81 



rank growth of cane was in poor condition to withstand hot, drying 

 winds and baked soil. Many of the varieties which were planted on 

 hard soil were injured by dry weather. 



The tonnage or yield of cane was large, being equaled only by that 

 of 1889. The average quality of the juice was good, as is shown in the 

 general averages given to each variety. In some of the varieties the 

 quality of juice was not as good as in the more favorable season of 

 1889. The varieties which usually mature early matured later than in 

 any previous season. Their juice had low purities until late in the sea- 

 son. The early and the late varieties matured nearly at the same time. 

 These facts can be explained only as the eifect of drought upon the un- 

 usually rank growth of June and July, for the weather in 1890 was as 

 dry as in 1891, at the same season, and had not the same effect upon 

 the smaller canes of that season. 



The object of the work at this station is, first, to select the varieties 

 best suited to sugar manufacture, so that all other varieties may be dis- 

 carded; and, second, to improve the selected varieties by selection of 

 seed from the best canes. In order to select the best varieties it is 

 necessary to learn the practical value of each variety during its working 

 season. A single analysis of a variety is of little value. The mean of 

 many analyses of a single lot may mislead. In the work which has 

 been done this season an effort has been made to arrive at the value of 

 each variety by grouping together all the maximum analyses of all the 

 lots of each variety which were grown here this season, and also by 

 grouping together the analyses of each variety in the four years in 

 which the experimental work has been carried on at this station. In 

 endeavoring to express the value of the Collier variety, for instance, a- 

 single analysis of a sample of the cane may have shown 18 per cent of 

 sugar. The mean of all the analyses of a single lot of that variety may 

 have shown 17 per cent of sugar, but the mean, or the general average, 

 of two hundred and forty- two analyses of canes from sixty- five lots of 

 that variety, which were made in 1888, 1889, 1890, and 1891, shows 

 that its juice contains, on an average, 14 per cent of sugar during a 

 working period of twenty-five days. It may be called a 14 per cent 

 variety. As its yearly average has not varied 1 per cent above or below 

 that figure for four years, it may be assumed that it will continue to 

 have that value. From the results of two hundred and six analyses of 

 Colman cane in four years it may be said that it also is a 14 per cent 

 variety. The McLean variety, as the result of seventy-five analyses in 

 two years, may be hoped to prove to be a 15 per cent variety. Early 

 Orange may thus be considered a 12 per cent variety, and Early Amber 

 an 11 per cent variety. 



These figures may seem low, for the Amber sometimes reaches 14 per 

 cent, the Orange 15 per cent, the Collier, the McLean, and the Colman 

 sometimes reach 18 per cent; but general averages for a proper working- 

 period are always low, and it is upon general averages that manufac- 

 ture depends. 



21383 No. 34 6 



