A NEW FAMILY OF FLOWERING PLANTS. 49 
А New Natural Family of Flowering  Plants—Tristichaceæ. Ву 
J. C. Wiis, M.A., Sc.D., Director of the Botanic Gardens, 
Rio de Janeiro. 
[Read 7th May, 1914.] 
Havine during the last few months been occupied with the Podostemaceæ 
of Brazil, I have begun to find that these plants are in reality as little known 
as those of India and Ceylon, which I have described in detail in previous 
monographs *. Already four absolutely new species have been discovered, 
some of them with morphological features of great interest, some of. them 
combining existing genera, some apparently representing new genera. The 
results of this work will provide the material for several papers, of which 
this is the first. 
The existing family Podostemaceæ contains many plants which are not 
in reality very closely related. The most aberrant of these— Hydrostachys, 
has already, and very properly, been removed by Warming, and placed in a 
special family, Hydrostachydaceæ ; but the remainder of the family still 
exhibits two great divisions, which are perhaps more widely separated one 
from the other than the divisions in any other family of the flowering plants. 
These divisions are the Chlamydatæ and the Achlamydatæ, whose characters 
may be briefly summed up as follows :— 
Chlamydate. Flower regular, or slightly irregular in the andrœcium. 
Perianth 3 or 5-merous, free or united, sepaloid, marcescent. Stamens 3 or 5, 
or 20-25, or 2, or 1. Ovary, 3-2 loc. Leaves small, simple, entire or 
nearly so, exstipulate. 
Achlamydate. Flower regular, or slightly or highly zygomorphie, naked, 
enclosed in a spathe springing from the base of the stalk. Stamens æ —1, 
sometimes monadelphous, usually with as many staminodes. Ovary, 2 loc. 
Leaves often large and much branched, usually stipulate. 
This division has long appeared to me unsatisfaetory. There are families 
with great gaps between their sub-orders, but none, or very few, in which 
it is not possible to conceive of fairly easy steps by which the transition from 
one to the other might be effected, and within a family this should always, 
it seems to me, be reasonably possible. 
Now this is exacily what is impossible with the two groups into which 
the existing family of Podostemaceæ is divided. How can one pass easily 
* Willis: “A Revision of the Podostemaceæ ot India and Ceylon,” Ann. R. B. G. 
Peradeniya, vol. i. 1902, p. 181. 
“Studies in the Morphology and Ecology of the Podostemaceæ of Ceylon and India." 
L. с, p. 257. 
LINN. JOURN.—BOTANY, VOL. XLII. E 
