It is only in the minority of cases that the accidents of preservation of fragments of ancient floras have given us the means of investigating the internal structure of the plant organs. It is far more frequently the case that fossil plants are represented only by a carbonised film on the surface of a piece of shale or other rock: the actual substance of the plant has been converted into a thin layer of coal, and though the venation and other surface-features may be clearly revealed, the internal tissues have been destroyed. If a lump of clay containing a piece of fern frond is heated, the result is an impression of the leaf on the hardened matrix and a coaly substance in place of the plant substance. It is occasionally possible by detaching a piece of the black film from a fossil, and heating it with nitric acid and chlorate of potash and then dipping it in ammonia, to obtain a transparent preparation suitable for microscopical examination of the cell-outlines of the superficial layer of the leaf or other plant-fragment. This method of examination, used by several students of fossil plants and with conspicuous success by Professor Nathorst of Stockholm, often affords valuable aids to identification. [Illustration: Fig. 7. Flood-plain of the delta of Rio Colorado. The hills in the background are 25 miles distant. (From a photograph by Prof. MacDougal.)] Pieces of plants embedded in sandy sediment, if not preserved by petrifaction, that is by the introduction into the tissues of some siliceous or calcareous solution, gradually decay and their fragmentary remains may be washed away by percolating water, leaving a hollow mould in the gradually hardening sediment, which is afterwards filled with sand or other material. The plant itself is destroyed, but a cast is taken which in the case of fine-grained sediments reproduces the form and surface-pattern of the original specimen. The incrustation of plants by the falsely named petrifying springs of Knaresborough and other places illustrate another method of fossilisation. [Illustration: Fig. 8. Flower of _Cinnamomum prototypum_ Conw. preserved in amber. × 10. (After Conwentz.)]