1890.] On the Question of the Fixation of Free Nitrogen. 103 



the prepared quartz sand without soil-extract ; No. 2 with lupin-soil- 

 extract added ; No. 3 was duplicate of No. 2 ; and No. 4 was with the 

 lupin soil itself, to which 01 per cent, of the plant-ash was added. 



The soil-extracts were in all cases added on July 9, before the 

 sowing of the seed ; 25 c.c. in the case of the peas, vetches, and clover, 

 and 50 c.c. in that of the lupins and lucerne. 



The seeds, carefully selected and weighed as in 1888, were sown on 

 July 10, that is, about four weeks earlier than in the previous year, 

 but still not so early as was desirable. In the case of the clover, ten 

 seeds were sown in each pot; in that of the blue and yellow lupins 

 three, and in that of the peas, vetches, and lucerne, only two seeds 

 were put in each pot. 



In all four pots, the peas germinated and grew well from the begin- 

 ning. In the No. 1 pot of vetches, one seed failed and had to be 

 replaced. Several of the red clover seeds failed, and eventually four 

 plants only were left in each pot. As in 1888, most of the blue 

 lupins failed ; and eventually only one plant in only one of the four 

 pots, remained. Some of both the yellow lupins and the lucerne also 

 failed; but, as will be seen further on, eventually two good plants 

 of each remained in each pot. 



No analytical details relating to the experiments of 1889 are yet 

 available ; but the notes on growth, and the photographs of the 

 plants and of their roots convey a clear idea of the importance and 

 significance of the results obtained. 



The peas were taken up on October 23 and 24. Photographs of 

 the four pots of plants were taken on August 3, August 20, Septem- 

 ber 27, and October 22, that is the day before taking up ; and an en- 

 largement of the last taken is exhibited. It is seen that, unlike the 

 result obtained in pot 1 in 1888 with the impure and non-sterilised 

 sand, the plants in the purer and sterilised quartz sand, show 

 extremely limited growth. Before the end of Jnly, the plants in 

 both pots 2 and 3, with soil-extract, began to show enhanced growth 

 compared with that in pot 1, without soil-extract seeding; and 

 eventually, whilst the plants in pot 1 were only 8| and 8J inches in 

 height, those in pot 2 with soil-extract were 14 and 50| inches ; and 

 those in pot 3, also with soil-extract, were 52 \ and 50^ inches high. 

 In pot 4, with the garden soil, the plants showed even somewhat less 

 extended growth than those in pots 2 and 3 with the soil-extract 

 only. But the plants in pot 4 were more vigorous, and whilst they 

 flowered and seeded, neither of those in either pot 2 or 3 did so ; but 

 continued to vegetate, the upper parts apparently at the expense of 

 the lower. 



The root development should be briefly noticed. In pot 1, without 

 soil-extract, it was altogether much less than in either pot 2 or pot 3 

 with soil-extract, or than in pot 4 with garden soil. Enlarged photo- 



