DEPAETMENT REPOKTS. 107 



In 1877, I began by testing the seeds which were sold on commission by 

 grocerynien all over our State. They were found very poor and unreliable. 

 The results were published. At least three of the worst firms wont out of the 

 business soon after or ceased to sell seeds in our State. The seeds still sold at 

 our grocery stores in small packets are many of them unreliable, and if unre- 

 liable are worthless. I find the greatest fraud exists in grass seeds, many of 

 which are imported. Seeds of the commonest grasses, as timothy, orchard 

 grass, June grass, red top, are generally very good, but seeds of perennial rye 

 grass, the fescues, meadow foxtail, oat grass, crested dog's tail, sweet vernal, 

 and others are very low in vitality, almost without exception. In some lots 

 tested by me there were no seeds which would germinate. Grass seeds are not 

 known by many persons. At one time a good dealer sent empty chaS of Bur- 

 mud a grass, which often flowers, but seldom seeds; at another time he sent 

 samples of red top, or something like it, which had apparently been cooked. 

 The seeds were all rotten. Even the Agricultural Department at Washington 

 has sent out many very poor seeds. 



The great importance of testing seeds to ascertain their vitality, purity, 

 identity, freedom from foreign substances, has already been clearly demon- 

 strated by the efforts of many persons. This is true of seeds which are home 

 grown, and especially true of seeds which are obtained by purchase. In 

 America the subject can hardly yet be said to have been fairly started. Peo- 

 ple are just beginning to think of the subject, but have done very little. Once 

 decided that it is important to test seeds, the next step is, ''How shall they 

 be tested?" 



In making tests, the fact must not be lost siglit of that the finest looking 

 seeds, and those which show the greatest proportion of living seeds, may not 

 be the most desirable for the field or garden. Much depends on the variety, 

 the choicest strains of some kinds producing seeds of inferior appearance and 

 vitality. One of the objects of the following experiments is to consider some 

 of the points in regard to the best modes of finding the vitality of a few kinds 

 of seeds. 



In several countries of Europe this part of the subject has commanded much 

 attention, and as one of the results several kinds of apparatus have been 

 manufactured and put on the market. I have had the privilege of using 

 but one kind of this apparatus. That one is the plates for testing seeds man- 

 ufactured by E. H. Krelage & Son, Haarlem, Holland. The plates are made 

 of fine potter's chiy, and are fifteen centimeters in diameter, or nearly six 

 inches. These plates are of two sorts, one containing twenty-four numbered 

 pits or depressions ; the other, five large depressions. 



To get the highest per cent to germinate, we have concluded that particular 

 modes of testing must be employed for the seeds of certain species, gendera, or 

 families. In a number of cases where the seeds possessed a low vitality it has 

 been found by others that certain chemical substances increased the per cent 

 which would germinate. I have not employed any of these. 



To hit upon the best method in a certain case, as that of wheat, we need a 

 quantity of seed which has been well grown, carefully harvested, and is known 

 not to have been subjected to severe trials or long exposure to great extremes 

 of drought and moisture, heat and cold. This reliable seed wheat can be 

 tested by some of the best known methods, and if found to be of prime vitality, 

 it can be used side by side in the different methods employed for testing other 

 lots whose history is unknown. 



In these modes, attention must be given to all the details of moisture, air, 



