311 



tiRed foi- the separation of tobacco seeds, in tlie experiments on tobacco 

 breeding performed at tlie Connecticut State Experiment Station. This 

 apparatus is described in the Yearbooli of the Department of Agriculture 

 for 1901. Slight modifications were found necessary to adapt it to this 

 type of seed. A number of seeds from both the heavy and light portions 

 were accurately weighed. The average of tlie light seeds was 0.0035 

 grams and of the heavy 0.0084, a ratio of 1 :2.40. The variation in the 

 I'.eavy seeds was from 0.0077 to 0.0091 grams and in the light from 0.0017 

 to 0.0046. The greater variation in the light seeds was found to be due 

 to the force of the air current employed in the separation. That a sep- 

 aration into liglit, medium and heavy can be made just as readily as into 

 light and heavy will be shown in connection with another form. 



A planting of three acres of burdock of the present year has been 

 chosen as a type upon which to calculate the increase in yield which 

 might have been obtained, by means of seed selection. This planting was 

 made upon a deep mellow loam and the total yield of 33,890 pounds is 

 rather unusual. Assuming that the seed supply used on this planting con- 

 sisted of light and heavy seeds in the determined proportion of 1 :2.0, the 

 total yield can be theoretically divided into two portions of G,77S and 

 27,112 pounds, respectively produced by the light and heavy seed. Had 

 the light seed used been equal in \\'eight to the heavy seed they would 

 luive produced twice as much as they theoretically did, which would have 

 been 13,550 pounds. This would make a total yield of 40,6GS pounds in- 

 stead of 33,890, an increase of 20 per cent. 



Even of greater importance, however, to the medical and pharma- 

 ceutical professions is the improvement of henbane and digitalis, repre- 

 senting as they do two valuable drugs of the I'uited States Pharmacopoeia. 

 F.\ their being more amenable to chemical and physiological methods of 

 standardization, the investigator is furnished with additional means of 

 following the progress of various methods of im[)rovement. OtFieial hen- 

 l>ane is supposed to consist of the dried leaves and flowering tops of 

 llyoscyamus niger L. collected from plants of the second year's growth 

 and yielding not less than 0.08 per cent, of mydriatic alkaloids. That the 

 above chemical and botanical conditions of this drug rarely obtain has 

 been clearly <lemonstrated. An average of a large number of samples 

 examined from August 25, 1908, to September 23, 1911, indicates that only 

 13 per cent, conformed to the requirements of the Ignited States Pharma- 



