132 AROIDEJE. 
The lobes have a curious vernation on similar principles, 
one may be regarded as external, its lamina involving the 
other, though not completely; Pl. CLX XIII. Fig. II. а 
inner, б, b, b, outer. 
The inner is involved on itself, and is much shorter, ceasing 
opposite the shorter one of the other lobe; this shorter one 
is its inner, the outer being much larger, becoming involuted ; 
within this, and passing round first between it and the outer 
limb of the outer lobe; subsequently the peduncle has a si- 
milar leafy sheath. 
In Caladium fornicatum it is really leafy at last, although as 
usual it is membranous at first. i 
Have Aroideæ unilocular or bilocular anthers? On this 
point, much will depend on the composition of the mul- 
tilocular anthers, as if the cells are always in pairs, they 
are bilocular; but in which ever light we look upon them, all 
those which have more than 4 cells to an anther, are evidently 
not monandrous in the strict sense of the word, for although 
we have evidence of anthers being many celled, yet in all 
these the formation is evidently bilocular, this is indicated by 
the formation of the anther in Viscum, and by the two pores 
in Rafflesia. Besides, in both these, the cells are irregular, 
and indefinite in number. See also Rhizophora. 
The types of formation dependent on the situation of the 
anthers. Are typical, really hypogynous, or are they dislocated 
and placed higher up than the female flowers, are there glands 
or not intervening, of which it may be said, that the lower re- 
present abortive ovaria, the upper abortive stamina, at least in 
Arum flagelliforum. 
Calla Aromatica, in which the fertile stamens are dislocated, 
is intermediate between these two distinct forms, but there 
is an abortive stamen attached to the base of each ovary. I 
consider this an abortive stamen, rather than a slightly deve- 
loped bract, from its appearance, its want of vascularity and 
