2 10 
Notes on Recent Literature. 
and Ephedra, upon which both Wettstein and Warming lay stress, 
have no more taxonomic significance than the presence of root- 
tubers in Cycas, Myrica, A Inns, etc.; and the direct relationship 
of Angiosperms and Gymnosperms in general, and of Ephedra and 
Casuarina in particular, is a view which cannot be upheld. It may 
be that systematists will not be disposed to reduce the Verticillatse 
to tribal rank, equivalent with the Coryleae and Betuleae, as Benson 
at first suggested; but that her work vindicates the position 
accorded to Casuarina by Eichler admits, in our judgment, of little 
or no doubt. 
In fact, we have the most serious doubts as to whether any of 
the modern speculations which attempt to derive directly the 
Angiospermae from the Gymnospermae have any permanent value ; 
and hence we are not disposed to take very seriously some of the 
phylogenetic tables given by Lotsy. In one of the latter (see p. 442), 
a hypothetical group—Drimytomagnolieae—is directly derived from 
the Bennettiteae. This supposition, in our opinion, places too much 
weight on a hypothesis of the so-called flower of the Bennettiteae, 
and too much weight on the very isolated phenomenon of the simple 
anatomy, seen in only a single tribe of the Magnoliaceae. Whatever 
view be taken of the simple wood of the Drimydeae, its phylo¬ 
genetic value is greatly discounted by the fact that normal 
Angiospermous wood occurs in tne remaining tribes of the family. 
Generally, we believe that the Angiospermae and Gymnospermae 
are connected only by some ancient and unknown ancestors which 
gave rise on the one hand to the more primitive Gymnospermae, 
and on the other hand to some unknown group of primitive Angio¬ 
spermae. On this view, the union of the Gymnospermae and 
Angiospermae by Engler and others into a single division (Abteilung) 
—Embryophyta Siphonogama—is not very useful, as it tends to 
obscure the real relationships of the two sub-divisions. 
Warming’s treatment of the Centrospermae, a name which he 
now adopts in place of Curvembryae which he formerly used, differs 
essentially from that accorded to the order by Engler. Wettstein, 
who adopts with little modification the sequence of families given 
by Engler, begins with the simple and presumably primitive 
Chenopodiaceae and goes on to the more complex and presumably 
derived Aizoaceae and Caryophyllaceae. Warming (p. 200) states 
that, with regard to the families of the order, “ it is uncertain what 
genetic relationships they bear to one another, and in what sequence 
they with most justice should be placed. Some authors begin with 
the incomplete types, and then take the more highly differentiated 
ones which possess both calyx and corolla, split stamens, epigyny, etc. 
Here, for the sake of convenience (“af praktiske Grunde,”) the 
Caryophyllaceae are placed first. After them are taken the simpler, 
formally reduced types which are considered to be most directly 
derived from them. After these, follow others with a single perianth. 
Lastly, families with epigyny and numerous stamens formed by 
division are taken.” Danish botanists who have become accustomed 
to the arrangement of the Centrospermae in Warming’s former 
work will doubtless appreciate these “praktiske Grunde”; and 
doubtless the more conservative among them will even enter a more 
or less mild protest against the change of name from the (to them) 
familiar Curvembryae to the (to them) unfamiliar Centrospermae; 
but those of us who are constitutionally inclined to regard systematic 
