the season, bad though this had been. The weeds were no doubt 

 closely connected with the failure, and some observations were 

 begun, and are being continued, to discover their effect on the crop. 

 It is not a sufficient explanation to say that the weeds took water 

 and food from the soil ; in point of fact the Broadbalk soil during 

 the growing season was moister and richer in nitrate than the soil of 

 the other fields. Alopecunis is one of the worst of the weeds; it 

 seems first to have caused trouble on the dunged plot, where it was 

 noticed among other weeds in 1869, and was very rampant in the wet 

 seasons 1878 and 1879. In 1886 and succeeding years it was so 

 bad that extensive hand weeding became necessary and many of the 

 wheat plants had their roots loosened during the process ; by this 

 time it had become bad on Plot 11 also (receiving ammonium salts 

 and superphosphate). The extreme course of fallowing the land was 

 adopted in 1904 and 1905, one half of the field only being sown in 

 each year, but still the grass survived : since then wider drills have 

 been adopted (12-in.) to facilitate hoeing and hand weeding, but 

 again without success ; and now narrower drills are being tried. 



A hypothesis has been put forward that the wheat plant excretes 

 a substance toxic to itself which is gradually accumulating in the 

 Broadbalk soil ; laboratory experiments have so far failed to support 

 this view. In any case Alopecunis seems to excrete nothing toxic 

 to itself. 



The continuous barley plots on Hoos Field were completely 

 fallowed this year in the hope of clearing the land of weeds. 



The mangolds (Sutton's Yellow Globe) were drilled in Barnfield 

 on May 6th. As usual of late years, the plants on the dunged strips 

 started well and continued well ; they were, however, somewhat at- 

 tacked by a leaf spot. The fate of those on the strips receiving no 

 organic manure varied with the state of the ground ; where a fine seed 

 bed had been obtained (as on Strip 5, receiving superphosphate only) 

 germination was rapid, but the young plant was killed by the drought 

 in May, and reseeding became necessary. On the other hand, where 

 the ground was rough (as on Strip 4, receiving complete mineral 

 manure) germination was delayed till the drought was ended, and 

 then the plants made satisfactory growth. One of the most interest- 

 ing features of the plots receiving no organic manures is the way in 

 which the operation of manures upon the texture of the soil becomes 

 magnified. 



The 17th four course Rotation was begun on Agdell Field with 

 a crop of swedes, the variety being Sutton's Magnum Bonum. Seed 

 was drilled on May 24th and germination was greatly favoured by 

 the rain early in June, so that a good plant was obtained, which grew 

 well right through the season. The completely manured plots 

 gave a yield of 26i tons per acre. The plot receiving super- 

 phosphate and potassium salts only, but no nitrogen compounds and 

 on which no clover is grown, gave 7^ tons of roots and \ ton of 

 leaves, but the plant showed all the signs of nitrogen starvation. 

 Last year this plot gave an extraordinarily high yield of wheat; almost 

 32 bushels of grain and 29 cwts..of straw. The source from which 

 the necessary nitrogen is obtained is not wholly evident. The con- 

 tinuously unmanured plot gave the usual diminutive roots, 8 cwt. 

 per acre only bemg (jbtained where no clover is grown, and Z'i 

 cwt. where clover had been taken two years before. It was early 



