KIDSTON : THE FLORA OF THE CARBONIFEROUS PERIOD. 355 
enfeebled conditions, the leaf-scars become approximated (Sigillaria 
Sauveuri Zeiller, Plate LIV., fig. 4). There are, therefore, really only 
two sections, those with ribs and those with smooth or unribbed stems. 
Those species with the cones arranged in two vertical rows are 
few in number. Sigillaria discophora Konig sp, { = Ulodendron 
minus L. and H.) (Plate LX., fig. 1) is frequent in the Upper Carbon- 
iferous, while another Ulodendroid form, Sigillaria Taylori Carr sp., 
occurs in the Lower Carboniferous. 
The leaves in the great majority 
of Sigillarice are long and grass-like, 
with a single nerve, but a few had 
shorter lanceolate leaves. 
The cones of Sigillaria (Sigil- 
lariostrobus) have hitherto always been 
discovered separated from their parent 
trunks though frequently attached to 
their pedicels, and from one such 
showing the Sigillarian leaf-scar, M. 
Zeiller was able to prove their Sigil- 
larian origin. The Sigillarian origin 
of these cones had long been suspected, 
but satisfactory proof was wanting. 
Owing to our inability of asso- 
ciating these cones with the species 
to which they belong they are placed 
in the genus Sigillariostrobus (Plate 
LV., fig. 4). These cones differ essen- 
tially in their structure from those of 
Lepidostrobus. In Lepidostrobus the 
sporangia are placed on the pedicel- 
like portion of the bract, whereas in Sigillariostrobus the sporangia 
are developed within the inflated and hollowed-out substance of 
the base of the bract. Text Fig. 9. The cones of Sigillaria were 
probably heterosporous ; at least one specimen I possess points to 
this conclusion, but owing to the imperfection of this example, the 
opinion requires confirmation before being definitely adopted, 
c 
Fig. 9. — SigiUiariostrohiis cili- 
atus Kidstoii. A, two spor- 
angia containing macrospores 
( X 4). B, restoration of spor- 
angium—a, axis ; b, wall of 
sporangium. 
