326 Field Experiments on Swedish Turnips. 
roots, which is a large produce, considering the light sandy 
character of the soil of the experimental field. 
2. That 5 cwts. of dissolved coprolites per acre on an average 
produced 1 ton 11 cwts. 1 qr. and 10 lbs. more swedes than the 
same quantity of finely ground coprolites. The cost of the 
manure in both instances was 1/. per acre. The increase over 
the unmanured plots, due to the application of mineral super- 
phosphate, amounted to 7 tons 5 cwts. 2 qrs. and 14 lbs. 
It will he seen that no other manure, not even the mixture of 
superphosphate and Peruvian guano, costing rather more than 
twice as much as the dissolved coprolites applied to plot 3, pro- 
duced so large an increase. On the light soil of Lansome-field, 
mineral superphosphate on plot 3 in Section A produced 27 tons 
9 cwts. and 4 lbs. of clean swedes, topped and tailed ; and on 
the duplicate plot 28 tons G cwts. 1 qr. and 16 lbs. ; or an 
average of 27 tons 7 cwts. 2 qrs. and 24 lbs. 
3. That finely ground coprolites on an average gave an in- 
crease of 6 tons 4 cwts. 1 qr. and 7 lbs. of clean swedes over 
the average yield of the two unmanured plots. 
4. That Redonda phosphate (hydrated phosphate of alumina 
and iron) increased the swede-crop by 7 tons 1 cwt. 2 qrs. on 
an average. Redonda phosphate gave nearly 1 ton more roots 
than ground Cambridge coprolites. 
5. That in these trials ground coprolites gave a larger in- 
crease than bone-meal, but less than dissolved bones. 
6. That precipitated phosphate of lime in this year's experi- 
ments gave disappointing results. On plot 5 B, as stated already, 
many roots were attacked by anbury, which, no doubt, told on the 
weight of roots ; but I do not see any reason why the produce on 
plot 5 in Section A was smaller than on any of the other 
manured plots. In similar trials in Warren-field precipitated 
phosphate of lime last year was used with very satisfactory re- 
sults, and I cannot therefore explain the unsatisfactory effect of 
this phosphatic material upon the swede-crop in Lansome-field. 
7. That the mixture of Peruvian guano and superphosphate 
had a less favourable effect upon the swede-crop on the light 
soil of Lansome-field than on the stronger soil of Warren-field 
in the preceding year. 
It will be further noticed that the weight of leaves on plots A 8 
and B 8 (manured with guano and superphosphate) was greater 
than on any of the remaining experimental plots. 
The effect of the nitrogen in guano on the light soil of 
Lansome-field appears to have been to favour too much the 
production of large tops, without a corresponding advantage to 
the development of the bulbs. 
I append analyses of the artificial manures which were used 
in these experiments. 
