122 CANADIAN: FOSSILS. 
axis. This curvature becomes less in the higher! cellules'; those of the 
middle open nearly parallel to the axis, and finally at the summit open 
in a direction opposite to those of the base. Cell-apertures mucronate, 
by the extension of the cell-partitions beyond the opening, in nearly their 
full width ; but seen upon the edge, they appear as setiform processes. The 
test is striated parallel to the cell-apertures, which have a concave outline. 
As the specimens lie upon the surface of the shale, the central portion, 
for about half a line in width, is usually rough, and the broken cellules 
are clearly distinguishable ; while on each side are the semi-oval divisions 
of the frond, with the cellules spreading from the central axis. 
The entire frond in reality consists of four semi-oval or semi-ovate 
folia, which are jomed rectangularly by their longitudinal axes, and in a 
transverse section present a regular cruciform figure. The expansions 
of the two sides, when laterally compressed, show distinct cellules with 
projecting mucronate extensions: those which are vertically compressed 
have the outer portions broken off in the separated lamine of slate, 
and present the bases of the cells; which have been sometimes filled with 
mineral matter, and distended before being imbedded. In a few instances 
the cells of the lateral portions are filled in the same manner, appearing 
as curving conical tubes with the broader extremities outwards. 
When the bases of the cellules of the upper or nearest of the folia 
remain, they are seen'to be directed obliquely downwards to the axis; but — 
sometimes in the process of separation these bases are removed wholly or 
in part, and the bases of the opposite folium are seen below the plane of 
the two lateral folia, or their impressions, which are spread out on the sur- 
face of the slaty lamina: these cellules are then clearly observed to be 
directed upwards, as we see them from below. 
It not unfrequently happens that this broad celluliferous axis is reduced 
to an undulating line, which results from compression in a direction oblique 
to the rectangularly-arranged folia, as in fig. 2, page 119, so that the two 
adjacent parts are spread out, and consequently no central line of cellules 
would be seen. When these have been divided longitudinally a little 
on one side of the centre, two sets of cellules are often seen penetrating 
the stone in oblique directions to the lamine of shale. 
The condition of preservation in several specimens examined is such as 
to render unavoidable the conclusion which I have given above, as to their 
mode of growth, however anomalous it may seem. 
This species differs from P. typus in its thicker substance, propor- 
tionally shorter and broader form, and more closely-arranged cellules. 
