Dec., 1923] 
ALLISON AND SHIVE — GROWTH OF SOYBEANS 
555 
plants each, were placed both in solution and in sand. The nutrient solu¬ 
tion used for all cultures was that found by Shive (7) to be well adapted for 
the growth of soybeans. This solution was used at a total concentration 
of 1.75 atmospheres and contained the three salts KH 2 P 0 4 , Ca(N 0 3 ) 2 , 
and MgS 0 4 in the proportions of .00406 M, .0034 M, and .0225 M, re¬ 
spectively. As a source of iron, one milligram of the element in the form of 
ferrous sulphate was used per liter of solution for all cultures. 
The four treatments accorded the plants in both sand and solution 
cultures may be outlined as follows: 
Series I. Simple non-aerated cultures. Solutions renewed every three and one 
half days. 
Series II. Simple aerated cultures. Like I, but aerated by a continuous bubbling of 
air through the medium. 
Series III. Drip cultures. Solutions continuously renewed, but not aerated. 
Series IV. Aerated-drip cultures. Like III, but aerated by a continuous bubbling of 
air through the medium. 
For all solution cultures quart jars were used in the manner described 
by Shive (6) and adapted to the various treatments. For those cultures 
not receiving a constant drip the period between solution renewals adopted 
for both solution and sand cultures was three and one half days. 
For the sand cultures percolators were prepared by cementing together, 
after removing the bottoms from all except the smallest, a series of four 
ordinary cone-shaped clay pots, the largest of which was 23 cm. in the upper 
diameter and the smallest 7.5 cm. In cementing the pots together, a fairly 
thin mixture of Portland cement was used, the bottom of the larger being 
fitted into the top of the smaller consecutively, the bottom of the smallest 
pot forming the bottom of the cone-shaped vessel thus constructed. These 
percolators were 50 cm. in height and were capable of holding nine kilograms 
of dry sand. After air-drying they were heated in a hot-air sterilizer for 
the purpose of removing hygroscopic moisture preparatory to treating them 
both inside and outside, while still warm, with hot paraffin. This treatment 
was found to prevent the penetration of the walls by the solution. The 
opening through the bottom of each percolator was closed by a rubber 
stopper carrying a short glass tube through which the percolating solution 
could escape. The upper end of this tube was loosely closed with a small 
plug of glass wool. 
Each vessel when completed was filled with nine kilograms of clean 
white sand which had been washed with tap water followed by distilled 
water. All the sand cultures employed in the experiments, whether solu¬ 
tions were renewed continuously or only intermittently, were prepared in 
these culture vessels. 
For sand-culture work in which the nutrient solutions are continuously 
renewed by percolation through the sand, a deep vessel having a form some¬ 
what like those here described is desirable if not an actual necessity in order 
to prevent excessive moisture in the median and upper portions of the sand 
