92 THE BOOK OF THE ROSE CHAP. 
which are gritty, it would be interesting to know if 
their ashes would show a larger proportion of silica, 
or whether it is only the extra drainage and conse- 
quent heat of the soil that proves favourable to them. 
Iron is present in but small quantities, highest in 
the stem and lowest in the flowers. Sulphate of 
iron is said to give a better colour to the foliage, 
and in some sorts, such as Anna Olivier, to the 
bloom as well, but I do not place much reliance on 
it. In fact I have sometimes omitted the iron from 
the formula below, for the small quantities cause a 
good deal of trouble to the makers, but it should be 
remembered that every one of the constituents is 
necessary, or the others are by so much rendered 
valueless; so perhaps it had better be included, 
though most soils and natural manures have it in 
sufficient quantity. 
The following is the mixture recommended by Mr. 
Tonks for the Rose, as deduced from the ash analysis : 
Superphosphate of lime, 12 parts. 
Nitrate of potash, 10 parts. 
Sulphate of magnesia, 2 parts. 
Sulphate of iron, 1 part. 
Sulphate of lime, 8 parts. 
30 parts. 
The manure is to be applied in early spring: 
immediately after pruning would generally be soon 
enough perhaps, but February would be better. It 
is to be evenly scattered on the previously hoed 
surface, at the rate of } lb. to the square yard, so 
that the Rose grower can soon calculate how much 
