Coals with similar properties are not found at haphazard amongst a number having widely different compositions , but are grouped together by reason of similar ultimate analyses .sx Nevertheless , the correlation of the ultimate compositions of coals , from all sources , with their properties is far from complete , and , as far as it goes , is not as precise as could be desired .sx Reasons for this appear when the constitution of coal is studied more closely .sx Coal .sx The term coal is conveniently restricted in normal British usage to the black varieties of technical importance .sx Else-where , the term may be , and is , extended to include such fuels as lignites and brown coals , when the deposits are of such an extent and nature as to be of value commercially .sx Carbonaceous materials , such as peats at the one end of the scale of carbon-contents and graphites at the other , are not usually referred to as " coals .sx " Nevertheless , by reason of the close relationship that exists between them , it is desirable to consider them in any discussion of the constitution of coal .sx The relationship between successive members of the series peats , brown coals , lignites , bituminous coals , anthracites and ( perhaps ) graphites has long been recognised .sx Each is derived from purely vegetable accumulations .sx Each covers a closely graded range of materials , and there is no sharp line of demarcation as between one member and the next .sx From Peat to Anthracite .sx There has thus naturally arisen a belief that the " coal " fuels form a continuous series in which the individual members differ one from another , not because of differences in the materials from which they were originally formed , nor even because of differences in their mode of formation , but because they represent successive stages in a slow process of " coalification .sx " This " peat-to-anthracite theory , or modifications of it , is now widely held , though it has been hotly debated in the past .sx Microscopical Evidence .sx Evidence in support of the peat-to-anthracite theory is to be found in the readily demonstrable similarity of type amongst the materials from which the fuels were formed .sx Aside from such preferential accumulations of certain types of plant structures as characterise a few special coals , similarity in the structures and in their distribution has been demonstrated , for example , by Thiessen ( U.S. Bureau of Mines , Bull .sx 117 , 1920 ) as between the peats , brown coals , and bituminous coals of North America ; and by Seyler ( Fuel ; 1923 , 2 , 217 ) as between the bituminous coals and anthracites of Great Britain .sx The persistence of the same types of resistant plant-tissues , such as spore exines and .sx cuticles , changed little by decay , is a feature common to all members of the series .sx Resins also , though not so resistant , have been shown ( by David White ( U.S. Geol .sx Survey , Prof .sx Paper 85 , part E , 1914 ) , .sx for example ) to persist in almost their original physical forms throughout the series up to , at least , the semi-anthracite coals .sx Chemical Evidence .sx There is a similarity in the chemical nature of the materials of which each fuel is compounded .sx The noticeable differences in properties between successive members of the series can be related , chemically , to minor changes in the external groupings of the compounds present .sx The internal structures of the molecules , whether the compounds are contained in a peat or an anthracite , appear to be much the same .sx Analytical Evidence .sx Perhaps the most striking evidence of the continuity of the " coal " series is to be found in analytical data , such as have been adduced by Seyler ( Fuel , 1924 , 3 , 15 ) , Ralston ( U.S. Bureau of Mines , Tech .sx Paper 93 , 1915 ) , Drakeley ( Trans .sx Chem .sx Soc. , 1922 , 121 , 221 ) , and , more recently , by Hickling ( Trans .sx Inst .sx Min .sx Eng .sx , 1927 , 72 , 261 ) , in a noteworthy paper to which reference should be made .sx In conformity with the practice of the earlier investigators , Hickling has plotted , on rectangular co-ordinates , the elementary compositions of a large number of coals , and has shown that they occupy a narrow and continuous band over the whole range .sx The proportions of nitrogen and of sulphur appear to be constant through-out the range , varying but little from 1.5 and 1.0 per cent .sx respectively .sx The proportion of hydrogen in the coals ( excluding the cannels and the anthracites of highest carbon-contents ) varies between 4.5 and 5.5 per cent .sx , that is to say , by only 10 per cent .sx ; on the mean value 5.0 per cent .sx The main variation in composition falls , therefore , on the carbon and the oxygen , and the variations in these two elements are nearly complementary .sx The carbon varies between 50 per cent .sx in the peats and 95 per cent .sx in the anthracites .sx Of this total variation of 45 per cent .sx , 17 per cent .sx covers the groups of bituminous coals and anthracites .sx The approximate constancy of the proportions of nitrogen , sulphur , and hydrogen is expressed by the narrowness of the band when the carbon-contents are plotted against the oxygen-contents of the coals ( see Fig. 44) .sx That the band should be narrow during the early ( low-carbon ) stages is an indication that the hydrogen-content of the original coal-forming materials did not vary much from a mean value .sx That the band should remain narrow up to the high-carbon end of the range is evidence that similar changes have operated in the formation of all the coals , for any marked variation in the character of the coalification would have resulted in products of markedly different hydrogen-content for a given carbon-content .sx Thus , at any period during the coalification process , the changes appear to be of definite and unvarying character , dependent on the nature of the material , and not on the immediate conditions to which it is subjected .sx No doubt the immediate conditions would affect the rapidity of the changes .sx A consideration of the distribution curve of coal analyses leads to a further suggestion .sx The curve , with carbon and oxygen percentages as axes , is at the outset nearly straight .sx This suggests .sx the gradual progress of but one kind of change .sx Assuming that the only kind of change , affecting ultimate composition , that can take place during progressive coalification is one involving loss of material , Hickling shows that the straight part of the distribution curve con-forms with a gradual loss of carbon dioxide and water .sx The first straight portion of the curve ends at about 83 per cent .sx carbon , giving place gradually to a second straight portion over the range of 89 to 95 per cent .sx carbon .sx This range covers but a slight loss of .sx oxygen , with a proportionately higher loss of hydrogen , perhaps as methane .sx There appears to be a scarcity of coals of about 86 per cent .sx carbon-content ( i.e. , at the juncture of the two straight portions of the distribution curve ) , otherwise the distribution over the whole range is remarkably uniform .sx The Formation of Coal .sx The validity of the peat-to-anthracite theory is not universally admitted .sx It is maintained , for example , by some geologists that the process by which any given coal has been formed has not necessarily been slow and gradual .sx If this is so , it follows that each class , or each individual , in the series has been the outcome of individual coal-forming conditions .sx For example , Strahan and Pollard ( Mem .sx Geol .sx Survey , England and Wales , 1915 ) have suggested that the partial anthracitisation of the South Wales .sx coalfield is due to differences in the nature of the deposited plant materials , a suggestion which has little to support it .sx Donath ( Brennst .sx Chem .sx , 1922 , 3 , 231 ) has long maintained the view that for the formation of a bituminous coal a high temperature , involving partial distillation , is essential .sx A more moderate view , for which Erdmann ( Brennst .sx Chem .sx , 1924 , 5 , 177 ) has made out a plausible ( but by no means flawless ) case , is that , whilst brown coals have resulted from coalification at normal temperatures , a temperature of at least 300 C. is required for the formation of a bituminous coal .sx Both these suggestions fail to account for the regular distribution of coals over the wide range of carbon-contents .sx It seems probable that , whilst a moderate increase of temperature can be regarded as one factor in the process of coalification , it would influence rather the rapidity of the process than its end-point .sx A new theory of coal-formation , definitely opposed to the peatto-anthracite theory , has recently been put forward by McKenzie Taylor ( Fuel , 1928 , 7 , 203 ) , based on the following novel observations .sx Beneath an impervious alkaline layer , such as is formed by the hydrolysis of sodium alumino-silica soils , resulting from base-exchange , bacterial decomposition of vegetable debris proceeds rapidly and to completion , mainly with the elimination of oxygen .sx This is due chiefly to the maintenance of mildly alkaline conditions , the acid products first formed , which would otherwise bring the decomposition to a premature close , being continuously neutralised .sx Whilst most peats are formed under aerobic conditions , the accumulation of acid products limiting the extent of the decay , examination of the roofs of bituminous coal and anthracite seams shows that they have , in general , an alkaline reaction , and that the coals have been formed under alkaline anaerobic conditions .sx This differentiates them from the lignites and brown coals , for an examination of the roofs of their seams shows that they have not been formed under anaerobic conditions and that they may have undergone incomplete bacterial decomposition and partial oxidation .sx McKenzie Taylor's view of coal formation , therefore , is that the vegetable material has first undergone a peat-like decay , under aerobic , acidic conditions .sx The duration of the extent of this " peatification " determines the nature of the product .sx If the decay is checked , as under a normal acidic or neutral cover , the peat deposit may ultimately form a lignite , but it can never become a bituminous coal .sx On the other hand , if the material in its particular state of arrested peat-decay is maintained under alkaline , anaerobic conditions , e.g. , under a cover of alkaline soils , a secondary decomposition will ensue .sx This , the " coalification " process proper , is rapidly carried to completion , the end-product being a coal the nature of which depends not at all on the geological conditions , nor on time , temperature and pressure , but solely on the degree of " peatification " before " coalification " began , each class of coal , sub-bituminous , bituminous or anthracitic , being a definite end-product .sx This view , whilst differing fundamentally from the peat-to- .sx anthracite theory , where each coal is regarded as an intermediate product leading progressively to the anthracites of high carbon-content , yet accepts and attempts to explain the gradual transition in composition observable throughout the range .sx It is early to judge as to the ability of McKenzie Taylor's theory to explain the apparently restricted and uniform distribution of coal compositions , for more data are required , but of the importance of the theory there should be no question .sx The discovery of the part played by a cover of alkaline soil in determining the extent of bacterial decay of vegetable debris provides a solution to the hitherto puzzling differences observed in the degree of preservation of particular plant entities , both in peat and in coal .sx The pronouncement that bituminous coals have of necessity been formed under an alkaline cover demands close consideration .sx To sum up , whatever may be the agencies which have advanced or limited the degree of " coalification , " there can be no doubt but that the different members of the series , peat-to-anthracite , represent the effects of increasingly prolonged or increasingly severe conditions on materials generally similar .sx The Coal-forming Materials .sx The nature of the changes that have taken place during coal formation can be followed roughly from the changes in composition of the coal-forming materials .sx Thus , as Hickling has pointed out , the transition from the original materials to lignites has involved a proportionate loss of hydrogen , probably as methane ; the bituminous coals differ from the lignites as their carbon-content increases in that they have suffered an increasingly greater loss of carbon dioxide and water ; whilst to form the anthracites , elimination of hydrogen , perhaps as methane , has again occurred .sx More precisely , the nature of the changes can be deduced from a study of the behaviour of different plant entities under the coal-forming agencies .sx The original plant materials , though botanically of widely varied source and form , were , from a chemical standpoint , not so diverse .sx