Geophilons Species of Peperomia. 423 
cotyledon \ the single active cotyledon functions first as an absorbent and 
later as an assimilating organ, and it is only under exceptional cases that 
the rudiment of the second cotyledon can be induced to develop 1 2 . 
Cyclamen thus appears to have worked out its own special line of 
evolution to a geophilous condition, which may perhaps be considered 
as a failure, in so far as such a method does not appear to be capable of very 
great possibilities 3 or to have been followed to any extent in other cases. 
Another attempt at geophily with its correlated modification of the 
seedling structure appears to have been made by the Ranunculaceae, where 
a fusion of the cotyledonary petioles, and perhaps even of their laminae 
as well, has occurred in some cases 4 , but it does not seem clear that this 
kind of method has been followed in the past or has led to anything more 
than modification within this order. 
Miss Sargant, of course, contends that the anomalous Ranunculaceae, 
such as Eranthis with its fused cotyledons, afford the clue to the origin 
of the monocotyledonous habit of these Monocotyledons. It is certainly 
true that the internal structure of the cotyledonary tube of Eranthis 5 shows 
considerable resemblance to that of the * single cotyledon ’ of a form like 
Annemqrhena 6 which is considered by Miss Sargant to be a primitive type, 
but it seems an equally possible suggestion that the single median bundle 
of the monocotyledonous * single cotyledon ’ has become divided into two 
bundles, in correlation in some way with the parallel venation so common to 
the leaves of Liliaceae, &c. 
1 Gressner, Bot. Zeit., 1874, p. 837. Darwin, Movements of Plants, p. 96 ; cf. Schmid, Bot. 
Zeit., 1902, p. 217. Gressner held the view that the embryo of Cyclamen was dicotyledonous with 
one cotyledon rudimentary, and this view is confirmed by Schmid, Taf. ix, Fig. 47, and also by my 
own work on various species of the genus. 
2 It is interesting to notice that the anatomical character of the petiole of the single cotyledon is 
suggestive of a double structure. 
3 Amongst other pseudo-monocotyledonous Dicotyledons, which may perhaps show somewhat 
similar modifications to Cyclamen , may be mentioned Claytonia Virginica, L. ; cf. Holm, Mem. 
Amer. Acad. Sc., vol. x ; Mem. II, p. 30, PI. II, Figs. 10, 11, 13, 14. The ‘minute leaf’ figured 
by Holm, apparently opposite to the cotyledon, may represent the undeveloped second cotyledon. 
Bunium bulbocastanum , L., according to Hegelmaier, Vergl. UnterSuch., 1878, pp. 1 39-140, 
PI. VII, Figs. 40, 41, cf. Schmid, p. 216, has a dicotyledonous embryo, and may represent a 
more reduced condition than Cyclamen. The condition of affairs in Erigenia bulbosa is uncertain, 
vide Holm, Amer. Journ. Sc., 1901, p. 63; seedlings are at present under observation. In Cordy- 
dalis cava , Schmid, 1 . c., p. 213, the second cotyledon appears to be entirely aborted. 
4 Ranunculus ficaria seems to afford a very good case of fused cotyledons, but a transverse 
section of the petiole of the ‘cotyledon’ shows a single entire median bundle. R. bulbosus also 
shows a slight cotyledonary tube, and Dodecatheon (Primulaceae), &c. , also appear to be working on 
similar lines. In this connexion the frequent tendency to splitting among cotyledons may be 
referred to. This peculiarity is well seen in seedlings of Acer pseudoplatanus, which often appear 
to have three or four cotyledons ; and the same thing is seen in Primula sinensis , where the apical 
depression of the cotyledon becomes a deep cut, &c. In Cyclamen persicum (cult.) the lamina 
of the single cotyledon is occasionally divided into two distinct laminae. 
5 Sargant, 1 . c., p. 52 et seq., Pis. VI, VII, Figs. 1-3. 
G Sargant, Ann. Bot., xiv, p. 633, PI. XXXIII, Figs. 2, 3 ; New Phytologist, i, 1902, p. 107. 
