Salt Experiments and Mangolds. 241 
general no better answer can be given to the inquiry " What 
have we learned from these trials ? " than simply this : " in some 
cases salt did good — and in others, not." Such a reply clearly 
can be no guide for successful applications in future. 
Impressed with the reputation which salt has gained as a 
manure for mangolds, I tried some experiments in 1862 in the 
hope of ascertaining the quantities which might be used to 
the best advantage. In recording the results of these and similar 
experiments, I consider it very important that the physical and 
chemical properties of the soil should be described as carefully 
as possible. Observations on a large scale have convinced me 
that the nature of the soil has a great deal to do with the success 
or failure in all cases where salt is tried as a manure. 
The field in which the experiments were tried was in excel- 
lent condition, both as regards manuring and mechanical work- 
ing. It bore clover in I860, and a good crop of wheat in 1861. 
It received a good dressing of dung in the autumn of 1861, and 
was in capital order when the mangolds were sown in 1862. 
The soil was a calcareous clay, containing no sand, and a suffi- 
cient amount of clay to render it decidedly heavy and difficult to 
work in wet weather. The subsoil on which it rests is stiff clay. 
A portion of the soil, taken from a large sample removed from 
different parts of the field, was submitted to analysis, and the 
following results obtained : — 
Composition of Soil from Experimental Mangold-field of the Royal 
Agricultural College Farm, Cirencester. 
Moisture (when analysed) 4-710 
Organic matter and water of combination .. .. 1O022 
Oxides of iron and alumina 15'602 
Carbonate of lime 16-448 
Sulphate of lime -375 
Phosphoric acid - 072 
Carbonate of magnesia -798 
Potash -832 
Soda • -098 
Insoluble siliceous matter (clay) .. • 50-787 
99-744 
It will be seen that clay is the preponderating constituent of 
this soil, which contains also a considerable proportion of car- 
bonate of lime. 
Nine plots of l-20th of an acre were selected for the experi- 
ments from the most even part of the mangold-field. The man- 
golds grew vigorously where the salt was sown by hand along both 
sides of the drills on the 29th of July. Each plot consisted of four 
drills. One plot received no salt ; the eight others were dressed at 
the rate of from 1 to 8 cwts. of common salt per acre. The roots 
were counted and carefully weighed after cleaning and dressing. 
