srCAK 1\ .MAMii:i.S 117 



or willi Vi\\)c viikc as llic sourci' of nitmi^M-n arc less watfiv 

 tlian tliDse ^^"()\^•n witli iiiti'alr of soda, rapi' cake |»ro(lu('iii;: 

 the richest roots for tlirir size. As reirai'ds its {'\]\'{\ lioth on 

 the ina<j;iiitiiile and (.-oinpoNition of tlic crop. I'do 11». of iiiiid;,'(Mi 

 in (lung are less effective than 80 11). in nitrate of soda or O.s jh. 

 in rape cake, and have about the same value as SO 11). of nitro^'en 

 in annnoniuni-salts. 



It has already been noticed that the use of anniioniuni 

 salts promotes an earlier maturity than does nitrate of soda : 

 this is seen in the generally higher "quotient «>f purity " of the 

 "A" as compared with the "N" series. The glucose co- 

 efficient is correspondingly lower in the *'A" scries. Excess 

 of nitrogen has the same effect as the sul)stitution of nitrate of 

 soda in lowering the quotient of purity and raising the glucose 

 coetticient. For instance, Plot 2 N gives a (|uoticnt of pni-ity 

 of 55'5 as compared with 64'9 and 62*2 on IMot^ 4 N and «) N. 

 which receive the same mineral manures but not the extr.i 

 nitrogen of the dung on Plot 2 X; .similarly, ilie glucose co- 

 efficient is 6 5 on Plot 2 X and only ;3 and oA on 4 N and 

 6 X"". Again, all the plots on the AC series, receiving both 

 rape cake and ammonium-salts, show worse results as regards 

 purity than the corresponding plots on either the A or the C 

 series, which receive only one portion of the nitrogenous 

 manure on series AC. 



The dependence of sugar-formation upon potasli is well 

 seen by comparing the weights of sugar per acre produced on 

 Plots 4 or 0, receiving potash, with the corresponding weights 

 from Plot 5, without potash ; or by comparing Plots 2. where 

 dung, nitrogenous manures, phosphates, and potash arc applied, 

 with Plots 1, which receive dung and nitrogenons mainn*cs only. 

 To this latter statement the nitrate Plots 1 and 2 affbrd an 

 exception. As a rnle, however, the percentage tif sugar in tin- 

 root is little if at all increased by the u.se of jMitash ; the effVct 

 comes from the increased crop, and is aj)parent in tli*- amount 

 of sugar grown per acre. The ([uotient of purity is. however, 

 better on IMots 4 and <3, with potash, than on Plots ."i, witiioul 



