IOWA ACADEMY OF SCIENCE 
167 
reduction. It cannot be regarded as a protective organ, nor could it 
ever, in its present condition, have been of use in entomophily. We are 
forced to the conclusion either that it is rudimentary, or that it is a 
nascent organ awaiting some change in the economy of thie plant to make 
it of use ; and the lattier of course is ridiculous. A form like Typha, with 
rudimentary perianth and spathe is reduced rather than primitive. If 
it is to he classed with the Palms it should follow rather than precede. 
The formation of a spathe and spadix links the Arales with the 
Palmales. The latter show true spiral characteristics in the individual 
blossom, the arrangement, of the stamens being spiral while that of the 
carpels is cyclic. The well developed perianth has led to its being re- 
garded as one of the higher orders, but in view of the floral spirality it 
would seem that it ought to be regarded as a more primitive group. 
This merely bears out the writer’s view that the Pandanales are a re- 
duced type with rudimentary perianth and in the Palmales we find a 
type which has not suffered reduction. 
While the Graminales are a comparatively primitive group, they stand 
farther from the Arales than do any of the other distinctly primitive 
groups, and the connection between the two must be very remote. Re- 
duction of parts has proceeded farther among the Graminales, yet they 
have retained the primitive anemophilous habit. The Farinales come 
somewhat closer to the Arales than do the Graminales. The well de- 
veloped perianth would indicate this. But the Farinales , Liliales , Or- 
chidales, and Scitaminales all represent a comparatively advanced stage 
of development, as indicated by the cyclic flowers, land the absence of 
spiral grouping of blossoms on the stalk. All consist for the most part 
of terrestrial, entomophilous plants which, in some cases, show decided 
tendencies in the direction of syncarpy, perigyny, and sympetaly. Were 
the writer to suggest a grouping which would put the Aroid order in 
what is probably its natural position, it would be as follows: 
Primitive Forms 
I 
Arales 
Synanthales 
Pandanaies 
Liliales 
Scitaminaler 
Orchidales 
