THE 
NEW PHYTOhOGIST. 
Vol. i., No. 7. July 24TH, 1902. 
ON SOME POINTS OF APPARENT 
RESEMBLANCE IN CERTAIN FOSSIL AND RECENT 
GYMNOSPERMOUS SEEDS. 
By F. YV. Oliver. 
O F the considerable number of fossil seeds with whose organi¬ 
sation we have become acquainted through the labours of French 
and of English palaeo-botanists, none shew a higher degree of 
complexity than Pachy testa from the French Permo-carboniferous, 
and Lagenostoma from the Lower Coal Measures of Lancashire. 
In the present note it is proposed to direct attention to some of the 
complex features shewn by the two seeds just named whilst some 
reference will be made to Torreya, an existing genus of Taxaceae. 
Of known palaeozoic seeds many seem to be organised on what 
may be quite generally termed the Cycadean type. Such seeds 
are straight, the upper portion of the nucellus is transformed 
into a pollen chamber of varying magnitude, whilst the integument 
(often bearing an outer sarcotesta) is adherent to the nucellus 
throughout the prothallial region. In other words the pollen- 
chamber and the micropylar region of the integument alone are 
free from one another. Treub, 1 who drew a comparison between 
this type of ovule and the sporangium of Ophioglossmn, has expressed 
the view that these free portions are new developments. 
In other palaeozoic seeds, of which Stephanospennum akenioides , 
Brongn. may be cited as an example, nucellus and integument 
appear to have been free, from the chalaza upwards. This condition 
may be regarded as more primitive, and the adnate Cycadean type 
as derived from it by the intercalation of an extensive basal region 
(phylogenetically younger) in which the prothallium and embryo find 
accommodation. Whatever view may be taken of the Cycadean 
1 Ann. du jard. hot. de Buitenzorg, vol. II., p. 48. “ be Nucelle 
et le tegument seraient des creations nouvelles dont on lie 
trouve pas d'honiologues dans les cryptogames,” 
