52 _ BOTANICAL GAZETTE _ [SEPTEMBER 
seeds were fully imbibed and most of them had sprouted. The 
clover was sown January 22, and on February 6, when the plants 
were 4 inches high, they were transplanted to the sand. A pure 
culture inoculum obtained from the Department of Bacteriology 
of the University of Wisconsin was added to the sand containing 
the clover on February 13. 
The nutrient solution used was the same as that used by Kraus 
and KRAYBILL (13), except that magnesium chloride was sub- 
stituted for magnesium sulphate. Perhaps a solution better 
suited to corn and clover might have been found, but it gave good 
growth in both cases. It was made up as follows: 
SOLUTION A SoOLuTION B 
Per cent Per cent 
Magnesium chloride.............. a Galen Witte. ie 
Dibasic potassium chloride........ 2 
Potassiuit nitrate. 0 2 
Equal parts of A and B were diluted 1 to 70 with water and 
then mixed. The solution was applied in this strength to the 
corn. The solution applied to the clover was just half this strength. 
Five hundred cc. of these solutions were added on an average of 
once a week to the corn and the clover. While the plants were 
small and the light poor, not so much was applied, but later the 
supply was increased. Both the corn and clover grew well, but 
no marked differences in color or size of plants developed in either 
In fact, in the case of the corn, that in the control series was as 
green as the corn of any of the other series. It was not deemed 
the right kind of material for studying the effect of sulphur on 
chlorophyll development, and no chemical analyses were made. 
Since no marked differences in color or growth due to sulphur 
deficiency had developed in the clover of the different series, it 
was decided to modify the experiment somewhat. It was thought 
the nitrate supply might be too high. On April 27 each series 
was divided into two parts. The nitrates were kept up in one- 
half of the pots and discontinued entirely in the other half, but 
no marked differences in color had developed at the time the 
experiment was stopped, on May 27. At this time the clover of 
the control series was somewhat paler than the clover of the high 
