Experiments on the Artificial Fecundation of Wheat. 217 



annum on each over 12 years, 1852-1863, is given in the 

 following Table (I.) : — 



Table I. 



Average Produce of Wheat per Acre per Annum on Duplicate Tlots. 

 Over 12 Years, 1852-1863. 



Thus, taking the average over 12 years, the duplicates, in 

 each case, give almost identical results. It is true that in indi- 

 vidual years the difference was sometimes much greater. For 

 instance, to take the most extreme case which the 12 years afford, 

 in 1855 the difference between the plots a and h amounted in each 

 case to about 3J bushels of dressed corn, and from oh to 4 cwts. 

 of straw per acre, the advantage always being with the plot b ; 

 and for several years afterwards the plots a gave in their turn 

 the higher produce, until the inequality had been in a great 

 degree compensated. From these facts it is evident that even 

 Avhen the greatest care is taken there may sometimes be a con- 

 siderable difference in the produce of plots treated as far as 

 possible in the same manner, arising from circumstances of the 

 season over which we have no control. It will, at the same time, 

 be admitted that the plots a and h were, upon the whole, in a 

 very uniform condition, and well suited, therefore, for the pur- 

 poses of an exact comparative experiment. 



The duplicates being again manured exactly alike as in former 

 years, plots 12a, 13a, and 14a were submitted to M. Hooibrenk's 

 process ; whilst plots 12Z', lob, and ]4Z» were not so treated. In 

 accordance with the directions given, the mode of proceeding 

 was as follows : — To a rope long enough to stretch across the 

 plot to be operated upon, pieces of woollen twist 2 feet in length 



