576 Benson.—Mazocarpon or the Structural Sigillariostrobus. 
The main features can be recognized from Text-fig. i, but it is neces¬ 
sary to explain by what means the form of the bract has been arrived at. 
Sections are available tangential to the bract throughout the radially 
extended portion—that is, the series from which Fig. 8 is obtained. 
Horizontal sections are available in the transverse sections of two cones 
and from fallen sporangia (see Figs. 2 and 20). 
Sections approximately radial, together with a series of three sections 
in a plane at an angle of about 45 0 with the horizontal and radial planes 
(H. Cn. 530, 14-16), have all been brought into requisition. 
In harmony with all these sections we find there was a narrow lamina 
with a keeled midrib on the horizontal part to which the sporange was 
attached. In spite of the well-marked keel, the vascular bundle and 
parichnos did not run in the bract, but, as already stated, in the pedicel or 
radially extended stalk of the sporange. 1 
The bract thickened considerably just at the distal end of the sporangial 
attachment, the under surface sinking to form a convex cushion which, 
might be 8 mm. below the plane of the keel. Thus the convex base is not 
shown in Fig. 2. In this region there was no projecting ridge nor keel. 
The free lamina was hood-like in form and measured 6 mm. in the 
widest part. 
As the distal part overlapped the sporophylls above it on the cone we 
are able to determine that the width increased from 1 mm. to 2 mm., and 
then to 6 mm., tapering again to a point a little over 6 mm. above the bend. 
The erect part had an adaxial groove, so that, as there was no dorsal 
ridge, the middle line is the thinnest part of the bract. The merest trace of 
a vascular bundle has been detected near the adaxial surface in the groove. 
The whole of the abaxial surface of the bract is strengthened by 
a layer of thick-walled cells. The wide erect free part of the distal lamina 
thins out to the margin, but the narrow free part, in the transition region 
where the bract thickens* has the sclerized layer duplicated like a hem. 
The upper surface in the concavity (upon which lies the sporangial lamella) 
is covered with a delicate layer of cells with small lumen under which the 
tissue usually perishes. Traces of a ligule are seen (H* Cn. 530, 16) in 
a groove immediately distal to the sporange and beneath its lamella. At 
this level the bract is still narrow, so that the ligule may be regarded as 
occupying the distal end of the keel. It may be of interest to point out 
that the structure of the bract was such that any hygroscopic contraction of 
the outer surface would have tended to straighten or bend back the structure. 
If this took place in the cone where the convex surface fitted into the 
1 Dr. Scott has pointed out to me that a similar route of the vascular bundle is shown in 
Lepidostrobus Brownii, as was demonstrated by Zeilleiy Etude sur le Lepidostrobus Brownii , 1911, 
and in Lepidostrobus Fischeri (now Z. Kcntuckiensis). See Scott and Jeffrey : On Fossil Plants . . . 
of Kentucky. Trans. Roy. Soc., vol. ccv, 1914, p. 358. 
