Cretaceous Pine Leaves. 
397 
The actual bundle is double , and is surrounded entirely by a sclerized 
zone or sheath, which forms a tongue of hard tissue between the two strands 
(see PL XXVIII, Fig. i, v.sc.). These sclerized cells are small, and have not 
such thick walls asthe hypodermic sclerenchyma. They are comparable to the 
sheath cells noted by Jeffrey (’ 08 , PL XIV, Fig. 17) in his Cretaceous Hard 
Pines. They are not represented in his Prepinus as splitting the bundle, 
but otherwise seem much like the inner thick-walled sheath he describes. 
Whether it is right to compare this sheath with that described by Stopes (’ 03 ) 
for Cordaites , is a point about which we wish to reserve our opinion for the 
present. 
The xylem in the two strands in our leaf is in small quantities, and the 
elements very small, o-oi mm. in diameter. The position of the phloem is 
apparent, but its cell-structure is not sufficiently well petrified for description. 
There is no indication of centripetal xylem or other unusual structure in the 
bundle. 
The transftisio7i sheath is a broad zone of closely packed, fairly uniform 
cells (see t, Phot. 1, PL XXVII, and Fig. 1, PL XXVIII) about 0-05-0*06 mm. 
in diameter. These fit closely together, have fairly thick walls, and appear to 
have been pitted, though only in a few cases are the pits still to be seen on 
their walls. There seems to have been no admixture of parenchyma cells 
with them. The outer zones of these cells may possibly be looked on as 
something of the nature of an endodermis : in the outermost row of cells the 
radial walls show an appearance suggestive of a slightly crushed endodermis, 
but the sheath is not a distinctive one. 
Jeffrey (’ 08 , p. 211) notes the absence of an endodermis in his Prepinus , 
but his photographs do not absolutely determine this point, and the sections 
of the American Prepinus which Prof. F. W. Oliver of London University 
kindly lent us do not determine it either. 
The reasons for placing our leaf in the genus Prepinus are as follows. 
Neither Jeffrey (’ 08 ) nor Hollick and Jeffrey (’ 09 ) give an actual diagnosis 
of their genus. When we pick out the more essential features from Jeffrey’s 
description, however, they are : (1) the contour of the leaf, with five approxi¬ 
mately plane surfaces, which differs from any known Pinus ; (2) the 
number of leaves in a fascicle; (3) the two resin-canals, placed laterally ; 
(4) the broad transfusion zone ; (5) the sheath of thick cells round the 
bundle ; (6) the single bundle ; (7) the large amount of centripetal wood in 
the bundle ; (8) the apparent absence of an endodermis ; (9) the presence 
of strong hypodermal isles beneath the epidermis. 
Jeffrey places much stress on the shape of the leaf and the correlation 
of this with the number of leaves in a fascicle, in which this fossil genus 
differs from any type of Pinus . In his view regarding this feature we 
entirely concur, and think that for diagnostic features it is the most 
important. 
■i 
