LEAF IN THE VASCULAR CRYPTOGAMS AND GYMNOSPERMS. 
581 
fig. 15). They are at first similar in every external aspect to the pinnse of Cycos 
of like age. Examination of the apices of the pinnse showed that there is no single 
apical ceil, nor are there, at least in early stages of development, any marginal series 
of cells, like those described for the leaves of the Leptosporangiate Ferns. The num¬ 
ber of pinnse formed is small as compared with the latter, and the pbyllopodium, at 
least in some cases if not in all, terminates abruptly in a blunt cone, or terminal 
spine, such as is found so frequently among the Cycadacece (figs. 16 and 17).* 
This abrupt ending of the phyllopodium may sometimes be still recognised in the 
mature leaf, but it is usually obliterated by the more rapid growth of the parts im¬ 
mediately below it. When the succession of the pinnse is clearly alternate, an 
appearance is then produced as though the development of the leaf were sympodial, 
and the last lateral pinnse might be mistaken for the apex of the phyllopodium : 
observations of the development show however that this is not the case. 
The pinnse themselves assume the circinate curvature at an early stage (Plate 38, 
fig. 16) : as their development proceeds they may either become themselves winged, 
as is the case in young or weak buds ; or they may form pinnules, which also arise 
alternately, and in acropetal order, and are formed monopodially. Towards the apices 
of those pinnse which form pinnules, a similar appearance as of a sympodial develop¬ 
ment is often found, and it may be explained as in the case of the phyllopodium. 
It is well known how frequently transitional forms are found between the winged 
development of pinnse, and pinnules, and even higher forms of ramification, the latter 
being assumed by stronger plants : the result is a great irregularity of outline of the 
whole leaf. 
The phyllopodium itself is but slightly winged, if at all: above the stipules it 
appears almost cylindrical, but in the upper parts of it traces may often be found of 
longitudinal markings where the wings would normally be situated, but the winged 
development is chiefly relegated to the branchings of higher order. WTren a winged 
development of a pinnse or pinnule begins, the central mass of tissue of the member 
is already beginning to pass over to the condition of permanent tissue : active 
meristematic division is still continued at those peripheral points where the wings 
arise. Here the cells are arranged somewhat after the manner of the Leptosporangiate 
Ferns; but whereas in these the marginal cells constitute a simple linear series, in 
Angiopteris the structure is much more complicated ; it is as though each cell of the 
marginal series of a leaf of one of the simpler Ferns were divided up by a number of 
anticlinal walls into cells of small and uniform size, arranged in regular series extending 
from the lower to the upper surface of the pinna or pinnule. Thus a transverse 
section would follow one of these series, as represented in Plate 38, fig. 19, in which 
may be seen in place of a single marginal cell, such as is seen in the transverse section 
* In some examples of Marattia tlie phyllopodium is not thus abruptly terminated, but it is continued 
as an elongated, winged structure ; the same is the case in Danoea. 
