TRANSACTIONS OF SECTION C. 335 



which conduct their operations with method and skill, and are financially in a very 

 prosperous condition, others work in a more or less slovenly manner. In the 

 Central Provinces one of the mines is worked by Government, hut has not as yet 

 repaid the outlay uuon it. 



In round figures it may be stated that at present 1,000,000 tons of coal are 

 consumed in British India per annum in locomotives and factories, the quantity 

 employed in the form of coke for domestic purposes being inconsiderable ; and 

 that of this 1,000,000 tons, about one-half is raised from Indian mines, the other 

 coming from England, France, and Australia. 



6. On Geological Episodes. By J. F. Blake, M.A., F.G.S. 



Geological nomenclature was first founded on the theory of universal deposits - t 

 then the idea of lateral changes was introduced with the necessary misuse of litho- 

 logically descriptive names ; ultimately all deposits were seen to have their boundaries. 

 Beds deposited in distinct areas can thus be proved only homotaxial, and these 

 are by no means necessarily synchronous. The object of this paper is to show that 

 a somewhat similar principle ought to govern all our geological classification. A 

 single area is defined to be one over which we can trace one or more related for- 

 mations consecutively, and which formations contain identical characteristic fossils. 

 Deposits in single areas may be compared as to time and divided into life zones \ 

 but these in different areas are homotaxial only. In each single area the outlines 

 and characters of the several deposits must first be determined and denoted ac- 

 cordingly. 



In studying any group of rocks in a single area it is seen that some members 

 have a much wider range than others. Such differences in range are accompanied 

 by marked differences in character and point to differences in the circumstances of 

 deposit. The wide-spread formation indicates uniform changes of level over the 

 area and a mixture of deposits — such circumstances may be called normal. But 

 mere local changes may bring more restricted areas into peculiar physical condi- 

 tions. Such local changes may be called ' geological episodes,' and they will result in 

 the formation of deposits of marked character easily distinguishable from the normal. 



The first point is to determine the characters by which an episodal deposit may 

 be differentiated from a normal one. The supreme test is that derived from its 

 definition, i.e., its local development ; but if it be very small, it may be insignificant ; 

 if relatively very large, the distinction may be of no consequence. As a rule argil- 

 laceous rocks are normal, and arenaceous and calcareous episodal ; but this is by no 

 means universal. When the normal formation of a period is determined, the 

 episodes are marked by their differing mineral nature. The two kinds of deposits 

 may also be determined by the nature of their fossils, after we have first dis- 

 covered what kinds of fossils are usually episodal. For this purpose those fossils 

 which are found in all kinds of rocks, and therefore appear to have been indifferent 

 as to their physical surroundings, may be called invariant, and those found only 

 under particular conditions, and which change their locality as these conditions 

 change, covariant. Invariants only are suited for zonal classification ; covariants 

 are characteristics of episodes. A table is drawn up showing the classes, families, 

 and genera which may be covariant, according to the imperfect observations of the 

 past. The chief covariants are a few Foraminifera — the sponges — a large number 

 of Hydrozoa and Actinozoa, some Crinoids, the Blastoids, a few Lamellibranchs, 

 and at least half the Gasteropod families. 



The main proposition is that similar, but distinct episodes, in a normal series of 

 strata are neither necessarily nor probably of the same aye. The true method of 



feological classification is therefore to arrange only the normal deposits in a series 

 y their stratigraphy and their invariant fossils, while the episodes are put in their 

 place as such. 



These doctrines applied to British strata yield the following results : No episodes 

 are recognised in Cambrian or Pre-Cambrian rocks. In the Lower Silurian, the Dur- 

 ness limestone, the Llandeilo flags, the Bala limestone,and the Caradoc sandstone, and. 



