C.—GEOLOGY 59 
the coal by about 1 per cent. The observed range of oxygen content 
in different coals is therefore many times greater than could be produced 
by any known variations of the plant constitution. This observation, 
taken in conjunction with the fact that the microscope fails to reveal any 
general difference of plant constitution between coals with high oxygen 
content and those with little, leaves no doubt that this aspect of the 
chemical history of the coal is determined by something other than the 
type of plant aggregate. It is this difference of oxygen content which 
chiefly denotes the rank of the coal, distinguishing the lignites, bituminous 
coals and anthracites. 
A very significant feature of this variation in rank is the fact that it is 
continuous throughout its long range. If any very large number of analyses 
of lignites, coals and anthracites be plotted, they will be found to form 
on the graph a very long and narrow belt which is unmistakably con- 
tinuous. The significance of this continuity is emphasized in an interest- 
ing way by the work of Prof. Bone and his colleagues, who have shown 
that the proportions of the different types of organic compounds present 
in the coal vary progressively throughout the series. 
Cannels and Anthracites—Cannel coals were excluded from the fore- 
going remarks on chemical composition merely because they form an 
extreme and relatively uncommon type. But just for this reason they 
afford the most striking illustration of the dependence of type on mode 
of origin and plant constitution, as also of the fact that in ultimate com- 
position it is the hydrogen which is most affected by these differences. 
Its increase in proportion to the number of algz in the mass is unmis- 
takable, and the figure of 12 per cent. reached in the pure algal cannels 
is sufficiently remarkable to ensure that no one will overlook its significance. 
Anthracites were omitted because in these very high-rank coals the rule 
that hydrogen content is independent of rank breaks down. In these 
coals the oxygen content has been reduced almost to the lowest possible 
limits, and as alteration proceeds further the hydrogen begins to be 
eliminated, and may be reduced to 2 per cent., or even 1 per cent. The 
last stage is graphite. That this hydrogen reduction is the result of a late 
stage of alteration is sufficiently proved by the fact that it is found only 
in those coals in which oxygen is reduced to 5 per cent. or less. Since 
a good deal of the discussion about the significance of rank in coal has 
centred round the anthracites, this fact that hydrogen elimination occurs 
only in the last stages has often been overlooked. For example, the use 
of the carbon : hydrogen ratio as a measure of rank is clearly permissible 
only in these high-rank coals ; but it has often been applied also to those 
of lower rank, in which, as we have seen, the hydrogen variation is ex- 
pressive of the plant constitution of the aggregate rather than of ‘ meta- 
morphic ’ changes. 
The Significance of Rank.—The thesis towards which all my remarks 
have been preparatory is that the rank of a coal is the measure of the 
alteration in composition which the deposit has suffered in consequence 
of the rise of temperature and increase of pressure resulting from its 
burial in the crust. I have been at some pains to distinguish between 
the differences in composition which denote rank and those due to the 
