18 Å DR. J. D. HOOKER ON THE STRUCTURE AND 
Jere, and some others; and from these, through the single-coated ovules of Menispermeæ, — 
many Monocotyledones, and other Orders, to the double-coated ovule of most Phænogams, « 
and lastly to the three ovular integuments of Gnetum. | 
This rudimentary ovule can by no means be compared to the archegonium of a moss; | 
nor does the reduction of the ovule to its simplest form argue any alliance between Bala- « 
nophoree and €ryptogamic plants. The affinity which Griffith endeavoured to establish, « 
is in this respect founded upon erroneous views of the origin and development of the ovule 1 
of Balanophora, with which he was not acquainted; and the development of this reduces 1 
the grounds of the argument to a casual external resemblance, or rude analogy, between - 
two organs which are not homologous, and have no similarity of origin, structure, or — 
function. ; 1 
 lvery much regret my having been unable to trace the development of the ovule in | 
any of the three embryonate and albuminous genera, Sarcophyte, Mystropetalon, and « 
Cynomorium; and can only suggest that in them the albumen is endospermic, or deve- - 
loped within the embryo-sac, and not in the substance of the nucleus. The position of — 
the radicle in Mystropetalon being close to the hilum of the seed, suggests the probability j 
of the ovule being anatropous, and hence somewhat more complex than its congeners; 1 
whilst the lateral position of the embryo of Cynomorium is consistent with an obliquely - 
pendulous ovule. Until, however, we become acquainted with the process of impregna- 1 
tion or the development of the ovule or the albumen, we have no materials for forming 
an opinion on the real nature either of the ovule or the seed. I have repeatedly dissected 
- half-grown ovules of Cynomorium preserved in spirits, but never found a trace of any coats 
to the ovule, which always appeared as a membranous sac, full of cells as in Balanophora, 
and amongst which cells one is free, and from it the embryo is developed. 
M. Weddell (Ann. Sc. Nat. sér.-3. v. 14) has also considered the ovule to be an embryo- 
sac, but he makes this opinion depend upon views of the nature of the ovary, style, stigma, 
and perianth, so different from my own, as to render our accordance upon this individual 
point purely accidental. I shall return to this subject after describing the seed. 
Seed.—There are two types of seed in this order; the embryonate, and what has been 
called by various authors the exembryonate, and which has been described as consisting: 
of a homogeneous or sporuliferous mass. The only known embryonate genera are Cyno- 
moriwm, Sarcophyte, and Mystropetalon. 
The seed is always pendulous from the summit of the cavity of the pericarp. The 
excessively thin testa contracts an intimate but not organie adhesion with the walls of the 
generally erustaceous endocarp, and is always so closely applied to the surface of the seed 
that it cannot be detached. "This structure is very frequent in various Orders of Exogens, 
as in Gunnera, whilst there is a manifest tendency to it in Araliaceæ, Boldoa (a South 
American genus of Monimiaceæ), and other plants. In the exalbuminous species the 
substance of the seed is uniformly cellular; the cells, which are loose in the ovule, and — 
fill the cavity of the pericarp previous to the swelling of the seed, become densely packed, 
probably from the cavity being limited in size, and its walls indurated before the seed has 
con E fall eaim When ripe, the seeds of most are densely corneous, especially 
periphery. The comparison of the seed to a loose cellular mass, so frequently 
