have a height of approximately 1 inch, at which time the surface of 

 the basket is sealed that is, it is covered with a piece of paper hav- 

 ing a small opening in the center sufficiently large to permit the 

 plants to pass through. The paper is dipped in melted paraffin just 

 before placing it over the soil, and then a small amount of paraffin 

 is run around the outer edge of the paper, thus forming contact with 

 the side of the basket. In this way all evaporation from the soil is 

 prevented, excepting the minute amount which may pass through 

 the small opening immediately around the plants. The loss from this 

 source is so slight in comparison to that which is transpired by the 

 plants that no account is taken of it, but even if the loss were consid- 

 erable it should be practically the same from all baskets. The weight 

 of the basket is taken immediately before sealing and immediately 

 afterwards in order to ascertain the weight which has been added to 

 it in the process of sealing. This increase is now added to the orig- 

 inal weight of the basket, and the result is what is known as the 

 u optimum weight," or that weight at which the contents of the 

 basket contain the most favorable amount of moisture for the growth 

 of plants. 



During the growing of the plants, which usually continues from 

 eighteen to twenty-one days from the date of sealing, the baskets 

 are weighed at intervals of two or three days and watered with 

 distilled water in order to retain a favorable moisture content 

 for plant growth. By this method the loss of water or the amount 

 transpired by the plants is ascertained periodically, and at the end 

 of the experiment the total amount of water given off through the 

 plants of each basket is obtained for comparison with the growth 

 and green weight of the plants, which is ascertained by cutting and 

 weighing the plants at the time the experiment is concluded. All 

 conditions of the experiment are so carefully controlled that the 

 average result of five baskets rarely differs more than 5 per cent 

 from the average result of any other five baskets that have been 

 treated throughout in precisely the same manner. Differences which 

 occur beyond this amount may therefore safely be attributed to the 

 different manurial treatments which have been given. This method 

 has several advantages over the growing of plants in open and porous 

 pots. The method of coating the baskets with paraffin prevents any 

 accumulation of roots between the soil and receptacle, a trouble which 

 is so common in pot experiments. The complete sealing up of the 

 soil also enables the experimenter to determine the amount of water 

 which the plant has actually used and transpired in its process of 

 growth, and this, together with the small size of the baskets, enables 

 the moisture content and its fluctuations to be carefully controlled. 

 When the plants have attained considerable size, the draft made 

 upon the soil moisture is very great, and on warm days with bright 



