is exhausted, or the
physical character of the bed is changed by pressure and consolidation
till it is no longer favourable to further accretion. The process
resembles the growth of a crystal in a solution by slowly attracting to
itself molecules of suitable nature from the surrounding medium. But in
the majority of cases it is not the crystalline forces, or not these
alone, which attract the particles. The structure of a flint, for
example, shows that the material had so little tendency to crystallize
that it remained permanently in cryptocrystalline or sub-crystalline
state. That the concretions grew in the solid sediment is proved by the
manner in which lines of bedding pass through them and not round them.
This is beautifully shown by many siliceous and calcareous nodules out
of recent clays. That the sediment was in a soft condition may be
inferred from the purity and perfect crystalline form of some of these
bodies, e.g. gyp