5 16 
Notes. 
sing a number of distinct vascular cylinders in the main stem. Both 
genera indeed include monostelic species, such as Auricula replans , 
and Gunnera monoica etc., but the majority are polystelic, the num- 
ber of cylinders varying in different species from two to eighty. In 
the monostelic species of each genera the single cylinder has precisely 
the same structure as each of the many cylinders of the polystelic 
forms. In every case each 4 stele ' is almost destitute of pith, has its 
vascular bundles crowded together, and shows little or no secondary 
growth in thickness. 
Now it is remarkable that both Auricula and Gunnera have near 
relations which are aquatic in habit, and which have the reduced 
vascular structure characteristic of aquatic plants. Among the Primu- 
laceae there is Hottonia , in which the submerged stem has a central 
cylinder of simple structure, with little pith, confused vascular bundles, 
and little or no secondary thickening, just as is found in Auricula 
replans , or in a single 4 stele’ of any of its allies. Van Tieghem, 
after pointing out this identity of structure, adds : 4 Mais FHottonie 
doit sans doute l’&roitesse de son cylindre central a sa vdg&ation sub- 
merge, de sorte que sa structure n’est pas rigoureusement comparable 
a celle des Auricules ’ 1 . 
Gunnera also, as a member of the Order Halorageae, has a number 
of aquatic relations, the central cylinder of which (at least in Myrio- 
phyllum and Hippuris) perfectly agrees with that of the monostelic 
Gunneras, or with the several cylinders of the polystelic species. In 
drawing this comparison, Van Tieghem repeats the remark just quoted 
as to Hottonia and Auricula. 
In the polystelic Auriculas the flower-stalks and the leaves have 
normal structure. In the Gunneras on the other hand polystely 
extends both to the pedicels and to the petioles and larger veins of 
the leaf. 
It is a striking fact that the two Dicotyledonous genera, in which 
alone, so far as we know, polystely prevails, belong to families remote 
from each other, but agreeing in the fact that they include aquatic 
representatives of reduced monostelic structure. 
The suggestion which I wish to bring forward is this : is it not 
possible that these polystelic Dicotyledons may owe their exceptional 
structure to descent from aquatic ancestors ? When a plant becomes 
adapted to submerged aquatic life, we find as a rule that its vascular 
1 Sur lu polystelie, 1 . c., p. 295. 
