130 EXPERIMENT STATION RECORD. 



Data n-latini,'' lo chaiiircs in llif iiitr()t.'('ii cnntciit of sandy soils arc reported and 

 discussed in detail. 



The duration of the effect of green manures, A. PetkkmaXxV {Bui. Ind. Chim. 

 et Jktct. (iemUoiix, 1903, No. 7S, pp. 11-21; Jhd. Ayr. [Brvssel^], 19 [1908), No. 4, 

 pp. 504-514). — Experiments with a nninher of crops grown on soil which had been 

 in grass, the sod being tnrned nr.der, and on soi which had been underclean culture, 

 indicated that tlu^ effect of such green manuring is quite noticeable on early potatoes, 

 less marked on late potatoes, still less marked the second year, and disappears the 

 tliird. The use of lime to accelerate the action of green manures is suggested. 



Fertilizer experiments on the sewage-irrigation fields of the town of 

 Dortmund, IIeppk and Gxikss [DeiiL L(tndn: I'res.'<e, oO {1903), No. 55, pp. 490, 491, 

 1i(j. 1). — l'',xperinients are reported which show the importance of using potash and 

 pliosphoric acid in connection witli sewage irrigation. 



Some remarks on the object and method of conducting field experiments 

 with fertilizers, F. W. D.vfekt {ImikIm: Jaltrh., 32 {1903), No. 1, pp. 149-159).— X 

 reply to a criticism by T. Pfeiffer of the failure to include check plats in the plan of 

 exjieriments carried out by the author, E. Meissl, and O. Reitmair (E. S. K., 12, 

 p. 889). The general questions of the purpose, plan, and method of conducting 

 field experiments are discussed. 



The action of commercial fertilizers on sandy soils of different degrees 

 of fertility, JiAcuMAXX {FuhUiu/'.^ Lnndir. Zlij., 52 (1903), No. 11, }>j>. 373-378).— 

 The results of 14 experiments with rye on sandy soils of different degrees of fertility 

 show that on the average the yield of grain was dou])led by the use of mixtures of 

 Thomas slag, kainit, or 40 per cent pcjtash salt, and nitrate of soda or sulphate of 

 ammonia. The use of fertilizers was profital)le in all cases. 



Investigations on the value of the new 40 per cent potash fertilizer as 

 compared with kainit, W. Schneidewind et al. {Arh. Deut. Landw. GeselL, 1903, 

 No. 31, fij). 168). — The results of 3 years' comparative tests on a variety of crops are 

 discu.^sed in detail. Previous ex}ieriments have been referred to (E. S. R., 14, p. 852). 



Conditions determining the poisonous action of chlorids, H. J. Wheeler 

 and B. L. Haktwell {Pdtode Ishind Sta. lipt. 1902, pp. 287-304, fig^. 6') .—This ques- 

 tion has been studied in a series of pot experiments l)egun in 1894. The experiments 

 incluvled tests of the effect of calcium chlorid on potatoes, and magnesium and 

 ammonium chhjrids on barley, rye, and oats. 



"Calcium chlorid and ammonium chlorid were fotmd to exert a marked poisonous 

 action upon certain plants, w'hen applied to a soil which was already somewhat acid. 

 Magnesium chlorid was not found to be poisonous under conditions where great 

 injury from calcium chlorid and ammonium chlorid resulted. 



"Calcium carbonate and caustic magnesia used singly, also a mixture of basic slag 

 meal with the carlxmates of jwlash and magnesia, were found to prevent or over- 

 come the ill effect produced by applications of either calcium chlorid or ammonium 

 chlorid. 



"TluM-e seems to l)i' no good foundation for the caution about using sulphate of 

 ammonia and nmriate of potasli or kainit in the same mixture on account of the for- 

 mation of anunonium chlorid, provided the soil is duly tested with blue litmus 

 paper, and lime, wood ashes, or other suitable substances are applied to correct or 

 partially overcome the existing acidity. 



"It is probable that all of the apparently conflicting evidence regarding the 

 poisonous and nonpoisonous action of chlorids upon plants, when they are ai)plied 

 in reasonable (juantities, could be explained Avithout difficulty, had due attention been 

 paid in all cases to the chemical reaction of the soil." 



Field experiments with ammonium sulphate and sodium nitrate, Kloepfer 

 {Fidditufx L<nidir. Zt;/., 52 {1903), Nos. 10, pp. 341-34-5; 11, pp. 388-393) .—The 

 results obtaine<l in Held experinients w^ere variable, depending apparently largely 



