404 SPHENOPHYLLALES. A. SPHENOPHYLLEAE 
were united at the base into a narrow sheath or collar surrounding the 
axis. The transition to the strobilus is plainly seen at the lower limit of 
the large specimen from the Brussels Museum, described and figured by 
Kidston: towards the upper limit of the specimen, where the sporangia 
cease, the axis is continued in the vegetative manner, with longer internodes. 
These facts plainly point to the absence of a highly differentiated strobilus, 
and. the existence in this species of a “ Se/ago” condition, where the fertile 
region is a mere zone on a continued axis. Not 
only does S. majus stand as yet alone in the genus 
by the indefiniteness of its cone, but also in the 
character of its sporangiophores. One of these is 
borne near to the base of each forked sporophyll 
(Fig. 222): the sporangia, which are 4-6 in 
number, but usually four, are grouped round a 
central attachment; and though no_ elongated 
pedicel can be seen, still the fact that when they 
are removed from the bracts they still remain 
united in groups of four to six indicates that 
Sphenophyllum Roemeri. Dia- 
grammatic sketch of the arrange. they had a common base. In favourable cases 
Rilcton’) the sporangia. (After 1 idston has been able to demonstrate that a 
radial line of dehiscence is clearly marked, 
corresponding in position to that of the synangium of Psélotum, to which 
the whole structure shows a remarkable resemblance. As a last type, and 
not the least remarkable of this variable genus, may be mentioned the 
fructification of S. fertile, recently described by Scott.1 It is characterised 
by the fact that both the “dorsal and ventral lobes are fertile,” by which 
is meant that the bract bears sporangia as well as the sporangiophore, 
which it subtends. Dr. Scott remarks that this is “more probably due to 
special modification than to retention of a primitive condition.” With 
this opinion I readily concur, adding the further comparison of this 
condition with the common variation of Botrychium Lunaria, where the 
sterile leaf is often partially, or even completely fertile (compare Fig. 85). 
Lastly, there remains to be described that remarkable cone from the 
Calciferous Sandstone of Burntisland, named by Scott Cheirostrobus, and 
placed by him in relation to the Sphenophyllales, while recognising also 
its affinities with the Equisetales and Lycopodiales.2 The vegetative system 
of the plant of which this is the fructification is still unknown. The cone 
itself is of large size, and shows greater complexity than any of the known 
sporangiophoric types. The robust axis shows structural characters sug- 
gestive of a Lycopodinous rather than of a Sphenophylloid affinity: the 
central stele in transverse section has a solid star-shaped xylem-core, with 
twelve projecting protoxylem-groups, corresponding to the series of sporo- 
Fic. 220. 
1 Proc. Roy. Soc., Dec., 1904, and 422. of Bot., xix., p. 168, also Progressus Ret 
Botanica, i. p. 151. 
2Scott, Phil. Trans., vol. 1898, 1897, ‘On Cheirostrobus.” 
