196 THE ROSE. 
are not very double will give more pollen 
and be better seed-bearers than the very full 
ones. Kinds like Bon Silene, Safrano, 
Fisher Holmes, Jean Cherpin, etc., will be 
found the most suitable for first experiments. 
If the operation is carried on out of doors, it 
is desirable to cover the flowers fertilized 
with fine gauze, to prevent the interference 
of insects. The insects are not apt to go to 
flowers from which the petals have been re- 
moved, but it is well not to depend on this. 
The heps should not be gathered until fully 
tipe, say after the first frost in October; 
they are then labelled and buried in pots of 
moist sand. The pots must be covered with 
glass or something of the kind to keep out 
mice, which are very fond of the pods. I 
allow the pods to remain in the sand till the 
first of January, they are then broken open, 
the seed taken out, and sown in boxes or 
pots. The seeds commence coming up three 
or four weeks after being planted. The 
seedlings appear with two leaves; so soon 
as they make a second growth they are 
pricked out by a knife blade and planted in 
small pots. They are very subject to mil- 
dew, and many are apt to pass away from 
this cause. In May they can be transplanted 
and put in open ground, in rich, well-drained 
