“7 
eS 
re 
we 
wa 
if 
% 
eS 
hceincrten Sts 
i 
NOTES ON THE FLOWERS OF ANTHOXANTHUM ODORATUM L 
BY 
THEO. Homo. 
(With Plate XLvItt.) 
It is a very interesting and highly instructive task to study the mor- 
phology of the Grass-flower. The numerous variations, which occur 
here are well fitted to confuse our ideas as to the identity of the con- 
stituents of the flower, and a mere look into the considerable literature 
upon this subjectis sufficient to prove the difficulty of the study. While 
some authors have considered the development of the flower as the only 
reliable guide, others have thought to find the best explanation in the 
fully but abnormally developed flowers, of which several forms have 
been recorded in the Gramineae. It may not be denied that these 
aberrant forms, in many cases, are really worthless; but there are, on 
_the other hand, not a few which seein to be of some use to morphological 
studies. But it would not be necessary to study, for instance, the 
nature of the glumes of viviparous specimens to find out that they are 
identical with bracts, because we are able to see that in the develop- 
ment of these organs; and in a similar instance has Goebel* tanght us 
that the history of the development of the inflorescence in Cenchrus 1s 
sufficient to show that the so-called involucre is an abundantly ramified, 
but rudimentary system of axes, in which each axis aborts and merely 
shows a spine in the mature flower. Now, in regard to Anthovanthum 
the true position of the floral organs and at the same time the mor- 
phological identity of these ave so much disputed, haying been studied 
from normal flowers, we propose to study the same, but from abnor- 
mally developed specimens. 
The explanations which have been given by different authors as to 
the correct understanding of the flower, or rather the spikelet of Anthow- 
_anthum, show a great divergency of opinion. Most commonly the 
spikelet is described as consisting of one pair of empty glumes, two 
neutral flowers, represented by two flowering glumes with distinet 
awns, and finally one perfect flower, of which the flowering glume and 
the palet are nearly uniform. We have, then, three flowering glumes in 
the same spikelet which do not resemble each other, a fact that has led to 
*K. Goebel: Vergleichende Entwicklungsgeschichte der Pflanzenorgane (A 
Schenk’s Handbueh der Botanik, Vol. u1, p. 126, 1884.) 
Proceedings National Museum, Vol, XVY—No. 910. 
399 
