72 



These experiments also show that both cotton-seed meal and cotton seed are 

 capable of supplying the plant abundantly with nitrogen. For soils similar to 

 these a combination of cotton-seed meal and acid phosphate, in varying propor- 

 tion from equal parts to one of the former to two of the latter, and used in quan- 

 tities from 200 to 500 pounds per acre, seems to be admirably adapted. 



(2) Distance ap<trt of foir.s (pp. 803, 30-t). — The object here was to 

 determine the best distance in width of rows for cotton on this pai"- 

 licnhir soih The distances varied from 2^ to 6 feet, and the results 

 suggest from 2^ to 4 feet as the proper width on such soils as this. 



(3) Distnnce a part of the plants in row (pp. 304, 305). — The ob- 

 ject here was " to test the distance required by cotton in the drill to 

 obtain best results." The rows were all 4 feet apart, and the distance 

 between plants varied from 8 to 48 inches. The value of the experi- 

 ment was impaired l)y the planting of several varieties of cotton 

 through mistake, but it Avas plain that distances from 8 to 20 inches 

 gave the best results. 



(4) Test of varieties (pp. 305-30T). — Thirty-one varieties of cot- 

 ton were tested. The results given in a table show that the yield of 

 cotton per acre varied from 1.444 t(^ 1.838 poimds and the yield of 

 lint (excluding Sea Island) from 420 to 586 pounds. The origin of 

 the seed which shoAved the highest yield of seed cotton per acre and 

 the next to the largest yield of lint is shrouded in mvstery. In this 

 trial and another at Baton Eouge this variety proved itself " the 

 superior of many of the so-called pedigreed cottons upon which time 

 and money have been prodigally exj^ended in propagating and de- 

 veloping." 



Field experirnents with corn (pp. 307-311). — (1) Maiuirial re- 

 quirements. — These were experiments with cotton seed and cotton- 

 seed meal, acid phosphate, and kainit, singly and in different com- 

 binations, on the plan of those with cotton above described. The 

 results were inconclusive, except as indicating that cotton seed and 

 cotton-seed meal were excellent soiu'ces of nitrogen. 



(2) Proper distances heticeen rows. — In this experiment the dis- 

 tances A^aried from 4 to 7 feet. The best results were with roAvs 5 

 feet apart. 



(3) Yield vitli late plantinc/ and effects of pnllinc/ off leaves for 

 fodder. — The results Avere decidedly unfaA'orable to pulling the fod- 

 der, the loss in yield of corn being eqiUA'alent to 7 to i) l)ushels per 

 acre, or from 15 to 20 per cent of the corn groAvn. Other experi- 

 ments haA'e uniformly brought like results. 



(4) Test of fourteen varieties of corn. — Tables shoAV ratios of corn, 

 cob, and shuck (husk) per acre. 



Forage crops {\)\). 312-320). — Notes on teosinte. pearl millet, Kaf- 

 fir corn, millo maize, a number of varieties of sorghum successfully 

 groAvn for fodder, and on Brazilian flour-corn, buckAvheat, Chap- 

 man's honey plant, and seA^eral Aarieties of pea-nuts and coAv-peas. 

 The prospect for. the successful groAvth of some of these plants in 



1 



