901 



composition, as talmlated, of the diaiuage water of a plat before aud after application 

 of ammonia salts shows tliat] only 40 hours after the aiipiication uf tlie ammonia, 

 uitriheatiou has made a distinct commencennMit, and the jiroportion of nitrate 

 increases in the successive drainage waters obtained, the maximum of nitrate (50.8 

 parts per million) being reached 3 weeks after the api)lica(ion of the animouia. [The 

 most complete experiments on this subject are those by Schh'ising {('him. Agricole, 

 165, and Compt. rend., 109, 423]. 



The nitrification of manures other than ammonium salts has at present been very 

 insufficiently studied. Perhaps the most iuierestingof recent experiments are those 

 bj' Miiutz and Girard {Ann. Agron.. 1891, 289). These investigators place guano next 

 to ammonium salts in the order of nitrifiability ; following guano come green manures 

 (luceru and lupines), which, compared with other forms of manure, appeared to 

 be especially active in clay soils; the third class includes dried blood and meat, and 

 powdered horn; far below these stand poudrettc. avooI, and leather. 



Lecture VI. — JJraiitayeand weU /caters (pp. 95-112). — In soils in which 

 there is no movement of wiiter (a very rare condition) the movements 

 of the dissolved salts are determined simply bj' the laws of molecnlar 

 diffnsion. While the movement from this cause is considerable, far 

 more ra|)id changes arc produced when tlu^ soil water is in motion. To 

 study this subject, 223 mo-, of sodium chloride were applied to the sur- 

 face of a saturated column of soil free from chloride, 8 inches deep and 

 4.5 inches wide. At the end of a week 120 c. c. of water was daily 

 placed on the soil and nearly an equal quantity of drcainage water 

 obtained and analyzed. 



[It appears] that during the first 3 days the drainage water obtained was free from 

 chlorine. On the fourth day, somelliiug less than -171.8 grams of drainage water 

 having passed, the discharge of the chlorine comme?nced. We have here at once 

 evidence of the downward diftusion of the cliloride during the 11 days that had 

 passed since its application. If no diffusion had taken place and the chloride had 

 remained at the surface, it w^ould have required the application of 850 grams of water 

 to cause its appearance in the drainage, this being the amoniTt of water necessarj'^ 

 to displace the water already held by the soil. The chloride having already 

 descended a considerable distance, the displacement of about 400 grams of water was 

 sufficient to bring some of it into the drainage water. * * * 



It took 8 days to remove all chloride from the soil aud the drainage water contain- 

 ing chlorine amounted to 9.50.1 grams, a quantity in striking contrast with the 150 

 grams ol drainage in which the whole of the chlorine was contained when w'ater was 

 quickly drawn through a similar weight of air-dried soil by the air pump. 



On looking at the strengths of the successive runnings, it will be evident that 

 the chloride came through the soil as a wide band, diffusing both at its lower and 

 upper edges. This is the usual mode in which diffusible salts descend through a soil, 

 and a knowledge of this fact is essential if we are to understand the results of the 

 analyses of drainage waters collected within a moderate distance of the surface. 



[In natural soils, how^ever, drainage is not simply a matter of diffusion and dis- 

 placement. The process is complicated] by the fact that such soils always contain a 

 multitude of fissures, consisting of cracks in the soil and channels formed by worms 

 or left by decaying roots. These fissures communicate directly with the surface, and 

 down them w ater passes which has never jiermeated the whole mass of soil. The 

 result is that the drainage collected at a moderate distance below the surface is a 

 mixture of two Avaters of different composition, and at different periods in the run- 

 ning one or other of these waters will greatly preponderate. 



