418 MISS M. BENSON--CONTRIBUTIONS TO THE 
two halves of the ovule which cause the very characteristic curvature of the ovules of 
Alnus, Betula, Corylus, and Carpinus. The position is strongly contrasted with that of 
Fagus and Quercus, although these ovules are also anatropous; but, at the same time, 
resembles what, from Treub’s figures, we observe to be the case in Casuarina (plates 
xxii. and xxiv.) *. 
From the difference in the one case and the agreement in the other we are, I think, 
entitled to regard this feature as a special adaptation for chalazogamy. 
I have given two figures of pollen-tubes in contact with the base of the embryo-sac 
ceca ( Corylus and Carpinus). In the case illustrated by figs. 46 and 47 (РІ. LXXI.) we 
find the tube abutting on the comparatively short caecum of Corylus. In figs. 44 
and 45, which also exhibit the tube ascending the nucellus, the lower part of the 
tube alone is represented. 
In figs. 55 and 56 (Pl. LXXII.), which illustrate a similar case in Carpinus, contact has 
taken place quite at the base of the nucellus. These instances, and especially the perfo- 
ration of the wall of the embryo-sac at the spot abutted upon by the pollen-tube, as 
shown in fig. 55, prepare us for the next stage, which we see delineated in figs. 50 and 51. 
The pollen-tube enters the base of the fertile embryo-sae іп Corylus and Carpinus, and, con- 
tinuing its course up the whole length of the czecum, eventually reaches and fertilizes 
the oosphere by direct contact. Whether this has 'already surrounded itself with a cell- 
wall or not I have not conclusively determined, although many of the preparations so far 
made suggest this to be the case, as I have repeatedly found a unicellular egg-cell 
clothed with a cell-wall where I have failed to demonstrate any vestige of a pollen-tube ; 
and yet fig. 50 proves it capable of persisting until the oospore has twice divided. Тһе 
inconspicuous development of the synergidze and their early dissolution are now explained. 
They are not required to assist in any way the act of fertilization, for the pollen-tube 
reaches the oosphere from below, having previously entered the embryo-sac in its basal 
region. 
Fig. 32 taken with fig. 34 (Pl. LXX.) has enabled me to construct the course taken by 
the tube in the nucellus of Alnus as given in the diagrammatic drawing, fig. 38. From the 
chalaza it runs up more or less by the tiers of large cells which constitute the stalk-like 
prolongation or pedicel (Treub) of the sporogenous tissue already described. It leaves 
this, and ascends to a level considerably above the apex of the embryo-sac, and then, 
turning down again, presents the appearance of a tube pursuing a normal course. It 
was doubtless this last part of the course that was observed by Hofmeister. 
Branching. —The pollen-tube presents another point of likeness with that of Caswarina, 
inasmuch as it sends out a short recurved branch on entering the nucellus (fig. 36, b.). 
This branching of the pollen-tube is a very general feature in the Amentifere. Even 
in those genera in which we do not find the chalazal course of the tube, we find a process 
of bifurcation still very widely present. Schacht, in the above-mentioned paper, gives 
drawings of the pollen-tubes of Pagus. I include a figure of the tubes of Quercus 
(РІ. LXIX. fig. 28). Тһе character of the branching is, in Fagus, the simple bifurcation 
of the tube into what appear to be equivalent tubes, but in the case of the chalazogamic 
* Treub, loc. cit. 
