1916] HOAR~—STERILITY IN RUBUS 381 
of about 40 species of Rubus. In every case I picked those buds 
which were about to open, and in this way I was able to obtain 
pollen grains which were mature. If these were morphologically 
good, they were large, round, and very full of protoplasmic sub- 
stances. Material, on the other hand, in which the pollen had not 
matured normally showed it to be shrunken and empty, having 
lost all or most of its protoplasm and having but a vestige of the 
nucleus present. 
During the following winter I made a study of the pollen con- 
dition present in these buds, the method in most instances being 
as follows. Owing to the fact that perfectly formed pollen is very 
densely filled with protoplasmic substances, no harm resulted from 
killing directly with 95 per cent alcohol. Other more tedious 
methods were tried, but no better results were obtained. The 
buds were then pricked, the air pumped out, and the buds thor- 
oughly dehydrated in absolute alcohol. After this they were 
placed in a 2 per cent solution of celloidin dissolved in equal parts 
of synthol and ether. The grades of celloidin were raised 2 per 
cent every 12 hours, until 16 per cent was reached. Then small 
pieces of dry celloidin were added at 12- or 24-hour intervals, until 
a sufficient density was reached. When such a condition was 
obtained, the buds were removed one at a time and allowed to 
harden in chloroform for about 12 hours. They were then placed 
in equal parts of 95 per cent alcohol and glycerine. When com- 
pletely hardened these were mounted upon blocks in a 6 per cent 
solution of celloidin and cut upon the microtome. After being cut 
and stained in Haidenhain’s iron alum hematoxylin and counter- 
stained in safranin, they were mounted in balsam. Such a method 
allows the observer actually to see the pollen as it occurs in the 
anther, and by so doing it gives one a more accurate idea of the 
true condition. 
The genus Rubus in the last edition of Gray’s Manual is divided 
into 5 groups with Focke as an authority for their names. Of the 
first group, which is called the IparosaTIs group, I have 3 different 
forms (Rubus strigosus Michx., R. neglectus Peck, and R. occidentalis 
L.). It is here made a very small group, having but 3 well authen- 
ticated species. R. strigosus Michx. (fig. 1) is given as a synonym 
