New York Agricultural Experiment Station. 167 



been sent to the Station by correspondents during the past year, 

 and also to discuss the seed-trade couditions in this State as re- 

 vealed by the samples submitted for analysis. The results of these 

 tests do not have the range and accuracy of those secured by the 

 official inspection; however, the results no doubt reflect very 

 accurately the condition of the seed sold upon the open market, 

 since the samples were received from all parts of the State and 

 from many of the larger seed companies. Such seed examination 

 is done free of charge to residents of the State if samples are 

 received not too frequently from one individual. During the 

 past year 1,140 samples of seed were examined and reported upon. 

 Samples received prior to June 24 were examined by Miss 

 Minerva Collins (now Mrs. Richard Wellington), Assistant 

 Botanist, who resigned June 30, 1912. 



purity test. 



The value of a purity test depends almost entirely upon the 

 manner in which the sample submitted for analysis is taken, 

 and whether it represents to the fullest extent the bulk of seed 

 from which it was drawn. A seed sample to be of value for ex- 

 amination should be a composite sample taken from the entire 

 bulk of seed, whether in a bin or in bags, and should weigh at 

 leant 2 ounce-Sf 



The number of samples received during the past few years 

 has reached such proportions that it has become impossible, dur- 

 ing the time available for such work, to give a detailed quantita- 

 tive report upon any one sample. However, in all cases a practi- 

 cal report relating to the presence of noxious weeds, adulterants, 

 and general appearance of the sample is given. Such general 

 reports are often supplemented by a note at the bottom of the 

 report sheet in cases where it seems warranted. 



Many of the samples sent for examination were entirely too 

 small for a dependable analysis. Some small samples, which 

 showed the seed to be apparently pure, might have revealed the 

 presence of noxious weed seeds if a larger sample had been sent 



