3o6 Journal of Agricultural Research voi. xviii. no. 6 



spread in future years and also whether any of the different rotation and 

 tillage treatments were really effective in retarding this spread. 



These rotation plots are X acre in size, being 264 feet long and 41.25 

 feet wide, and afford space for 10 rows of cotton 4.1 feet apart. Each 

 row was carefully measured, and the location of the portion of the row 

 in which the plants were dead was indicated on a diagram drawn to 

 scale. The survey of 191 6 was made near the end of the growing season, 

 October 21, after the final picking of cotton had been finished. One of 

 these plot diagrams, showing the areas of dead plants, is shown in figure i. 

 This diagram shows two main areas of infection, as shown by the brush- 

 like lines. A count of the living and dead plants in this plot at the time 

 the survey was made showed that 60.5 per cent of the total number of 

 plants were dead. This plot had been planted in cotton each year since 

 1909. It had been plowed in November each year and had received an 

 annual application of manure at the rate of 12 tons per acre. 



The plowing and other tillage operations were made lengthwise of the 

 plot so that any distribution of soil infection would naturally be favored. 



^ B«B«!iafWi^!'ygirim?,'gWgJ MMJKTM»mjV))>«w>n««nyj«MaliMimim««^^ 



<S Wa.WRIIWiyPFWMI'g >»»yK»i«a«»i« - ■4<-»ji<gf;«>T>>MW8>!w«t-.H.wa'»».^vMt».i>A«««a/.t.ira^ 



S" atM>»«»Rg»aa!gi»})iwj(BaCTii^ 



/O 'miSm. ».gi\\m»r»-'.aKfc »«.>i;>tg,.g<iiimilW!»!V!!VilK«!l 



/='S/? c-^/yr as^'^ XPAP /S>/CJ> XP/i^ XPxl^ AS>/G 



Fig. I. — Diagram of plot Bs-4, showing by the brushhke lines the portions of the rows in which the cotton 

 plants were killed by rootrot in 1916. 



Notwithstanding this fact, the limits of the affected areas were very 

 sharp, dead plants standing adjacent to living ones in each row at the 

 edge of the diseased area . 



This same plot was planted to cotton again in 1917, each row being 

 planted as nearly as possible in the same place as in the previous season. 

 The count of living and dead plants and the diagram of the areas of 

 dead plants were made on October 25, 191 7, at the end of the growing 

 season. The count of plants showed that 36.8 per cent of the total 

 number were dead with symptoms of rootrot. The diagram of the plot 

 for 1917 is shown in figure 2. 



The distribution of the disease in 191 7 was more scattered than in 

 1916; and the spot that is shown on the north side of the west half of 

 the plot in the 1916 diagram (fig. i) gives some indication of a pro- 

 gressive spread, in that living plants were found in 1917 where only dead 

 plants were noted in 19 16. This tendency for the spread of the disease 

 to take place like the spreading of a fairy ring is not very pronounced, 

 however, as can be seen in the diagram for the east half of the plot, nor 

 is it to be found so definitely expressed in the diagrams of other plots. 



