8 BULLETIN 195, U. S. DEPARTMENT OF AGRICULTURE. 
thern into the laboratory, where they were allowed to wilt slightly, 
after which it was generally found that the pollen could be j arred out 
of the terminal pores into a watch glass. In varieties which developed 
pollen sparingly, the cells of the anthers were opened and the pollen 
grains removed. Pollen secured in this way gave very indifferent 
results, as a rule, and the method was superseded by the one which is 
now in use. 
The present method consists in gathering the flowers from the plants 
about as needed. "When considerable numbers of crosses are to be 
made and when good flowers are abundant, a number of them are 
gathered from each male parent to be used. These flowers are kept 
in small paper bags similar to those used for covering the emasculated 
flowers, each bag being properly labeled with the name or field 
number of the variety. In this way the operator may carry a 
considerable quantity of readily available material with him. When 
pollen of any particular variety is desired, a flower is selected from 
the proper bag and the corolla is pushed back between the forefinger 
and thumb and held in such a way that the stamens he directly across 
the thumb-nail. After removing the pistil, the anthers are tapped 
sharply with the forceps, and the pollen is jarred out, falling upon the 
thumb-nail (PL III, fig. 2), whence it is readily applied to the pre- 
viously uncovered stigmas of the emasculated flowers (PL I, fig. 2). 
The bag is then replaced over the pollinated flowers, again inclosing 
as much f oliage as possible. Usually the success or failure of the cross 
can be determined about one week after the pollen is applied. If 
there is a seeming affinity between the plants crossed and if an 
abundance of good viable pollen has been used, one will frequently 
find an almost full-grown seed ball at the end of seven days. As a 
rule, the crosses should be examined in fixe to seven days from the 
date of pollination. In most cases, at the expiration of this time, 
either the flower will have dropped off or the ovary will have swollen 
sufficiently to show that the cross has been successful. In such cases 
the paper bags should be removed and all seed balls that are devel- 
oping should be inclosed in loose cheesecloth sacks, which should be 
securely tied to the stems of the plants. It is hardly necessary to say 
that a record should be made of each step in the process and that each 
cross should be properly labeled. In Plate IV a cluster of seed balls 
is shown and also lateral, sectional, and basal views of individual 
seed balls. 
In connection with the methods just presented for the protec- 
tion of the flowers from insect visitors or other possible sources 
of outside pollination, it should be stated that two leading plant 
breeders claim that it is unnecessary to cover the potato blossom. 
It has been suggested by one of these breeders that the flowers are 
