CONTRIBUTIONS FROM THE WISLEY LABORATORY. 315 



average weight of the plants was the same. In the grass no difference 

 was visible. The actual gains were therefore very slight, and nothing 

 like the result we had been led to hope for. Untreated peat had little 

 effect, depressing the yield of radishes (but the roots were the same as 

 when treated peat was used) and slightly increasing the yield of French 

 beans. Since only one comparison was possible in the case of the 

 radishes and French beans, differences of yield as great as those 

 mentioned might easily occur quite apart from any influence of 

 manure though they may have been caused by the manure. 



IV. Series 3. — Experiments in the Open Ground artificially 

 watered* 



The results obtained in the second series of experiments were far 

 less favourable to the peat than those obtained in the first. This 

 became evident while the turnips and radishes in the second series were 

 still growing, and the reason seemed to lie in either : 



1. A difference in the chemical composition of the peat used, or 



2. The operation of some limiting factor present in the open, and 

 absent from the series grown under glass, or 



3. The relatively small quantity of bacterized peat. 



The weather had been very dry during the season of growth, and 

 this suggested the possibility that insufficient water had been available 

 for the full benefit of the peat to be felt in the experiments in the garden. 

 In order to test this, Dr. Keeble suggested a further series of experi- 

 ments where some of the plots should be kept well watered and the 

 others should receive only the water which fell as rain. A series of this 

 nature was arranged, and at the same time the opportunity was taken 

 to compare the action of the peat with that of farmyard manure and of 

 other soil treatments. 



The site chosen was a piece of rather poor ground which had not 

 been manured recently, and, as will be seen from the plan (fig. 89), the 

 experiment occupied eleven plots, which were repeated four times, and 

 placed so as to neutralize errors arising from variation in the soil as far 

 as possible. 



Plots 7 and 8 were added to ascertain whether the presence of lime 

 to neutralize the acid of the peat exercised any influence on the yield. 

 As will be seen, the dressing of farmyard manure was quite light (at 

 the rate of 20 tons to the acre), that of the bacterized peat 1 ton to 

 the acre.*f Plots 8, 9, and 10 received no artificial watering, but 

 all the others were watered four times during the growth of the plants, 

 the same quantity of water being put on each plot on the same day, 

 about one gallon a square yard or a little more being used on each 



* I.e. twice as heavy as the dressing recommended in the recently published 

 book already alluded to. 



t It should be remembered, in comparing the weight of farmyard manure 

 with that of peat used, that the former contains a very much larger quantity of 

 water than the latter. The bacterized peat used in the second and subsequent 

 series had been dried before sending out. 



