16 



a summary is given of the results received from 19 stations in Europe, 

 4 in Ainerica and 1 in Japan.* 



Immediately after the Congress I wrote to these stations and drew 

 attention to those of their results which differed by more than a reasonable 

 latitude from the average figures of the results obtained at the large 

 stations, and which results agree generally within the latitudes fixed in the 

 Rules of the Danish State Seed Testing Station. 



After the Congress, a new series of samples was sent out, consisting of 

 24 seed samples, and a correspondence, similar to that mentioned above, 

 was carried on with the 43 stations (37 European, 4 American, 1 New 

 Zealand and 1 Japan), which sent in results. From Table 1 {see pp. 21-26) 

 it will be seen that these results agree, as a rule, within reasonable lati- 

 tudes, as far as a series of stations is concerned. Many of those stations 

 which arrived at similar results have rules for seed testing which differ in 

 many particulars ; some have very brief rules and others have no official 

 rules at all. On the other hand, one cannot help noticing that results 

 from stations which have somewhat detailed common rules for seed 

 testing are, in many cases, rather diSerent. 



The main points are that the lines of direction for seed testing are 

 similar and that the staffs are well-trained and experienced. It is, there- 

 fore, unfortunate when the seed testing work in a country is divided up 

 between many small stations, as, for instance, in Germany and Sweden, 

 where the seed testing stations are frequently offshoots of chemical insti- 

 tutions. A centraUsation of the work, with one, or a few, well equipped 

 seed testing stations, would undoubtedly be the best way to attain more 

 uniform results. 



A comprehensive correspondence has been carried on with the stations 

 partaking in the enquiry and also with others. I have endeavoured to 

 point out the main causes of the greatest discrepancies and have found that, 

 for instance, seeds attacked by larvae, or poorly developed seeds, shrivelled 

 and injured seeds in the leguminous species, " hard seeds," and especially 

 " broken seedlings," &c., have been subject to very different valuations. 

 The drawing of pure seed for the germination test, differences in tempera- 

 ture and moisture, and, possibly in one case only, conditions of Kght, 

 &c., are all factors that have occasioned discrepancies. It must, however, 

 be noticed that the samples in question were difficult to test; for 

 example, they contained many " doubtful " seeds, which have been 

 judged very differently. In one case the samples were not considered 

 satisfactory because of this ; but, in my opinion, it is best to send 

 samples of this kind for comparative analysis, as it is easy enough to 

 obtain corresponding results with samples which do not present any 

 difficulties in testing. Most of these and other questions will be 

 referred to in other papers read at this Conference, and an opportunity 

 will be given for discussing them. All seed testing stations have adopted 

 the so-called " Continental method," with the exception of the station 

 in Dublin, where the so-called " Irish method " is still used for grass 

 species. The Dublin statiop. has naade the comparative analysis according 

 to both methods. 



In November, 1923, new "Referee" seed samples (21 samples of 19 

 species) were sent out to 54 stations. The results obtained at 45 stations 

 (38 European, 6 American and 1 Japan) appear in Table 2 (sea pp. 27-32). 

 It is a pleasure to see that they agree, generally, better than was the case 

 with the two former series, although the last series of samples was the 

 most difficult to test. Great differences still exist, however, between 

 the results of some of the stations. 



" Other Crop Seeds " and Weed Seeds. 



In earlier comparative investigations the contents of " other crop seed " 

 axid weed seed have not been quoted in percentage by weight. Many 



* See pp. 76-83, " EUscussions at the International Seed Testing Conference 

 in Copenhagen, 6-10 June 1921," by K. Dorph-Petersen. 



