462 
Palaeontologie. 
nucleate type, and hence must be regarded as representing a derivea 
and not a primitive condition. A normal embryo-sac shows five 
successive divisions beginning with the Spore mother cell, while 
in such cases as Peperomia all four megaspore nuclei probably aid 
in forming the embryo-sac, and each of them divides onlj" four 
times. Other exceptional cases are explained in the same wa}^. 
M. A. Chrysler. 
Oliver, F. W., On Physostoma elegans, Williamson, an archaic 
type of Seed from the palaeozoic Rocks. (Annals of Botany, 
Vol. XXIII, p. 73 — 116. 3 pl. and 10 text-figs. 1909.) 
The seed described in this paper is the Lagenostoma physoides 
of Williamson’s 8 th memoir (1877) of which an abundance of material 
has recently become available. It belongs to the Lagenostoma-gr oup 
of seeds, from wich it differs however in important particulars so 
that generic distinction is required. The present name was Willi¬ 
amson’s original Suggestion, but owing to the incompleteness of 
his material he placed is provisionally in Lagenostoma. 
Physostoma is a small straight seed about 6 mm. long. The ribbed 
integument is coalescent with the body or nucellus, except at the 
apex where the ribs separate to form a whorl of tentacles surroun- 
ding the pollen-chamber. Ribs and tentacles alike are adorned with 
long tubulär hairs which give all sections of the seed a very cha- 
racteristic appearance. The multiple character of the integument is 
regarded as an archaic feature of which traces are discernible in 
Lagenostoma and allied seeds. The various regions are described 
in detail. The megaspore, which is surrounded by a well-developea 
tapetum, is peculiar from possessing an apical papilla which protru- 
des into the floor of the large pollen-chamber. This papilla is com- 
pared with the u tent-pole” of Ginkgo and Cordaitean seeds. The 
pollen-chamber is unusually rieh in pollen-grains with which are 
occasionally associated bodies which are interpreted as spermatozoids. 
In the discussion at the end the significance of the multiple 
integument is considered with especial regard to its being a new 
formation. The seed is compared in some detail with the allied forms 
in the concluding section. Though evidence of the parentage of 
Physostoma has not been obtained, it is regarded as probable that 
it belonged to a member of the Lyginodendreae. The paper is 
illustrated by photographic reproductions of sections, drawings and 
diagrams. Among the last is included a frequency curve shewing 
that the numerical Variation of the ribs was essentially the same as 
in similar cases at the present day. The seed is found at numerous 
localities in Lancashire; its horizon is the Lower Coal measures. 
F. W. Oliver. 
Reiss, K., Untersuchung über fossile Hölzer aus Japan. 
(Inauguraldissert. Leipzig. Rostock. 224 pp. 1 Tafel mit 10 Fig. 
1909.) 
Verf. gibt zunächst zur Erläuterung der Lagerungsverhältnisse 
der Holzreste, die ihm Felix zur Bearbeitung übergab, eine aus¬ 
führlichere compilatorische geologische Einleitung (p. 1—88), wo er 
auch die fossilen Floren Japans besonders nach Yokoyama und 
Nathorst anführt. 
Die Holzreste sind meist Coniferenhölzer, von denen Verf. eine 
ganze Reihe von „Arten” aufführt. Es sind vorhanden 2 Cedroxyla 
