332 H. W. Turpin 
by dividing the weekly carbon-dioxide percentage in the cropped soil after 
the carbon dioxide in the bare soil had been subtracted, by the weekly loss 
of water in pounds. The figures in column H, however, were obtained 
by dividing the weekly carben-dioxide percentage in the cropped soil by 
the weekly loss of water without first subtracting the carbon dioxide in 
the bare soil from that in the cropped soil. 
The relationship between the carbon dioxide produced in the cropped 
soil (from which has been subtracted the carbon dioxide in the bare soil), 
and the water transpired by the crop, is shown graphically in figures 
48 and 49. There seems to be a relationship between the amount of 
water transpired and the carbon dioxide produced by plants, as is indicated 
Per cent ; ; Pounds 
of COs of water 
OEP I OL75 O17 CL —— 
Carbor qoxwge 
OCCOTCO COCO OHHH HBR BR RRR Db tt ty ty Pt tO 
BNW RORDAD OOH NHN WHEMAIDWDOOH NWR OID 
- 5 
== 
May June July 
Fig. 48. RELATION BETWEEN WATER TRANSPIRED AND CARBON DIOXIDE PRODUCED 
BY AN OAT CROP FOR THE TEN WEEKS DURING WHICH ITS GROWTH WAS MOST VIG- 
orous, 1917 
by tables 3 and 4 and by figures 48 and 49. The illustrations show that 
the curves for the water transpired each week, and for the carbon dioxide 
obtained by subtracting the carbon dioxide in the bare soil from that 
in the cropped soil, follow each other closely. The data given in the 
