238 JOTTINGS OF A GENTLEMAN GARDENER 



the summer, as, for instance, those on the use of liquid foods, 

 while for others, such as those with soil fumigants, the 

 winter is the best. 



The arrangement of an experiment calls for some little 

 careful forethought and plan, in regard to the matter to 

 be investigated. By way of illustration I will take the 

 question of the relative value of different manures or soil 

 fumigants, and describe my own method ; which, however, 

 does not differ from others. 



Two or three similar plots are prepared, and at least 

 twelve similar plants carefully selected to be grown tinder 

 similar conditions, except as regards manure in each plot. 



The plots should, if possible, be at least i ft. from each 

 other, and before any dressing is applied, the soil in both 

 must be of a precisely similar nature. 



Except for the manure or fumigant, the plots are then 

 all treated in precisely the same way, and at the same 

 time. When the manure or fumigant is forked into one 

 plot, the other plots are also forked over the same day, 

 with perhaps some other manure for comparison, or it may 

 be without any at all ; but their mechanical treatment, 

 that is the forking, must be the same. When the plants are 

 set in one plot, they are set in the others at the same time ; 

 and they are selected as nearly as possible of the same state 

 of growth and vigour. When one plot is hoed, the same 

 operation is carried out in the other plot or plots. And 

 the same with weeding and watering. The essential point 

 throughout is that all the plots, whether there are two or 

 more, shall receive precisely " similar " treatment in 

 everything but that one in each plot which is the subject 

 of the test ; the object being to compare results between 

 the plots with reasonable certainty that any marked 

 differences between them may be ascribed to the one thing 

 in which their treatment was not similar. 



The objects of such experiments may, of course, differ 

 very widely. They may be to ascertain the effects of 

 different manures, or of different quantities of the same 

 manure, or of different treatment, such as hoeing, or water- 



