On Certain Archaic Features in the Seed of Taxus 
baccata, with Remarks on the Antiquity of the 
Taxineae. 
BY 
B. SAHNI, M.A., M.Sc. 
Emmanuel College , Cambridge ; Professor of Botany , Benares. 
With seven Figures in the Text. 
Introductory. 
T HE examination of Taxus seeds on which this paper is based was 
originally undertaken in connexion with an investigation of the New 
Caledonian genus Acmopyle , an account of which it is hoped to publish in 
the near future. In that paper the conclusion is expressed that the genera 
Taxus , Torreya , and Cephalotaxus are structurally so distinct from the 
Podocarps and other Conifers, that they deserve the rank of a separate 
phylum, Jaxales, 1 having a clearer and more direct relation with Ginkgo 
and the Cordaitales than may be claimed for the remaining plants generally 
grouped under the Conifers. Since the following observations — which have 
a strong bearing upon the Cordaitean theory of the ancestry of Conifers — 
would have unduly lengthened my paper on Acmopyle , they are here 
published separately. 
Professor Oliver’s work (1903) has clearly shown that Torreya possesses 
one of the most primitively organized seeds known among living Gymno- 
sperms. From his study of the seed of Cephalotaxus Worsdell (1900) has 
come to a similar conclusion with regard to this genus. These considera- 
tions, and the close similarity in general morphology between the ovuli- 
ferous shoots of Cordaites and Taxus, impelled me to make a search for 
possible archaic features in the seed of Taxus, and to compare the latter 
with what we know of the structure of Cordaitean seeds. 2 
As a result of this comparison I have been led to conclude that, so 
far as seed-structure is concerned, ( 1 ) Taxus is in some respects more 
1 A name previously adopted by Lotsy (1911, p. 160 ), but not in the sense of a group equal in 
rank with the Coniferales. 
2 The previous investigations on the seed of this common plant, numerous as they are, have 
chiefly been conducted from the point of view of the structure and development of the nucellus, 
gametophytes, and embryo, while the minute study of the integument has been confined, so far, as 
I know, to the comparatively young stages. 
(Annals of Botany, Vol. XXXIV. No. CXXXIII. January, 1920.] 
