AND ON POLYEMBRYONY IN THE HIGHER PLANTS. Laff 
wards the middle of April that a distinctly perceptible cavity 
made its appearance ; it became filled with a glutinous mucus, 
and, becoming more and more attenuated, extended to the stigma 
or summit of the nucleus ; in figs. 2 and 3 this cavity is marked 
i, and the embryo-sacs are seen in it in the state in which they 
were in the last ten days of April. In several flowers I already 
found the first trace of the embryo-sacs in the beginning of 
the same month; they originated in the centre of the base of 
the nucleus, and then increased in length from below upwards ; 
they grew therefore, as in the Leguminose, Santalinee, &c., to 
meet the subsequently penetrating fertilizing substance. Most 
frequently two embryo-sacs occur in the cavity of one and the 
same nucleus, as, for instance, in fig. 3; some mistletoe plants 
always presented in one out of every six or seven flowers two 
embryo-sacs; but three embryo-sacs were exceedingly rare in 
all the plants that passed through my hands ; for I noticed them 
only in two cases. On this plurality of the embryo-sacs, then, 
is the polyembryony founded in Viscum, by which it differs so 
remarkably from all other hitherto known cases. 
On itsfirst appearance the embryo-sac appears like a somewhat 
cylindrical utricle, which soon expands at the micropyle end, 
but retains at the opposite end its primitive size, even after 
several months (fig. 4. Pl. I. at a). The membrane of the em- 
bryo-sac of Viscum is remarkably thick and firm, such as I have 
scarcely ever observed in any other plant; nay, in a some- 
what advanced state, one might be induced to believe that a 
second layer of membrane had been precipitated on the inner 
surface. In Plate I., figs. 4 and 5 represent the simple embryo- 
sacs observed in the first week of May; they are transparent 
sacs, mostly without any solid contents, and occur from ten to 
fourteen days and more previous to the shedding of the pollen by 
the male flowers, whence, therefore, it is clear that the number 
of the embryo-sacs, and consequently the number of embryos, 
is not dependent on the number of pollen-tubes which have 
to descend into the nucleus. The process of impregnation, and, 
what is especially remarkable, the further development of the 
embryo proceeds with extreme slowness in Viscum; it is even 
three or four weeks and more before the first traces of impreg- 
nation become evident in the embryo-sac. 
The entrance of the pollen-tubes through the summit of 
the nucleus, and their connexion with the micropyle-end of 
