816 EXPERIMENT STATION RECORD. 



subsequent sale. So iiuieh depends upon the inspection of the grow- 

 ing grain and harvested product and the certificate furnished by the 

 breeding station that Swedish farmers evidently prefer to pay the 

 extra price charged for the certified seed. 



Throughout both Sweden and Finland the Svalof varieties of grain 

 are spoken of in the highest terms. As far north as the sixtj^-sixth 

 parallel they vield better returns than native sorts whenever the season 

 is sufficiently long. In short seasons, when the frosts come unusually 

 early, the southern varieties do not ripen. It could not be expected 

 that sorts from southern Sweden would prove well adapted in this 

 respect to the northernmost localities. 



The seed compan}^ carries in stock a limited number of standard 

 seed varieties that have not been lired at the Svalof station. These are, 

 however, all su))jected to inspection by the experts of the station, and 

 ever}^ sack of seed sold by the company bears the stamp of the latter 

 and contains a certificate as to its purity. All pedigreed grains which 

 are purchased b}- the companv from its correspondents must, before 

 purchase, be inspected in the field and officially recommended by an 

 expert of the breeding station. 



In a brochure ' for the guidance of a committee of the German Agri- 

 cultural Society, published on the ocrcasion of its visit to Svalof in 

 1898, Dr. Nilsson describes in more detail than the limits of this arti- 

 cle will allow the organization and scientific principles of the insti- 

 tution. This brochure is worthy the careful perusal of every American 

 plant breeder. 



The breeding station at Svalof is housed in a modern laboratory of 

 attractive exterior and convenientl}^ arranged interior, and is sur- 

 rounded hj a small neatly kept park. (PI. II, fig. 1.) The experi- 

 mental plats, of which there are more than 2,000, are scattered among 

 the larger fields of pedigreed grain and cover in all about 25 acres. 



Special stress is laid by Dr. Nilsson upon the fact that all of his 

 experimental plats are under as nearly as possil)le natural conditions, 

 and to insure still further accurac}^, each variety is tested on not less 

 than three diflerent plats. 



The work of the station naturally runs along two lines, i. e. , the 

 search for mother plants for starting new varieties, and the elimination 

 of all but the very best of the varieties started. 



All selection of mother plants is based upon (1) the general (qualities 

 of the plant itself, (2) its abilit}^ to produce plants of high average 

 quality, and (3) its ability to produce plants of nearly absolute uni- 

 formity in botanical characters. 



The method of work is practically the same for all crops handled, 



^N. H. Nilsson, Einige kiirze Notizen iiber die schwedische Pflanzen-Veredlung 

 zti Svalof. Malmo, 1898. 



