422 
IOWA ACADEMY OF SCIENCE Vol. XXIV, 1917 
In Melilotus the giant grains are obviously abnormal. Moreover 
the development differs from that described by Juel and others 
in that the wall of the giant grain forms inside the old mother 
cell wall, while in the Cyperaceae the wall of the pollen grain 
is the modified mother cell wall. 
Miss Lyon (4) in her study of Selaginella found that in certain 
cases the megaspore mother cells formed spores without dividing 
or by dividing only once. 
Shattuck (6) working oil Marsilia, induced microspores from 
mother cells without the formation of cross walls. The abnormal- 
ities were brought about by' exposing the plants to a spray of 
cold water for several days. 
The formation of the giant grains in Melilotus resembles in 
some respects the formation of some of the giant microspores of 
Marsilia. In both cases the growth of the mother cell continues 
past the normal point of mitosis while cleavage takes place not 
at all or only incompletely. The variations from the normal in 
the case of Marsilia can be deafly correlated with certain envir- 
onmental factors. Any correlation in the case of Melilotus is very 
difficult since there appeared in the same flower and even in the 
same anther both normal and abnormal pollen. All of the flowers 
must have been subject to practically the same temperature and 
moisture conditions, as they were collected at points within a few 
feet of one another. 
As pointed out by Juel, the development of the giant grains is 
of some theoretical interest, for the facts show a striking re- 
semblance on the one hand with the development of the mega- 
spores of the heterosporous ferns and on the other hand with 
the formation of the embryo-sac of the Phanerogams. The de- 
velopment of these abnormal microspores indicates the homology 
of microspores and megaspores of higher plants as strikingly as 
does the abnormal germination of microspores as described by 
Nemec (5). In the one case the microspores resemble mega- 
spores in their formation and in the other case in their germina- 
tion. 
Perhaps the point of chief interest in connection with the 
giant grains is the variation in the extent of internal wall di- 
vision and the possible bearing of this upon the cpiestion of the 
formation of cell plates in the tetrad divisions. Since the divi- 
sions of pollen mother cells began to be carefully investigated 
