14 Campbell.—Studies on the Araceae. 
in low wet ground, and closely resembles in general appear¬ 
ance Anthurium acaule . Indeed both in the leaves and flowers 
it seems to be nearer to Anthurium than to Symplocarpus , 
with which it is usually associated. It was late for the flowers 
at the time of my visit, and most of my material showed only 
the later stages of the embryo-sac and the embryos. For 
the earlier stages, I had to depend on the material furnished 
me by Professor Hill, of Seattle. 
The flowers, like those of Anthurium , are hermaphrodite, 
and like many species of Anthurium , the ovary is two-celled, 
with two ovules in each cell. The ovules themselves, how¬ 
ever, are much larger and more massive, and are orthotropous. 
They are attached to the inner wall of the ovarian cavity, near 
the base, but none of the material was young enough to deter¬ 
mine positively whether they arose from the axis, as in Dieffen- 
bachia^ or from the carpel itself. To judge from a comparison 
with the older carpels of Dieffenbachia> it is probable that in 
Lysichiton also, the ovules are outgrowths of an axial placenta. 
The young ovule (Fig. 34) has a very broad and short funi¬ 
culus. As in Dieffenbachia and Aglaonema, the nucellus is very 
small. As in these forms, the micropyle is formed by the 
inner integument alone, the outer integument being very short. 
The latter is, however, extraordinarily massive, so that the 
breadth of the ovule considerably exceeds its height, giving it 
a most characteristic form. At this time, an examination of 
the epidermal cells of the base of its outer integument shows 
that they have begun to enlarge, and their contents become 
dense and the nuclei large. Later, these cells grow out into 
elongated papillae, and the base of the ovule is surrounded by 
a thick ring of these secretory hairs which develop a mass of 
transparent mucilage in which the ovule is completely em¬ 
bedded, and which make it very difficult to preserve the older 
ovules without shrinkage ; even taking all the precautions I 
could, much of my material proved of little use. 
The earliest stages procurable had the parts of the embryo- 
sac already complete, and the young embryo-sac showed the 
egg-apparatus of the ordinary form, and the three antipodal 
