356 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 
Each of the trees — flowers of which were counted in the control 
experiment — are carrying excellent cropSi whilst others equally well 
flowered have only poor crops, so that the percentage is probably 
higher than would have been the case had a wider range of varieties 
been included. 
For several reasons it has not yet been found possible to collect 
suflicient data with regard to the viability of the pollen of the 
varieties, but an examination was made of pollen from sixteen varieties 
growing outdoors, in an endeavour to find, with the aid of the microscope, 
the approximate amount of good and bad grains in average samples. 
Of the sixteen varieties examined, the minimum amount of mis- 
shapen and apparently bad grains was 8 per cent., this being in pollen 
from Frogmore Orleans, whilst the maximum amount of bad pollen 
found was 25 per cent., in samples from Coe's Golden Drop and 
Bryanston Gage. The average amount was 17 per cent, misshapen 
and apparently bad in the other varieties. 
