232 
Review. 
The polystelic stem Steloxylon has long puzzled palaeobotanists; 
the latest view is that of P. Bertrand, who thinks that the plant is 
a form of Cladoxylon. It is a pity that the transverse section of the 
stem is not figured. 
Chap. XXXI is devoted to the seed-bearing fronds of 
Pteridosperms such as Pecopteris Pluckeneti and Aneimites (here 
called Wardin') fertilis, where only external characters are known. 
The author attributes the winged appearance of the seeds in 
P. Pluckeneti to the presence of a sarcotesta, and suggests an 
affinity with Medulloseae rather than Lyginopterideae; the same 
applies to Wardia fertilis. The seed Samaropsis acuta is believed 
to belong to an Eremopteris, as first suggested by Duns as long ago 
as 1872, while Nathorst’s Lageuosperinum Arberi is associated with 
an Adiantites from Spitzbergen (Lower Carboniferous). 
Chap. XXXII is headed Cycadofilices ; the distinction from 
Pteridosperms is clearly arbitrary, though unavoidable while our 
evidence is so scanty as to the fructifications. It may be asked 
however why Steloxylon should be placed in Pteridospermeae and 
the almost identical Cladoxylon in Cycadofilices. 
It is remarkable how many of the families of Cycadofilices— 
Megaloxyleae, Rhetinangieae, Stenomyeleae and Protopityeae 1 —are 
monotypic. Rhetinangieae might perhaps be merged in Lyginop¬ 
terideae, but the others are certainly isolated types. Our knowledge 
of the whole group must evidently be very fragmentary; possibly 
we may be dealing with the survivors of families which had their 
day in an earlier period. 
In the account of the Cycadoxyleae the Cycadean features might 
perhaps have been more emphasized. These plants seem to be the 
nearest approach we have to connecting links between Pteridosperms 
and Cycadophyta. 
Zalessky’s separation of Eristophylon from Calaiuopitys is 
adopted, but the new genus is included in the Calamopityeae—a 
reasonable compromise between opposing views. 
Two chapters are devoted to the Cordaitales, the great class 
of Palaeozoic Gymnosperms. The family Poroxyleae is taken first, 
and the description is illustrated by photographs supplied by Prof. 
C. E. Bertrand. The habit of the plants is compared with that of 
Bamboos. In describing the Cordaiteae the name Cordaites is still 
used in a comprehensive sense, while special genera such as 
Mesoxylon are wisely restricted to anatomical specimens. The 
genus Araucarioxylon is rejected, and all wood of the general 
Araucarian type which cannot be more definitely determined is 
included under Dadoxylon. 
Stress is laid on the wide distribution of the Cordaiteae (includ¬ 
ing N ceggerathiopsis) which spread practically all over the world. 
As regards range in time, while Cordaitales no doubt existed in pre- 
Carboniferoits Floras, there is no evidence that the Cordaites type 
was then represented. At the other end of the scale, it is probable 
that the family Cordaiteae came down at least to the Rhaetic. 
The Cordaitean structure is described as xerophilous, and it is 
suggested that the frequent absence or slight development of 
palisade tissue in the leaf may point to diffused rather than brilliant 
sunlight. 
> p.212, line 9 ; the statement refers to the primary, not to “the inner 
part of the secondary xylem,” 
