692 



(3) Nitrogen is very effective when used in medium cpiantities in two doses, in 

 combination with three doses of phosphoric acid. A larger jirojiortion of nitrogen 

 apjiears to be injurious. 



(4) Cotton-seed meal as a source of nitrogen is equally effective with nitrate of 

 soda in the same combinations as i)lant food lV)r cotton. 



(5) The most etlective com1>ination of the three elements as a fertilizer for cottxm 

 on the soil covered by this exi>erimeut is three doses of superphosidiate 4tiS pounds, 

 containing 6() i)Ounds phosphoric acid; two doses of muriate of potash 7S pounds, 

 containing 39 pounds of potash; two doses of nitrate of soda 65 i>ounds, contiiining 

 120 pounds of nitrogen. 



iHtoruItural fert'dizinfi (pp. 141-147). — Two experiiiuMits to observe 

 the eti'ect of ai»i)lyiiig a ;;iveii amount of fertilizer all at oiiee ami in 

 fractions, were made in 1.S91 during the growing season and were simi- 

 lar to one made in 1800, reported in Bulletin No. 11 of the station (see 

 Experiment Station llecord. vol, ir. p. 5.");3). One experiment was ineon- 

 elusive. In the other superphosphate and muriate of potash were used 

 with cotton-seed meal or nitrate of soda, lieiug applied either all 

 before planting, i>art before planting, or part at jilanting. and the 

 remainder in two or three fractions later (June 1 and .Inly 1). 



(1) Successive or intercultural ap]tlications of nitrogen in the form of cotton-seed 

 meal are not i)rolifable. It is mor»* etlective when all is ai>p]ied at or before plant- 

 ing than when applied in two dosis. and le-^s etVective in tliree than in two. 



(2) Nitrogen in the form of nitrate of soda may be prolitablv divided into not more 

 than two ajiplications, tin* secon<l not to be later jirobaldy than June 1. It doc« 

 not a]))>ear wliat wftuld liave been the result if one half the nitrate had been applied 

 July 1 inst<:id of June 1. 



Tojq)iiuf voiiini (pp. 147-lHt). — A previous i'xp«'rimetit on topping 

 cotton was reported in linlUtin >so. 11 of the station (see Kxiteii- 

 ment Station Kecord, vol. ii, p. 554). One at-re, containing 52 rows of 

 cotton, all of which were mainired ami cidtivated exactly the same, was 

 used for the experiment in ISIH. •• Kvt-ry lifth row, commeming with 

 the iirst ami ending with the tiftylirst, was left untopped. On .Inly 1, 

 July 15, and August 15 each 10 rows were topped, each row topped at 

 any of the dates being s«'parated by 5 rows from the nearest row topped 

 at the same date.'' The yii'hls of cotton for the topi)ed and unlopped 

 rows at 5 successive pickings are tabulated. 



(1) It is again jiroved (as last year) tliat topping ]iroduces a material cffe«t on tlie 

 yield. Holh hist year and this year topping from .Inly 1 to July 15 injured the yield 

 most. Last year the mws not topped gave the largest yield; this year those lopped 

 August 15 gave the largest. 



{'2) Topjiiiig cotton is a hazardous e\|HTinient and very uncertain. The jnobalile 

 effect can not be foreseen with sutlicieut certainty. It is probal>le that many farmers 

 have to]i|(ed their colt«ui and liejieved the eli'e<t was bencticial when in fa<t careful 

 com]>etitive ex]ieriment would \\:\\v jirovcd the lontrary. 



It is suggested that the conditions under which topi»iiig would be most likely to 

 prove bencticial are when the cotton plants will i>robably be large — i or 5 feet 

 high — and have good distance. If crowded in the ri:\v to])ping tends to crowd the 

 foliage still more by causing the branches to grow longer. 



Drrp rs. shtdloir citlfurr of mfton ([>. 14!>). — An exiu'riment on (» jdats 

 siinihir to the on*- leiiorted in Ilnbetin No. 11 of the station (see Kxper- 

 imeiil Sl:ilion iaHoid, vol. il. p. 554). 



