15 



there was no volatilization of phosphoric acid at the temperature at 

 wlucli the igtiitions were made. The separation of lime from iron 

 and alumina was effected by precipitating the iron and aluminum 

 phosphates in acetic acid .solution without addition of ferric chlorid, 

 as tliis method possessed some advantages over the official method.' 

 The iron was determined volumetrically with potassium perman- 

 ganate. Aji aliquot was twice evaporated with sulphuric acid until 

 fimies of sidphuric acid appeared. The diluted solution was then 

 reduced mth iron-free zinc, filtered, and titrated with ^ or y^tj normal 

 potassium permanganate. Tests with potassium tliiocyanate showed 

 that between the final fdtration and titration with permanganate no 

 ferric iron was formed. 



For the calculation of the results the sum of the constituents found 

 in the acid solution of the ash was not taken as the amount of carbon- 

 fi-ee ash, but the total ash was determined absolutely and the percent- 

 ages of the various constituents calculated from tliis determination. 

 The percentages of the constituents in the ash are tlius dependent 

 upon the accuracy of the determination of the carbon-free ash, wliich 

 is probably subject to an error of about one part in a hundred. The 

 inorganic elements present in the dry matter of the plant were cal- 

 culated by multiplying tlie percentages in the ash by the percentage 

 of carbon-fi'ee ash and dividing by 100. In this process the percent- 

 age of ash eliminates itself, so the percentages of the constituents 

 found in the dry substance of the plant are more accurate, bcijig 

 independent of the determination of carbon-free ash. 



GROWTH AND COMPOSITION OF VARIOUS PLANTS ON THE CAL- 

 CAREOUS AND NONCALCAREOUS SOILS. 



The gi'owths of the various plants on the different soils and their 

 ash compositions are detailed in the following pages. In all the 

 tables Plat I refere to the check plat, containing no carbonate of 

 lime, and Plats II, III, and IV refer to the plats containing, respec- 

 tively, 5, 18, and .35 per cent of carbonate of lime. 



The results obtained with pineapples are reported in a former 

 bulletm of this station. - 



BUSH BEANS. 



Six crops of bush beans, variety Improved Golden Wax, were 

 grown at various seasons of the year. WMIe the growth made at 

 different seasons varied greatly, plantings in March and April gi%ing 

 the maximum growth, the relative growths made on the diS'erent 

 plats appeared to be unaffected by the time of planting. The first 

 crop of beans being grown to seed no data were secured on the weight 

 of the plants, as when the seed were thoroughly mature the plants 



' A report on this method is given in Porto Rico Sta. Rpt. 1912, p. 21. 

 'Porto Rico Sta. Bui. 11. 



