RESULTS OF SEED TESTS 



Made July 1, 1920, to July 1, 1921 



The number of samples of seed tested during the year 

 was 609. Of this number 111 were private tests on sam- 

 ples submitted by individuals. Farmers who plant seed 

 of their own raising find the State Laboratories a conven- 

 ient place to learn its germinating qualities. Dealers who 

 plan to sell local seed or seed which has been held over a 

 year request tests for germination and purity for the pur- 

 pose of tagging such seed to comply with the State Law. 



The other 498 tests represent the official samples col- 

 lected by the State Department of Agriculture to check 

 up on the printed guarantees attached by the wholesalers 

 and thus protect the purchaser against possible misstate- 

 ments in labelling. All this inspection work is done with 

 a view of protecting not only the buyer but the legitimate 

 dealer against unscrupulous persons who might endeavor 

 to market inferior seed by means of unwarranted guar- 

 antees. There is little question that the inspection work 

 is making it possible for the intelligent farmer to elim- 

 inate a large percentage of the guesswork in buying seeds 

 if he will but read the guarantees and profit thereby. 



In presenting herewith the results of findings for the 

 498 official samples, the customary allowances for varia- 

 tion are made. A difference of 5 in germination for all 

 seeds guaranteed 90% and above is considered satisfactory 

 and a variation of 10 on guarantees below 90^? . In re- 

 lation to purity the so-called "double tolerance table" is 

 used. This table is adapted from a formula stated 

 ^~ "^^^r~^^ in which "T" stands for the varia- 

 tion allowed and "P" for the percentage of purity found. 

 Those interested should note that this formula is used in 

 conformity with the ideas of the Association of Official 

 Seed Analysts of North America. 



Samples of seed that test sufficiently high in germi- 

 nation or purity respectively to come within the variation 

 suggested are marked "Satisfactory," samples that test 

 higher than the guarantee are marked "Above," and sam- 



