240 Report of the Department of Chemistry of the 



however, of some of the more recent theories that have been pro- 

 pounded with relation to the nutrition of plants, it is felt that 

 these additional data should be made public. 



Plan of the experiments. — The forms of phosphoric acid used 

 in these experiments included all of those used at Maine with the 

 addition of Thomas slag and bone meal. 



This makes the list as follows: 



1. Acid phosphate (dissolved Florida rock). 



2. Finely ground Florida rock (floats). 



3. Thomas slag. 



4. Dehydrated Redonda phosphate. 



5. Bone meal. 



The method followed for comparing these materials has been 

 to apply in available form to all the experimental boxes equal 

 amounts of the needed constituents of plant food other than 

 phosphoric acid, this ingredient being added in the combinations 

 previously indicated. The form of phosphoric acid being the 

 only variable factor, any differences in crop production from the 

 several sets of boxes may reasonably be attributed to the com- 

 parative availability of the phosphatic materials used. This is 

 the most rational explanation, at least. 



The " soil "' in which the plants were grown in the several 

 experiments was as follows : 



1896-7. Natural sandy soil, supposedly poor, from pine plains 

 near the Station, 110 pounds per can. 



1898-9. Quartz sand, from Corning Glass Works, containing 

 99.5 to 99.7 per ct. Si0 2 , (which was bought, undoubtedly, from 

 the Berkshire Glass Sand Company, Cheshire, Mass.,) 46 pounds 

 per box, with 12 pounds of coarse material for drainage. 



1899-1900. In part with some crops the same sand that was 

 used in 1898-9 from which the roots of the 1898-9 crops had 

 been removed, and with some crops entirely new sand of the same 

 kind and in the same quantity. 



The cans and boxes were located in the Station forcing houses, 

 the water being taken from the ordinary city supply; viz., a 

 combination of lake and spring water. 



