1. THE ROCKS ERUPTED.--Mr. Allport has described the microscopic character of some of the toadstones,[29] and further details have been supplied by Mr. Teall.[30] The fullest account of the subject, however, is that given by Mr. Bemrose in the paper above referred to. This observer distinguishes the lava-form from the fragmental rocks, and gives the minute characters of each series. He does not, however, separate true interstratified lavas from injected sills, nor the bedded tuffs from the coarse agglomerates which fill up the vents. These distinctions are obviously required in order that the true nature and sequence of the materials in the volcanic eruptions may be traced, and that the phenomena exhibited in Derbyshire may be brought into comparison with those found in other Carboniferous districts. But to establish them satisfactorily the whole region must be carefully re-examined and even to some extent re-mapped. [Footnote 29: _Quart. Journ. Geol. Soc._ xxx. (1874), p. 529.] [Footnote 30: _British Petrography_, p. 209.] The lavas (including, in the meantime, sheets which there can be little doubt are sills) show three main types of minute structure and composition, which are discriminated by Mr. Bemrose as--(_a_) Olivine-dolerites; these, the most abundant of the series, consist of augite in grains, olivine in idiomorphic crystals, plagioclase giving lath-shaped and tabular sections, and magnetite or ilmenite in rods and grains; (_b_) Ophitic olivine-dolerites, consisting of augite in ophitic plates forming the groundmass, in which are imbedded idiomorphic olivine, plagioclase (often giving large lath-shaped sections and magnetite or ilmenite); (_c_) Olivine-basalts; these rocks are distinguished by containing crystals of augite and olivine in a groundmass of small felspar-laths, granular augite and magnetite or ilmenite, with very little interstitial matter. They have been noticed only in two of the outcrops of toadstone. The fragmental rocks have been shown by Mr. Bemrose to cover a much more extensive space than had been previously supposed. He has found them to be distinguished by an abundance of lapilli varying from minute fragments up to pieces about the size of a pea, and composed of a material that differs in structure from the dolerites and basalts with which the tuffs are associated. These lapilli consist largely of a glassy base more or less altered, which is generally finely vesicular and encloses abundant skeleton crystals and crystallites. The tuffs thus very closely resemble some of the Carboniferous basic tuffs of Fife, already referred to (vol. i. p. 422), and like these they include abundant blocks of dolerite and basalt.