454 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 
ing glume) and palet, these inclosing normal stamens. At the tips of 
most of the branches of the tassel the spikelets are normal. While 
there is a gradual reduction in the size of the branches (see Pl. LX X XV) 
there is a very abrupt transition from the last of the pistillate inflo- 
rescences to the male flowers with three apparently normal stamens. 
It would seem from this that the abnormality is not due merely toa 
gradual transformation of the individual floral primordia into leaves 
(phyllody), but rather that a change affecting the entire bud takes 
place early in its history, causing the young bud to develop as a branch 
or young plant instead of producing a normal staminate spikelet. 
Furthermore, the number of primordia required for one of these 
growths is vastly greater than the number required in a normal 
spikelet. 
That the inflorescences that terminate these branches or plants 
should be pistillate is to be expected from their position on the upper 
part of the plant. Branches from the lower nodes of ordinary plants 
are the so-called ‘‘suckers,” which terminate in staminate inflores- 
cences. Branches from the nodes farther up have the terminal inflor- 
escence pistillate, forming the ears, while branches from the inter- 
mediate nodes, below the normal ears, usually bear terminal inflores- 
cences that contain both staminate and pistillate flowers. 
In the axils of the first leaves, which correspond to the outer ¢lumes, 
small roots could be seen (see Pl. LX XXIV), and when separated from 
the tassel and placed in the ground these apogamous plants took root 
and made considerable growth. Though none lived to maturity, they 
continued to grow in an apparently normal manner for nearly two 
months and produced roots over 1 foot in length. 
The production of roots enabling these branches to maintain an inde- 
pendent existence would seem to make this a true case of apogamy 
similar to that in onions, Agave vivipara, and the Arctic species of 
saxifrage. It would only remain for these apogamous plants to effect 
a natural separation from the parent plant to make the agreement 
perfect.¢ 
“The definition of apogamy given by Winkler in his ‘‘Parthenogenesis und 
Apogamie im Pflanzenreiche,”’ as reviewed in Nature for March 18, 1909, would 
seem to exclude all observed cases. The definition is given as follows: ‘‘Apogamy 
is the apomictic formation of sporophytes from vegetative cells of the gametophyte.”’ 
Apomixis is previously defined as the production of a new individual not preceded 
by fusion of nuclei. Hence apogamy is restricted to the formation of a new indi- 
vidual with cells containing the double number of chromosomes (sporophytes) from . 
cells containing the single number of chromosomes (gametophytes) without any 
union of nuclei. 
Even Yamonouchi’s case of a plant of Nephrodium molle developing from the pro- 
thallus and retaining the single number of chromosomes could not be included, 
since Leavitt's interpretation of this phenomenon as a case of homcosis seems well 
taken and the plant can hardly be considered as a sporophyte. 
It seems desirable to retain the term apogamy with its more general application to 
cases where a new plant is produced asexually from tissues which normally give 
rise to sexual organs. 
