340 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 
not seen bearing pollen; and, besides this, they are more easy to fertilise 
than any other type. There are some varieties among the Japanese which 
produce bracts or green scales (if either term is botanically correct) 
between the florets, something after the style of a double Zinnia. These 
are among the easiest to fertilise. I cannot account for this unless their 
presence gives the ovary more room to develop, as these scales appear to 
take the place of florets. Consequently the seed vessels are not so tightly 
packed together as in other varieties. 
Being anxious to test whether my theory that the non-fertility of the 
flowers of large plants was due to too high cultivation, a few years ago I 
selected several large plants, treating them the same as the smaller ones; 
those intended for the female, or seed bearing, did not mature a seed, while 
almost all the centre florets of the pollen bearers matured seeds. These 
were sown in due time, but all came single or semi-double, and although 
all must have been fertilised with their own pollen (as I used separate 
brushes), not one resembled its parent. I also found on examining the 
outer florets of the flower intended for the female parent that, although 
the stigma looked very tempting to an hybridist, the ovary had taken the 
form of a small deformed floret; in fact some of them, though small, had 
thrust themselves up the tube of the floret. From this I formed the 
idea that the outer florets would produce the most double flowers, and 
from results obtained I found that the extreme outer row produced from 
80 to 90 per cent. of double flowers, while the nearer the florets fertilised 
are to the centre, the greater the percentage of single flowers. I now 
remove all but the outer row of florets before commencing operations, 
except in the case of incurved Japanese and true incuryeds, when several 
rows are left, as it is not desirable to have the centre too full in this type. 
I haye also tried various times of the year for propagating for this 
purpose, but have found none more suitable than that previously stated, 
for if started too early the buds which form first have a tendency to 
doubling in the outer florets the same as in the large plants. This would 
be a greater drawback still in the pollen-bearing plants, as an open centre 
is essential, and blooms from this bud in most cases would be too full in 
the centre, while if the plants are grown on to the next bud the plants 
would require'a larger pot: the plants would not be such a convenient 
size, and the raiser gain no better (if as good) results. 
In conclusion, with my short experience in hybridising the Chrys- 
anthemum, it would be assumption on my part to pose as an authority 
on the subject, as there are many who have had a longer experience than I 
who may differ with me in some points. Therefore, although this paper 
is based on observations made during my short experience, I may yet 
have reason to alter my convictions in some respects, as I believe there 
is yet a great deal to be learnt on this particular subject. 
