The Sources of the Nitrogen of o^ir Leguminous Crops. 671 
The pots used were made of glazed earthenware. They had 
j a hole half an inch in diameter at the bottom for drainage ; and 
another at the side, near the bottom, for aeration, into which a 
glass tube bent upwards, and lightly closed with cotton-wool to 
exclude insects, was fixed. The pots rested on slips of thick 
sheet glass, placed in basins of the same glazed earthenware as 
the pots. 
The plant-ashes used as mineral nutriment were prepared 
i by suspension in distilled water, adding sulphuric acid to the 
point of neutralisation, evaporating to dryness, and gently re- 
igniting. 
The drain hole at the bottom of each pot was loosely covered 
with a piece of thick glass, and 1 lb. of broken, washed, and 
dried flint was then put in. The pots held from 7 to 9 lb. of 
the yellow sand, from 6 to 7 lb. of the lupin sand, and about 
14 lb. of the garden soil. 
The soil-extracts supposed to supply the organisms were 
made by shaking, in a stoppered bottle, one part of the garden 
soil or lupin sand, with five parts of distilled water, and, after 
subsidence, pouring off the turbid liquid, which was then passed 
through platinum gauze to separate any floating matter. 
Determinations of nitric nitrogen by Schlcesing's method, and 
of total nitrogen by copper oxide, showed that the 25 c.c. of the 
garden-soil extract used for microbe-seeding contained little 
more than three-quarters of a milligram, and the 25 c.c. of the 
hipin-sand extract little more than a quarter of a milligram, of 
nitrogen — quantities which are quite immaterial considered as a 
supply of combined nitrogen. 
The seeds were selected for sowing and for analysis by 
weighing three or four lots of a hundred each to ascertain the aver- 
age weight per seed, and then a number of single seeds weigh- 
ing within five milligi-ams of the mean weight were taken. 
It was intended to commence the experiments early in the 
I summer ; but owing to the pressure of other work the necessary 
I preparations were not completed until early in August. All the 
i seeds were sown on August 6, three accurately weighed seeds 
being put into each pot. 
From the first the peas germinated and grew well in each 
of the four pots ; but in each of the four of blue lupins, and in 
each of the four of yellow lupins, one or moi'e seeds failed, and 
had to be replaced ; and in some cases these also failed. It is 
admittedly very difficult to secure healthy growth with lupins 
in pots. One essential condition seems to be that the soil must 
be kept open and porous ; and it is also important that the 
mineral matter added to the soil should be quite neutral. 
