55 



«' i» %) i' B' i' ^' ^^^^^^ ' these eight deplhs cuiiotituted 



a series. 



The method of procedure adopted in preparing the pois for 

 ^;(H-mination was as follows : Each pot was first filled to the 

 required depth — marked on the pot — with a definite aUiOunt of 

 soil in accordance with the volume of the pot below the depth 

 mark. After the soil had been carefully levelled to the depth 

 mark 100 seeds were sown at 1 cm. apart. The seeds were 

 covered to the appropriate depth by filling the pot with a calcu- 

 lated weight of soil, which was then compressed and levelled to 

 the *' surface mark." By this method all the pots received 

 equal weights of soil, which occupied equal volumes and which 

 was compressed to the same degree of compactness, while the 

 seeds were buried to the exact depths required. The pots 

 received the same initial amount of water, and were afterwards 

 watered every two or three days as required. Each series was 

 jillowed to germinate at room temperature and was treated alike 

 in all respects. The number of surface seedlings was counted 

 every seven days. For the purpose of these experiments the 

 term " surface seedhngs " is applied, in the case of surface 

 sowings, to seedhngs with radicles properly fixed in the soil, 

 and in the case of covered sowings only to seedhngs whicli had 

 reached the surface. 



Box Cultures. — During May, June and July, 1920, a similar 

 experiment was carried out in boxes in the open, but protected 

 from birds by means of a wire cage. Seeds wei'o sown on (he 

 surface and at eight depths from J- in. to 3 in., 500 seeds being 

 sown in each box in rows 1 in. apart and ^ in. between the 

 seeds in the rows. The ordinary soil, which had been freed of 

 stones, was used for this purpose. The method of procedure 

 adopted in preparing the boxes for gennination was similar to tl\ » 

 method described under pot cultures; the surface seedling'^ were 

 counted every 14 days. 



Bed Cultures. — This experiment was carried out on duplicated 

 beds 42 in. by 18 in. on one of the Station's experimental 

 grounds. 500 seeds were sown in each bed in rows 2 in. 

 apart and J in. between the seeds in the rov>s. The different 

 depths were gauged by means of a graduated wooden fi*ame of 

 the same dimensions as the beds. The frame was ])ress; d into 

 the soil until the required depth was reached. The soil was then 

 removed and the floor of the frame carefully levelled. The seeds 

 were covered by replacing the removed soil, which was pr( ssed 

 down to the original surface mark. 



