No. 4.] TOBACCO GROWING. 29 



curing barn, and, after sorting, large samples of first and 

 second wrappers from each plot were cased down and fer- 

 mented by a tobacco dealer. The next fall each crop, 

 labelled so as to distinguish it, without giving the expert an 

 idea from which plot it came, was very carefully examined 

 and tested by a dealer in leaf tobacco ; its burn, ash, color, 

 texture, yield, size, vein and stem were all noted ; and, 

 lastly, its grade or rank, as compared with the others. The 

 best was marked 1st, the poorest 29th, and the intermediate 

 ones in their order of value. During these five years, of 

 course, we got wet seasons and dry, large yields and small, 

 very good tobacco and very poor tobacco. 



Of course, the comparative value of the crop from a given 

 plot was not exactly the same in all years. For instance, 

 one lot of leaf was graded 14th in one year, the next year 

 2d, the next 1st, the next 7th. If, now, we average these 

 numbers, we get an expression of the relative average quality 

 for the whole period of experiment. 



The work was done jointly by this experiment company 

 and the Connecticut station at New Haven, the station 

 bearing most of the expense ; it was done as carefully and 

 with as much skill in the field and in the laboratory as was 

 possible, so that any defects in it are inherent in the nature 

 of the thing itself, and not in the way of doing it. 



Now, what, in brief, is the showing of these experiments 

 on the effect of different fertilizers on the quality and quan- 

 tity of the crop? 



A word as to yield. The average yield of all the plots, good 

 and poor, was 1,685 pounds sorted leaf per acre, ranging 

 from 1,568 in one year to 1,876 pounds in another. The 

 largest yield from any one plot was at the rate of 2,280 

 pounds. On the average of five years, from all plots, good 

 and bad, we got 60.7 per cent of wrappers, ranging from 

 47.2 to 66.6. On some plots we got as high as 78 per cent. 

 This much is to show that, one season with another, we 

 had about average success. 



1. As to the effect of quantity of fertilizer nitrogen on 

 the crop, 3,000 pounds of cotton-seed meal per acre (210 

 pounds nitrogen) gave a larger crop and a better quality of 

 leaf for five years than either 1,500 (105 pounds nitrogen) 



