ioo Oliver. — On Physostoma elegans , Williamson , 
which, though it often persisted to a fairly late stage in the history of the 
seed, probably culminated functionally at a somewhat younger stage than 
that which most of our specimens have reached (as in R. 99, see PL VII, 
Fig. 29). Its persistence may have been facilitated by suberization of the 
membranes, as Thomson has shown in the case of recent Cycads and other 
Gymnosperms. 1 In Lagenostoma , on the other hand, a tapetum has not 
been detected in the ordinary pollinated specimens ; so that, if one were 
present at all, its functional culmination must have occurred at a much 
earlier stage of seed-development. 
10. The Mcgaspore- Membrane and the Prothallus. 
Throughout the investigation of Physostoma careful search has been 
made for traces of the megaspore-membrane. In the majority of specimens 
this structure is conspicuous by its absence, and it is only in the rarest 
examples that a very delicate membrane has been found delimiting the 
prothallus (as in Professor Bertrand’s specimen, M. H., 370, PI. V, Fig. 2, mg). 
Though the evidence is mainly negative, the conclusion is difficult to resist 
that no robust membrane such as occurs in Lagenostoma (both species) was 
yet present in these seeds when they dropped and were petrified. In this 
connexion it is of interest to note that in Conostoma oblongum also — which 
like Physostoma has a many-layered tapetum — a megaspore-membrane has 
eluded observation. Under these circumstances it would appear probable 
that the megaspore-membrane in these seeds did not thicken so long as the 
tapetum was in a state of functional activity. 
As regards the prothallus, there is nothing to add to the statement 
that it was a very delicate tissue in the few cases that have come under 
observation (as in M. H. 370, PI. V, Fig. 1, psi). Archegonia have not 
been seen. 
11. Undersized or abortive seeds. 
Two sections of what appear to be specimens of small seeds or ovules 
have come under observation during the course of the present investigation. 
As these sections belong to the middle regions of the seed, and their 
preservation is indifferent, no full account is possible as in the corresponding 
specimens of Lagenostoma Lomaxi . 2 There are certain features, however, 
worth placing on record. The better specimen of the two is contained in 
the preparation U. C. L., R. 99 (see PI. VII, Fig. 29). This section seems to 
be cut a little below the height at which the tentacles were given off. 
Eleven ribs are present, and they are crowned by tubular hairs (the latter 
are not shown in the figure). The mean diameter of the section, excluding 
1 R. B. Thomson, The Megaspore-membrane of the Gymnosperms, Univ. of Toronto Studies, 
Biol. Ser., No. 4, 1905. 
2 Oliver and Scott, loc. cit., p. 21 1. 
