240 
PROFESSOR W. C. WILLIAMSON ON THE ORGANIZATION 
space surrounding this parenchyma are numerous small rounded bodies (l), apparently 
identical with those seen in fig. 70. These are the supposed pollen grains. 
Fig. 73 is another section of the same seed, made parallel to fig. 72, only nearer the 
base of the lagenostome. The tissues reappear as in the last section, except that the upper 
half of the wall of the lagenostome, c, exhibits the crenulated figure of that membrane in 
a more regular manner than before, and at g we now see the uppermost surface of a 
portion of the perispermic membrane. The appearances presented here are extremely 
significant. This membrane is altogether different from those already described ; a small 
portion of it, enlarged 175 times, is seen in fig. 75. It is apparently structureless, except 
that it is full of minute translucent spaces usually exhibiting sharply defined angular 
contours, and which look as if the membrane had been filled with minute crystals*, which 
I believe to have been the case, unless these objects were crystalloids, which is not 
improbable. It is perfectly obvious from this section that the prosenchymatous wall of 
the lagenostome did not extend across the base of the latter structure, but that, in its 
normal state, the cavity of that lagenostome was in direct contact with the membrane 
of the perisperm, and that consequently nothing intervened between the pollen grains 
in the lagenostome and the perispermic cells but this thin membrane. Thus it follows 
that in this seed the lagenostome (the “ cavite pollinique ” of M. Brongniart) has. been 
formed by a separation of the nucular membrane from the perispermic one at the upper 
extremity of the perisperm. 
I have entered upon the minute details of the preceding description because I believe 
this remarkable seed to be essentially a typical one. It appears to he much more 
elaborate in its organization than most of those described by M. Brongniart, yet it 
obviously belongs to the same class of seeds as those discovered by M. Grand’Eury. 
That no known recent seed corresponds exactly with these has already been pointed 
out by M. Brongniart. But that distinguished botanist read a brief memoir to the 
Academy of Sciences in Paris in September, 1875 f, in which he indicated the proba- 
bility, from observations which he had made upon the fertilized ovules of some recent 
Cycads, that in their young states the seeds of this class of living plants exhibit conditions 
not dissimilar from those found in the Carboniferous fossils ; but the notice published in 
the ‘ Comptes Rendus ’ is too brief to contain the details needful for a thorough settlement 
of the exact degree of identity. On the other hand it appears to me very important 
that the early development of the fruit of the Salisburia should be studied in reference 
to this question. Dr. Hooker long ago pointed out the many features in which this 
Chinese fruit resembled the fossil Trigonocarpons ; but there is one peculiarity in it which 
he has not noticed, but which appears to bear upon the present inquiry. On opening 
the hard endotesta (fig. 76, a) of one of the dried seeds we discover immediately within 
* This appearance is almost identical with that presented by the surfaces of the spicular sclerogen cells 
observed by Dr. Hooker in the parenchyma of the stock of Welwitschia mirabilis (Trans. Linnean Society, 
vol. xxiv. pi. xii. figs. 5, 6). 
t Comptes Rendus, tome Ixxxi. no. 47. 
