168 
IOWA ACADEMY OF SCIENCE 
3. Reduction of parts: Rudimentary stamens, or staminodia, are 
present in most of the Philodendroidcae ( Philodendron , Peltandra, Homa- 
lomena , Typhonodo ru m, and occasionally perhaps Pichardia) , and in a 
few of the Aroideae ( Gorgonidium, Spathantheum, Spathicarpa) , also in 
the Staurostigmoideae, and in Dieff enbachia of the Aglaonemoideae. In 
the stamiiniafte blossoms, rudimentary carpels are occasionally present 
(. Peltandra , Apatemone) . Engler notes that all forms having both a 
perianth and petaloid spathe have very inconspicuous sepals, and in 
many cases the latter must be regarded as rudimentary. Aroids having 
a bract-like spathe have comparatively conspicuous perianths. The peri- 
anth of Syrnplo carpus is plainly protective in its function. Possibly that 
of Acorus is also protective ; if not, it must be regarded as rudimentary, 
as it evidently is not in its present condition adapted to entomophily. 
Authurium possesses a rudimentary perianth, as does also Spathiphyllum. 
The other genera studied lack the perianth. 
The frequent presence of staminodia indicate that the perfect flower 
is, in the case of the Aroids, more primitive than the imperfect, 'and 
that the monoecious and dioecious species have suffered reduction of the 
floral parts. This is further suggested by the occasional presence of 
perfect flowers in Pichardia. The rudimentary character of the perianth 
in miany genera indicates reduction along that line also, and suggests 
that the sepaloid forms are the more primitive. The crowding of the 
blossoms on the stalk would naturally lead to a gradual elimination of 
the floral envelopes, and as they tend to disappear in the course of 
evolution, their place is taken by a more or less brightly colored and 
petaloid spathe. Such a form as Arisaema cannot but be regarded as 
quite highly developed, since it is dioecious (sometimes slightly mono- 
ecious) and has naked blossoms crowded on the spadix, and a highly 
developed spathe. Symplocarpus is more primitive, having perfect, 
sepaloid flowers, though the spathe here also is well developed and of 
use in entomophily. Pichardia is one of the more highly specialized 
forms, but shows vestigial characteristics in the form of staminodia, 
and occasional perfect flowers. 
If the theory here advanced is correct, the Aroids must trace back to 
some ancestral form that was sepaloid and adapted to the habit of 
entomophily. Far back of that, very likely, lies a group of yet more 
primitive anemophilous plants, but their existence is purely inferential, 
and rests upon no positive evidence to be found in the Aroids as we 
know them today. 
