144 The Principles of Vegetable -Gardening 



sarily, the latter tests are more difficult of determina- 

 tion, as they must be made from the product of the 

 plants, often requiring special and expert training on 

 the part of the investigator. They have apparently 

 not received the attention they deserve, largely from the 

 prevalent opinion that such matters lie beyond the con- 

 trol or check afforded by the tests of impartial investiga- 

 tors, an opinion no doubt strengthened by the so-called 

 contract printed on seed -packets to the effect that 

 the seller assumes no responsibility for the contents of 

 the packet. The seed dealer certainly cannot be held 

 responsible for failures which may be fairly associated 

 with conditions of weather, soil or method of sowing ; 

 but the warranty clause could not shield him if he 

 were to be negligent or remiss, or if he failed to exercise 

 reasonable caution in the care and selection of his 

 stock. 



Testing samples to determine the foreign matter or 

 the presence of seeds of other species is performed by 

 carefully examining small lots of seeds under a lens. 

 The operator should have at hand for comparison 

 reliable samples of the seeds of weeds and other plants 

 likely to occur in any samples.* 



Tests for purity of the sample have been carefully made in 

 Germany, extending over many years, especially at the famous 

 "seed control" station at Tharandt, in Saxony, organized under 

 the direction of Nobbe. This station, founded in 1869, was the. 

 first of its kind. The percentages of foreign matter found in 

 samples, by Nobbe, vary from nothing to over 80 per cent. The 



*Sainples of weed seeds, put np in bottles which are mounted in a serviceable 

 case, have been prepared by Dr. B. D. Halsted, Rutgers College, New Brunswick, 

 New Jersey. 



