88 
F. Cavers. 
leaves. A similar arrangement occurs in the genus Petalophyllum, 
and this, together with the tuber formation which occurs in both 
genera, led Goebel (13) to the view that Geothallus is closely allied 
to Petalophylliuu ; hut we meet with the same thing again in a 
species of Anthoceros, and it is obviously nothing more than a case 
of homoplasy—which, as has been pointed out by Farmer and by 
Goebel, the Hepaticas show to an extent that cannot be paralleled 
by any other group of plants. 
Riellace^e. 
Until 1902 the remarkable aquatic genus Riella was supposed to 
be confined to the Mediterranean basin, the seven species then known 
having been recorded from Algeria, France, Sardinia, and the Lake 
of Geneva ( R . Reuteri, described by Hofmeister, but long ago 
destroyed owing to the erection of a mill on the lake-side). New 
species have since been described from Turkestan by Porsild (26), 
from Texas and the Grand Canary by Howe and Underwood (19), 
and from Port Elizabeth, South Africa, by the writer (7)— R. capensis 
is of interest as extending the range of the genus to the Southern 
Hemisphere. A paper on R. capensis, and on the genus generally, 
is being prepared for publication. 
Riella closely resembles Sphcevocarpus and Geothallus in the 
structure of the sexual organs and sporophyte, but differs greatly 
from these genera in having a cylindrical axis or stem which is 
produced dorsally into a wing-like expansion, besides hearing a row 
of leaves on either side at the base of the wing. In R. bialata, 
recently described by Trabut (37), there are two wings, right and 
left, both arising from the dorsal side of the stem. 
The morphology of Riella has been much discussed. Several 
writers have doubted the accuracy of Montagne’s original description 
and figure (1, 2) of R. helicophylla, the first species to be discovered, 
in which the wing was represented as being wound in a close spiral, 
like the thread of a screw, around the erect stem. Leitgeb (22) 
confirmed this description, found that the cells of the axis showed 
spiral twisting, and suggested that the wing arose at a late stage, 
after the stem had coiled around some supporting water-plant; this 
suggestion of a climbing habit was quite erroneous, but Leitgeb had 
only herbarium material. Later writers supposed that the spiral 
appearance of the wing was simply an optical illusion, and that the 
wing was merely folded and wavy, as is certainly the case in 
R. capensis (Fig. 3). Quite recently, however, R. helicophylla has 
