22 BULLETIN 1386; U. s. DEPARTMENT OF AGRICULTURE 
THE AWN 
The sterile lemma, fertile lemma, and palea are for the most part 
concealed by the glumes, although occasionally they are somewhat 
exserted at their summits. Their fragile nature also lessens their 
value for purposes of classification or identification. The awn 
affords a positive difference, as it is either present or absent. It 
varies in regard to some minor features, as, for example, length and 
color, there being appreciable differences between some varieties 
regarding its length, and there sometimes being a short or rudi- 
mentary awn that does not project beyond the glumes, or at least 
not enough to be easily detected. Panicles have never been found 
in which the awns were reduced to short projecting points, nor have 
awned typo been encountered in which all the distinctive features 
of the awn were not generally present — that is to say, in which they 
were not twisted, geniculate, and scabrous toward the apex. Forms 
exist which are similar except for the presence or absence of the awn, 
and it has been possible to isolate from certain mixed populations 
either an awned or an awnless form. In apparently all varieties the 
lemmas drop the awns at or before the maturity of the panicle. If, 
however, the variety has awn-, they will always be found in the 
panicle in greater or less numbers, even in dried specimens, unless 
they have been handled with undue carelessness. 
THE CABTOPSIS 
The character- of the caryopsis are useful in helping to differen- 
tiate both the principal varieties and the secondary forms, the diver- 
sity being in form, color, length relative to that of the glumes, 
together with minor differences resulting from the variations in 
form. Comparative length and coloration show pronounced con- 
trasts, and these have been employed by a number of writers. The 
length has been used by some writers to differentiate groups, and 
coloration to distinguish between varieties within the groups. Dif- 
ferences in both these characters are useful in keys and in descrip- 
tions, but especially in the latter, where they are of value in substan- 
tiating determination- made on other features. 
The caryopses disarticulate from their rachillas, and in certain 
varieties they also separate in the main from the spikelets during the 
threshing process. In the case of some forms, however, only a part 
of the grains is dislodged, the proportion remaining in the spikelets 
varying both with the variety and with the severity of the threshing, 
so that this doe- not constitute a good character for the purpose of 
identification. When the glumes remain attached to the threshed seeds 
or are found intermixed among them in commercial samples, they are 
usually more or less mutilated or at least have been deprived of their 
awns, much of their pubescence, and the sterile spikelets. Points of 
distinction in the caryopsis, although often evident along varietal 
lines, generally speaking, are not so numerous or so sharply marked 
as in the glumes, and the extent of variation within varieties pertain- 
ing to features of this member is somewhat greater. Although a ten- 
dency toward a definite configuration can usually be seen in the cary- 
opses of each type, fluctuations are generally considerable, and differ- 
