SOILS FERTILIZERS. 



719 



amount of water applied at any one time was at the first irrigation, which 

 amounted to 2.!)-i in. In the three irrigations which followed 1.47 in. was 

 ai)i)lied, subseciuent irrigations being 2.2 in. each. 



The following table gives the results of determinations of nitric nitrogen 

 in the drainage water of the different lysimeters expressed in percentages of 

 the original amount of nitrogen aj)plie(l to each lysimeter, namely, 10.3 gm. 



Mtrir nilnxjuti in draiituyc for different periods. 

 Ll'erccntage of nitrogen added to soil.] 



Period and date. 



Tankage. 



Fish 

 scrap. 



Hoof 

 meal. 



Nitrate 

 of soda. 



Sulphate 



of aimmo- 



nla. 



Dried 

 blood. 



Period 1, Mar. 14-Apr. 4.. 

 Period 2, Apr. 4-May 31 . . 

 Period 3, Mav 31-Jiily 'Ih . 

 I'eriod 4, July 2.'i-Sept. 19. 

 Period .'), Sep"t. 19-Noy. 21 



Total 



4.64 

 2.47 

 4.26 

 2.94 

 3.53 



0.66 

 4.59 

 9.91 

 8.40 

 4.51 



0.72 

 8.37 

 18.07 

 7.65 

 3.63 



85. .50 



2.62 



.82 



.94 



.54 



1.53 



36. 74 



30.00 



5.91 



1.40 



17.84 



26.75 



8.44 



90.42 



75.58 



0.73 

 5.70 

 14.14 

 10.21 

 2.89 



33.67 



As will be noted, the amounts of nitrogen recovered as nitrate in the drainage 

 water varied from 90.42 per cent in the case of nitrate of soda to 17.84 per 

 cent in the case of tankage during a period of practically 8 moutlis. It will also 

 be observed that the nitrate of soda passed into the drainage water quite 

 rapidly, 85.5 per cent being so recovered in the first period. The sulphate of 

 ammonia was more slowly recovered and does not appear in large amounts as 

 nitric nitrogen in the drainage until tlie second and third periods, but as a rule 

 precedes in this respect the oi'ganic materials used. There was little evidence of 

 any considerable amount of denitrification notwitlistanding the fact that the 

 moisture content of the soils was maintained at the high average of about 87 

 per c(>nt saturation. Eight months after the various nitrogenous fertilizers 

 were applii'd the drainage waters contained varying amounts of nitric nitrogen 

 and this was as true of the nitrate of soda lysimeter as of the others. 



Observations on the effect of nitrate of soda, sulphate of ammonia, and dried 

 blood on the removal of lime from the soil showed that while soil to which none 

 of these materials were applied lost 10.02 gm. of lime, that to which nitrate of 

 soda was added lost 29.44 gm., 1.'>.42 gm. of which was thei'efore due to the 

 effect of the nitrate, and in the same way sulphate of ammonia was responsible 

 for the loss of 19.58 gm. and dried blood 2.25 gm. These results witli fresh 

 water are quite contradictory to those obtained with saline water as previously 

 reported (E. S. R., 14, p. 554). 



Seeoiid seiics (pp. 20-31). — In the e.xperinumts with different lime compounds 

 the materials used were applied at I'ates furnishing 100 gm. of lime per 100 lbs. 

 of water-free soil ; that is. at the rate of 3.92 tons per acre in case of burnt lime, 

 8.87 tons ground coral, and 11.08 tons of gypsum. 



Examinations of the drainage water show that thei'e was an approximate 

 decrease of nitric nitrogen per acre of 1.69 lbs. in case of burnt lime, a gain of 

 3.1 lbs. in case of ground coi'al, and a decrease of 7.54 lbs. in case of gypsum as 

 compared with the amount found in drainage of untreated soil. Calculated on 

 the basis of gain or loss per ton of material, the figures are a decrease of 1.66 

 lbs. in case of burnt lime, a gain of 1.35 lbs. in case of ground coral, and a de- 

 crease of 2.61 lbs. in case of gypsum. The prejudicial effect on nitrification of 

 the quicklime is attributed to excessive alkalinity produced by the use of this 

 material. " The deleterious action of gypsum in this particular is, in the 

 opinion of the writer, an indirect one, and was caused by the large amounts of 



