128 



and if the vitality of the seed is diSerent at different depths, experiments 

 were commenced in 1903, similar in principle to those mentioned above, 

 except that portions of 100 seeds of every species were biu'ied at depths^ 

 of 8, 20 and 30 cm. A portion from each depth was examined every spring 

 together with a sample stored under dry conditions at the seed testing 

 station. The main results are recorded in Table 4 {see p. 136). 



The rather considerable variations in the germinating capacity of the 

 same sample year by year are probably due to the comparatively small 

 nxmnber of seeds (100) annually tested. In some cases earthworms and other 

 animals had bored through the soil in the flower -pots ; the results obtained 

 are therefore not quite reliable. The conditions of germination were, 

 further, not quite equal in the various years on account of the different 

 atmospheric conditions. 



As might be expected, all the species examined retained their ger- 

 minating capacity most poorly at a depth of 8 cm., where the supply of 

 oxygen is largest and where the conditions of temperature and moisture are- 

 most variable. Seed of Daucus carota and Cirsium arvense was only buried 

 20 cm. deep. Besides the species mentioned in the table, seeds of Secale- 

 cereale, Avena sativa, Avena elatior, Lolium perenne ajid Agrostemma 

 githagowere investigated in the same way. Except a few seeds of Avena 

 elatior and Lolium perenne, all the seeds buried of these species died in 

 the first wintra-. -.^ --.-«,»-,™™.:,»«»-» 



The crop seeds, as a rule, retained their germinating capacity less in the- 

 soil than the weed' seeds. Even the oil-charged seed of Brassica campes- 

 tris rapif era, when lying in the soil, retained its germinating power much 

 less than the nearly related Sinapis arvensis. 



VI. — How inany weed seeds are contained in forage grain before and after 

 grinding, imported forage grain, chaffs, cleanings, etc. ? 



During 1907, 37 samples of cereal seed for forage purposes were tested 

 for content of weed seeds, before as well as after grinding. It became 

 apparent that the common supposition that the weed seeds are crushed 

 and made harmless by the grinding of the cereal seed, is by no means 

 correct. The samples tested contained on an average 16,400 weed seeds 

 per kg. before grinding, and after this 9,300 uninjured weed seeds per kg- 

 Only a little more than one -third of the weed seeds were thus crushed in 

 the mill. The samples contained 54 species of weed seeds altogether. As 

 might be expected, mainly the species with small and hard seeds were 

 found in the cereal seed after its treatment in the mill {see Table 5 on 

 p. 137). 



Other investigations with barley, which is imported into Denmark from 

 the countries along the Black Sea, are also mentioned, which show that 

 large quantities of weed seeds are frequently contained in this imported 

 forage grain ; in one case a cleaning of barley sold for forage purposes at 

 almost the same price as that of pure barley contained 55 • 8 per cent, weed 

 seeds, in another, 41 per cent. 



It is also mentioned in the Report that the amount of weed seed in 

 cleanings and chaff is so great that it is necessary, when using such 

 residual products, to take measures in order to prevent the spreading of 

 weed seeds contained in the said products on the field. 



VII. — How does weed seed that has passed through the digestive system of 

 the domestic animals germinate ? 



On this subject information is given in " Tidsskrift for Landbrugets 

 Planteavl," vol. 8, pp. 33-35, and vol. 12, pp. 51-53 (the reports of the 

 Danish State Seed Testing Station for 1899/1900 and 1903/04). 



In the first-mentioned report, O. Rostrup has given particulars of the 

 result of the examination of the manure of a cow which had been fed with 

 seed-bearing plants of ten different species. Of eight of these, stated 



