122 



fegrioulture. Under present conditions, farmers aiid gardeners are fre- 

 quently disappointed in finding that their crops, as they approach maturity, 

 are not what they expected from the seeds purchased, and frequently 

 suffer much inconvenience and even severe losses. I suppose in Europe 

 and in America similar results are not infrequent occurrences when seeds 

 other than those of standajd brands are purchased. 



In Northern Japan and in Korea sugar-beet cultivation has been 

 established, and the areas devoted to this crop are constantly being in- 

 creased. To supply the demands of the sugar-beet planters, large quan- 

 tities of seeds must be imported from foreign countries. It frequently 

 happens that farmers receive seeds of the field beet rather than seeds of 

 the sugar beet. The seeds of both varieties are very similar in appearance, 

 and, naturally, the average farmer is unable to distinguish between them. 



Many tests made by me show that frequently the percentage of ger- 

 mination of the seeds is very low. The chief difficulty is, however, that 

 the seeds on sale are often not true to name. Mixed strains are often sold 

 as pure strains. The loss to the farmer may be very great, if, for example, 

 he finds that, as his crop approaches maturity, his beets are of the field 

 variety rather than of the sugar variety. It thus becomes as important 

 to determine the correctness of the varietal name as it does to determine 

 the percentage of germination, purity, etc. 



In agriculture, horticulture and general gardening, the correct identifi- 

 cation of the seeds to the genus and the species miost be supplemented by 

 further accurate identifications as to the variety, form, or sort, because, even 

 though seeds may be correctly identified as to the genus and species, they 

 may have no value as seeds for agricultural purposes when the identification 

 as to the particular variety or form is incorrect. 



It is, of course, a very difficult and exacting matter to determine 

 purity of sort and correctness of varietal identification from seeds alone — ■ 

 at least, in very numerous cases. As a pre-requisite we must have a most 

 thorough knowledge of the seed characters of all standard varieties and 

 forms of every species of agricultural or horticultural importance. For 

 instance, in Japan among the cultivated brassicas, variously classified as 

 Brassica campestris, B. japonica, B. nigra, B. juncea and B. oleracea, about 

 fifty varieties or forms are commonly cultivated. The seeds of all these 

 varieties closely resemble each other, and it is, therefore, a very difficult 

 matter to distinguish them properly. We also find, in cultivation in 

 Japan, about thirty garden varieties of the common radish, Bapfianus 

 sativua. As with the cultivated brassicas, while the growing plants of the 

 different varieties are readily distinguishable, it is exceedingly difficult 

 to distinguish these radish varieties from the seeds alone, or, I might say, 

 for the ordinary farmer or untrained observer it is practically impossible 

 to distinguish them. Farmers and gardeners are therefore often puzzled 

 by incorrectly -named seeds and by mixed seeds which are not infrequently 

 offered for sale by unscrupulous merchants. The matter of the correct 

 identification and certification of garden varieties of seeds thus assumes 

 an important position, and we are forced to an intensive macroscopic and 

 microscopic examination of the numerous varieties of seeds in order to 

 determine just what the varietal differences, as presented by the seeds, 

 may be. 



In general practice in Europe and in America seed testing is carried 

 on chiefly with a view to determining the percentage of germination and 

 the relative number of weed seeds present, and determination as to the 

 correctness of identification of the variety or form is not usually attempted. 

 I believe, however, that the latter should form an important part of seed 

 testing. 



Since 1908, I have carried on an intensive investigation of the Japanese 

 agricultural seeds which are more commonly offered for sale with a view 

 to correlating the seed characters of the numerous species, varieties, and 

 forms, with the distinctive specific, varietal, and form characters of the 

 plants themselves. The results of my investigations are in part recorded 

 in the " Berichte des Ohara-Instituts fiir landwirtsohaftliche Porschungen." 



