A CHEMICAL STUDY OF THE ASPARAGUS PLANT. 267 



The material was next spread on a large sheet of paper in a cool place 

 until the surface was free of adhering moisture. Each individual crown 

 and its accompanying roots were then weighed and the weight noted 

 down for the subsequent calculations as the fresh or green weight from 

 the field. 



The first stage of preparation of the material for analysis was to pass 

 a sample, consisting of one crown and its corresponding roots, through 

 a hand-lever feed-cutter, by which they were cut to lengths of about 

 1 inch (2.5 centimeters). The sample was then placed in a large steam- 

 heated drying oven, where the temperature was about 55° C, and dried 

 until sufficiently brittle to be easily pulverized. 



In pulverizing asparagus roots for analysis certain properties of their 

 constituents made serious trouble. During the preparation of the first 

 lot of roots in 190S Mr. Smith found the dried material to be so hygro- 

 scopic that in damp weather it would quickly become sticky and gum 

 the mill. The friction of grinding was also apt to produce sufficient heat 

 to make the material sticky and hopelessly cement the grinding plates 

 together. By using a ball mill in dry weather he finally succeeded in 

 reducing the samples to powder. 



The TVTiter's procedure with the samples of 1910 was as follows: im- 

 mediately after removing the dried sample from the oven the material 

 was allowed to cool a short time in the air and then weighed. Directly 

 after weighing the sample was passed through a large drug mill, by which 

 it was reduced to a mixture of coarse fiber and fine powder, the fiber 

 coming from the outer walls of the roots and the powder from the interior 

 and the crown. The mixture was subsampled by two successive quar- 

 terings. 



The subsample was next sifted by means of a millimeter sieve, which 

 separated nearly all of the fine powder from the fibrous shreds. By this 

 step the hygroscopic, gummy constituent was largely eliminated from 

 subsequent milling and the coarse fiber was pulverized about as readily 

 as wheat bran, until it also passed through the millimeter sieve. The 

 entire material of the subsample was thoroughly mixed and preserved in 

 a tightly corked bottle for analysis. Care was taken to conduct all the 

 operations in a dry atmosphere. 



On June 23, 1911, at the end of the cutting season, a third lot of samples 

 was taken for the purpose of determining the amount of exhaustion which 

 the reserve material in the roots had undergone in producing the crop 

 recently harvested. This lot of roots was collected under the supervision 

 of Mr. C. W. Prescott, who was in charge of the Concord experiment 

 field. There was practically no top growth by which to judge the size 

 of a crown, and the roots were therefore necessarily chosen more at random 

 than in the previous cases. On arrival of the roots at the laboratory they 

 were treated in the manner described for the samples of 1910. 



The average fresh weight of forty-four roots gathered from eleven 

 different plots was found to be for each of two years, as follows: 1908, 



