34 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



When the various substances had been watered into the subsoil 

 and walls (as far as possible), they were watered about a week later 

 with plain water and then sprinkled with hot lime. A layer of spent 

 tan i inch thick was then put on the subsoil throughout every house, 

 treated and untreated alike. The fresh virgin soil previously mixed 

 with farmyard and artificial manures was placed in the houses about 

 a fortnight after the plots had been chemically treated. 



The ends of each plot were indicated throughout the growing 

 season by vertical lines of paint on the whitewash of the walls. 



Selection of the Plots for Experimentation. 



During the seasons immediately preceding that of the experiment 

 here described, the owner of the greenhouses had made a practice 

 of marking the places where diseased plants (i.e. plants with nodulated 

 roots) were found. The diseased places were therefore located without 

 difficulty, and each plot of ground selected for experiment included 

 one or more of these infected places. To make sure of the presence 

 of the disease (as well as to determine the limits of the plot) , the sub- 

 soil was in all cases forked over to a depth of 3 inches, and if nodules 

 were found therein the place was chosen. 



Though the selected plots were in this way proved to harbour 

 the parasite in the subsoil, yet it is to be clearly pointed out that 

 the infection was not equal. For example, Plots 23, 24, 20, 29, 4, 

 &c. were very heavily infected, whilst Plots 1, 9, 36, 37, 39, &c. were 

 less so. The plots were of varying lengths according to the lengths 

 of diseased subsoil found at each place. 



Difficulties. 



From the experimental point of view this inequality in the infection 

 was rather unfortunate, but in the circumstances one could not 

 make an attempt to infect each one heavily before treatment. Un- 

 fortunately, too, the area at our disposal was not great enough to 

 have a full series for each trial substance, and to make a large number 

 of comparatively tiny plots would have been out of the question. 

 It was simply a matter of trying as many substances as the number 

 of infected places admitted of, and gaining as much information as 

 the circumstances allowed. Even in the first year, difficulties occurred 

 in getting some of the substances for experiment, and each year this 

 difficulty increased. 



Treatment of Subsoil. 

 Trial Substances. — In 1915 the substances tried were as follows : 



1 Mineral Acids . . Nitric acid and Chromic acid. 



2 Alkalies . . . Caustic soda. 



