44 



INVESTIGATION INTO THE AMOUNT OF WATER 



cess of being weighed. The knife-edge of the balance was 

 relieved in the usual way by a support to the beam when not in 

 actual use ; this being removed or applied at pleasure by means 

 of a lever, (lie arrangement for which is not, however, indicated 

 in the drawing. The whole was, moreover, covered by a frame 

 of glass, provided with a door by which to gain access to the 

 weight-pan, and another for the attachment of the loose arm by 

 which the jar and plant are suspended. A standard counterpoise, 

 consisting of two leaden weights, was kept in the weight-pan, 

 the deviations only above and below this amount being deter- 

 mined by weights — a set of which, from ten thousand grains 

 down to one-tenth of a grain, was provided for this purpose. 

 As will shortly be seen, however, the amounts of water given off 

 by the plants were very large ; and after a time it was not deemed 

 necessary in practice to determine the weight within one grain, 

 or frequently even two. 



Between the time of planting and the full growth of the plants 

 more than twenty weighings of most of them were taken ; and 

 weighed quantities of water were supplied whenever it seemed to 

 be required. 



The collected results of the water supplied or given off by the 

 plants are exhibited in the following tables : — 



In the opposite Table (No. 1), as well as in those which 

 follow it, the results are arranged in two sections ; the upper 

 one bringing more prominently to view the comparisons between 

 the different plants with one and the same condition of soil, and 

 the lower one those of the same description of plant with the 

 varying conditions of soil. 



The summary of the total water given off during the growth 

 of the plants, as shown in the third column of this Table, is, of 

 course, chiefly of interest in connection with the coincident 

 accumulation of vegetable substance, and it will therefore be 

 repeated, and further considered, when we come to treat of that 

 part of the subject. Attention may here be called, however, to 

 the evidence afforded by a glance at the figures of this column — 

 whether in the upper or the lower section — at the much greater 

 regularity in Series 1 without manure, and in Series 2 with 

 mineral manure only, than in Series 3 with both mineral and 

 ammoniacal manures. Indeed, from the beginning, the plants of 

 Series 3 were unhealthy, and, as indicated in the tables, only the 

 wheat and the barley survived to the end of the experiment, and 

 these even gave a produce far inferior to the same description of 

 plants under the other conditions, though the ammonia provided 

 by the manure amounted in this case to only about 0.1 per cent, 

 upon the weight of the soil. 



It is seen that, in the cases of the healthy plants, there has 



