THE PALJEOZOIC SEED LAGENOSTOMA OVOIDES, 419 
In В. 29 (PI. 93. fig. 11) and R. 44, just within the megaspore membrane, 
small groups of carbonaceous material are seen, and in R. 24 (РІ. 23. fig. 18) 
there is much the same appearance, only that within this layer, which is 
here multiseriate, is another of similar masses pale brown in colour, while 
the outer are black. The inner die out towards the centre. The megaspore 
of R. 42 is entirely filled with these cell-like masses, and in places, par- 
ticularly at the base, thin streaks appear between them. It is difficult to 
resist the inference that we have in these preparations three stages in the 
formation of the prothallus: the lining layer of carbonaceous groups may 
represent the first stage, where, as in recent Cycads (6, p. 132), а parietal 
cytoplasmie layer with free nuclei is first formed; we then have this 
spreading inwards in В. 24, where the lighter colour may indicate 
the younger nuclei; and finally, in R. 42 the whole megaspore is full 
of nuclei, the fine streaks suggesting the beginning of partition. walls. 
Indications of a cellular tissue within the megaspore membrane are also 
shown in the section 1429 of the Williamson Collection, and Bertrand (4) 
has already noted the occurrence of “ membranes cellulaires ? in the endo- 
sperm of some silicified seeds. 
It is probable that the archegonia of the Lagenostomex were situated at the 
base of the pollen-chamber (cf. Oliver (12), fig. 9, pl. 24), and in В. 39 
a dark brown mass occurs, which is certainly suggestive of an arehegonium. 
It is true that dark masses of similar size occur promiscuously in other parts 
of the seed (Pl. 24. figs. 32 & 33), but these have more the appearance of 
a fungal origin. 
К. Fungus. 
No description of the material available would be complete without an 
allusion to the fungus which several preparations show infesting the decaying 
seed, though for a full elucidation we must no doubt await the palæo- 
mycologist of the future. 
(1.) Mycelium.—This is shown to its greatest extent in К. 29 (РІ. 83 
figs. 11 & 13), where the hyphæ ramify in all the soft parts of the seed, 
i. e. protballus, nucellar cone (though not in the dark apical portion), and 
canopy chambers. ‘They are extremely fine, perhaps 2 ш across, and are so 
closely interwoven that it 13 almost impossible to determine such points as to 
whether or not they were septate, branched, ete. Very probably they were 
both, for in some preparations where hyphæ occur singly these points can 
be satisfactorily demonstrated, R. 40 (1) (РІ. 24. fig. 25) showing a septate 
and R. 49 (Pl. 24. fig. 23) a branched hypha. But of course it is impossible 
to be certain in these cases that one is dealing with the same fungus. The 
hyphe in the two latter cases are stouter, measuring 4 ш across. 
2L2 
