If, then, lignites have not become soft coal, and if the latter has not become anthracite, it is not that time was wanting, but climatic conditions and environment.  Most analyses of specimens of coal have been made up to the present with fragments so selected as to give a mean composition of the mass; it is rare that trouble has been taken to select bits of wood, bark, etc., of the same plant, determined in advance by means of thin and transparent sections in order to assure the chemist of the sole origin and of the absolute purity of the coal submitted to analysis.  This void has been partially fitted, and we give in the following table the results published by Mr. Carnot of analyses made of different portions of plants previously determined by us: 

Carbon Hydrogen Oxygen Nitrogen
1.  Calamodendron (5 specimens) 82.95 4.78 11.89 0.48 2.  Cordaites (4 specimens) 82.94 4.88 11.84 0.44 3.  Lepidodendron (3 specimens) 83.28 4.88 11.45 0.39 4.  Psaronius (4 specimens) 81.64 4.80 13.11 0.44
\——­v——­/
5.  Ptychopteris (1 specimen) 80.62 4.85 14.53 6.  Megaphyton (1 specimen) 83.37 4.40 12.23

As seen from this table, the elementary composition of the various specimens is nearly the same, notwithstanding that the selection was made from among plants that are widely separated in the botanical scale, or from among very different parts of plants.  In fact, with Numbers 1 and 2 the analysis was made solely of the wood, and with No. 3 only of the prosenchymatous and suberose parts of the bark.  Here we remark a slight increase in carbon, as should be the case.  With No. 4 the analysis was of the roots and the parenchymatous tissue that descends along the stem, and with No. 6 of the bark and small roots.  One will remark here again a slight increase in the proportion of carbon, as was to be foreseen.  The elementary composition found nearly corresponds with that of the coal taken from the large Commentry deposit.

Carbon.   Hydrogen.  Oxygen and
Nitrogen. 
Regnault     82.92     5.39      11.78
Mr Carnot    83.21     5.57      11.22

Although the chemical composition is nearly the same, the manner in which the different species or fragments of vegetables behave under distillation is quite different.

In fact, according to Mr. Carnot, the plants already cited furnish the following results on distillation: 

Volatile Fixed Coke.
matters. residue. 
Calamodendron 35.5 64.7 Well agglomerated.  Cordaites 42.1 57.8 Quite porous.  Lepidodendron 34.7 55.3 Well agglomerated.  Psaronius 29.4 60.5 Slightly porous.  Ptychopteris 39.4 60.5
Megaphyton 35.5 64.5 Well agglomerated.  Coal of the Great Bed 40.5 59.5 Slightly porous.

These differences in the proportions of volatile substances, of fixed residua, and of density in the coke obtained seem to be in harmony with the primitive organic nature of the carbonized tissues.  We know, in fact, that the wood of the Calamodendrons is composed of alternately radiating bands formed of ligneous and thick walled prosenchymatous tissue, while the wood of Cordaites, which is less dense, recalls that of certain coniferae of the present day (Araucariae).