14 
INVESTIGATIONS ON ROTHAMSTED SOILS. 
Table 1. — Broadbalk wheat plats, etc. — Continued. 
Plat. 
Annual manuring per acre (over 
50 years). 
Average yield per acre. 
21 years 
(1852-1872). 
21 years 
(1873-1893). 
42 years 
(1852-1893). 
b years 
(1889-1894). 
17 (A and B) 
and 18 (A 
and B). 
19. 
Average of 
min e r a 1 
years 
These plats in alter- 
nate years receive, 
the one mineral 
manures as on plat 
5, the other ammo- 
nium salts (400 
pounds), the treat- 
ment being year by 
year transposed. 
Thus, one year plat- 
17 receives the min- 
eral manures and 
plat 18 the ammo 
nium salts, while in 
the following year 
plat 17 receives the 
ammonium salts 
and plat 18 the min- 
eral manures 
(Rape cake plat.) 1852-1878 (27 
years), rape cake (500 pounds), 
ammonium salts (300 pounds), 
superphosphate (392 pounds); 
(average produce for these 27 
years, 29} bushels grain, 27* cwt. 
straw); 1878-1882. rape cake 
(1,700 pounds): 1883 and since, 
rape cake (1,889 pounds) applied 
in autumn 
Average of 
ammoni- 
um years. 
Mean 
nual. 
Bush. 
m 
3ii 
Cwt. 
16 
31* 
Bush. Cwt. 
30f 
23* 
12| 
29 
29* 25* 
91 
37| 
18| 
Bush. Civt. 
15* 12} 
30* 29* 
22| 21 
Bush. Cwt. 
29? 27 
22i 
25f m 
19} 
25j 
MECHANICAL COMPOSITION (STONES, FINE SOIL, ETC.). 
Having thus given a brief tabular history of the plats and their 
produce, we may return to the subject of soil sampling, giving an 
account of the results of the various systematic samplings that have 
been made of the Broadbalk soils. For the mere purpose of generali- 
zation or calculation it would no doubt suffice to give merely sum- 
marized or average results of the weights of the samples and their 
mechanical or physical composition ; but, seeing that avast quantity 
of detailed analytical work has been done the chief interest of which 
ultimately rests in its representation of quantities per acre, it seems 
desirable in a general record and discussion of such analytical data 
to give also an account of the detailed facts and figures which are 
la ken by Sir Henry Gilbert as the basis of the weights of fine dry soil 
per acre adopted in acreage calculations of the soil constituents. 
Furthermore, the various columns of figures are Interesting as show- 
ing the general nature of the variations in the distribution of the 
stones, fine soil, moisture, etc., in the various parts of this historic 
field, both as regards surface soil and as regards any given layer of 
subsoil, and also the differences shown to exist between one layer or 
stratum of subsoil and another. To some of these variations, as well 
as to the general indical ions of t he results, at tent ion may be specific- 
ally directed. 
