THE PRIMITIVE VEGETATION OF THE EARTH. 481 
clay," which is a fossil soil on which a forest of Sigillariz 
has grown, and the remains of these trees are very abundant 
in the coal and the accompanying beds. Hence the Sig- 
illariæ of the coal-period are regarded as the plants most 
important in the accumulation of coal. ln the Devonian, as 
far as we yet know, they did not attain to this utility, and in 
the lower part of the system at least, the rhizomata of Psil- 
ophyton seem to have occupied tbe place afterwards held by 
the Stigmarie. In connection with this it is to be remarked 
that the Sigillarie of the Erian period seem to have been 
few, and of small dimensions in comparison with those of 
the coal. 
Rising still higher in the vegetable kingdom, and arriving 
at unquestionable Gymnosperms, we find in the Devonian of 
Eastern America, and also, I believe, in that of Scotland 
and Germany, trunks which may be referred to Conifere. 
In the Middle and Upper Devonian these present the struc- 
ture of modern Araucarian pines, or that modification of it 
belonging to the Carboniferous trees of the genus Dadoxy- 
lon. In the Lower Devonian we have what seems to be a 
simplification of the Coniferous structure, in the cylindrical 
wood-cells, marked only with spiral threads, found in the 
genus Prototaxites. These trees are very abundant as drift 
trunks in the Lower Devonian, down almost to its bottom 
beds, and sometimes attain to a diameter of three feet. 
Though of a structure so lax that it is comparable only with 
the youngest stems of ordinary Conifers, these trees must 
have been durable, and they are furnished both with medul- 
lary rays and rings of annual growth. Unfortunately we 
know nothing of their foliage or fruit. 
But for one little fragment of wood we should have had 
no indication of the existence in the Erian of any trees of 
higher organization than the Conifers. This fragment, found 
by Professor Hall at Eighteen-mile Creek, Lake Erie, has 
the, dotted vessels characteristic of ordinary Exogens, and 
unquestionably indieates a plant of the highest kind of 
AMER. NATURALIST, VOL. 1V. 61 
