1G2 
CONTRIBUTIONS TO BOTANY, 
intimately agglutinated to it. Within this, again, is found an- 
other thin, hyaline, reticulated integument, which is quite free 
from the former, but adherent to the albumen. The albumen 
is deshy, of an oblong form in Strigilia, with a small nipple- 
shaped protuberance at its base, somewhat excentrically placed, 
in which is imbedded the extremity of the slender, terete, infe- 
rior radicle, while the oval, compressed, foliaceous cotyledons 
are situated in the middle of the albuminous mass, and are about 
|ths of its length, and somewhat narrower. This structure, 
with very little variation, I have found constant in Strigilia, 
Styrax, and Cyrta. The pericarp in Styrax and Cyrta is very 
analogous to that above described in Strigilia ; in the two former 
the endocarp is more intimately combined with the mesocarp, so 
that in Cyrta the entire pericarp opens by three equal valves, 
although only unilocular and monospermous ; in Styrax, these 
valves open only at the apex, the pericarp thus becoming cupu- 
liform in shape, and closely investing the seed. On the side 
opposite to that of the origin of the raphe, a little beyond the 
l3> 2. large hilum, is seen a small scar, which closes a foramen beneath 
it open to the interior ; and in this cavity the prominent nipple, 
containing the extremity of the radicle, enters. In Styrax offi- 
cinale, where the albumen is in the form of a depressed globe, 
the embryo lies nearly in a horizontal position, with the coty- 
ledons slightly inclined to the hilum ; in Cyrta, where the albu- 
men is oval, the embryo lies in a more diagonal position ; but 
in Strigilia, where the albumen is oblong, the embi-yo is vertical : 
in all three cases, the radicle points to the cicatrix seen a little 
on one side of the hilum. The structure of the fruit and seed 
in Halesia differs from the foregoing in many essential respects, 
as I will presently show ; but in every case throughout the Sty- 
racece it is totally unlike that existing in the Symplocacece. 
It is important to notice here, that the external shell of the 
seed above-mentioned is in no way analogous to the outer osseous 
tunics which I have described in the families of the Canellacea, 
IVinteracecE, and Lardizabalacece, and which I have shown to be 
arillous in their nature. Nor can it be compared to the bony 
shell of the Clusiaceee and Magnoliacea, where it constitutes a 
tunic lying within the fleshy coat that bears the raphe. Here 
the position and course of the raphe prove that in the Stijracinea 
the osseous shell of the seed is the proper testa, originating from 
the growth of the primine of the ovule : we see that the whole of 
its fleshy mesodermic tissue has become solidified* by the depo- 
* This offers a. strong confirmation of the view I have taken of the 
nature of. the bony shell in the see<l of Magnolia, which by a few eminent 
botanists has been thought to result from a deposition of sclerogen upon the 
inner face of the primine, leaving the outer face transformed into a thick 
