312 CARNEGIE INSTITUTION OF WASHINGTON. 



Growth-rate and transpiration-losses were obtained at the three last- 

 named stations by the use of batteries of 6 to 12 plants of each of the following 

 species in sealed and aerated containers of appropriate size to permit fairly 

 normal root development. Helianthus annuus, Acer negundo, Avena sativa, 

 and Elymus canadensis were grown, when two or more weeks old, for periods 

 of 14 days at each of the stations. The results are not only in fair agreement 

 with factors as measured by instruments, but also correlate well with the 

 behavior of the seeded and transplanted vegetation. The results of the ex- 

 periments over a period of 5 years are now being assembled for publication. 



Water-Loss from Vegetation in its Normal Soil Relation, by J. E. Weaver and 



John W. Crist. 

 The following method has been devised to overcome the objection of 

 measuring transpiration from plants grown in soil of disturbed structure and 

 to make it possible to measure the water-loss from areas of undisturbed 

 stabilized vegetation. A heavy galvanized iron cyhnder with cross-sectional 

 area of 1 square foot was started into the soil in the opening left by a heavy 

 steel cutter of similar diameter which had been driven to a depth of 4 inches 

 and then withdrawn. A trench 2 feet wide, dug around a row of these 

 cyUnders, permitted columns of soil, over which the cylinders fit tightly 

 when forced downward, to be cut to a depth of 3 feet. Loose-fitting bottoms 

 with upturned edges were then placed over the lower ends of the cylinders after 

 the soil columns had been cut away. Containers and soil were then weighed, 

 placed in an appropriate trench in undisturbed grass-land, the bottoms 

 made water-tight by the addition of a wax seal, and the narrow trench refilled. 

 After a 15-day interval, during which the containers were covered in case of 

 rain but watered as necessary, they were reweighed and the water-losses 

 determined. Thus, it was found that Bulhilis dactyloides sod in the short- 

 grass plains (Burlington, Colorado) lost nearly a pound of water per day, 

 Agropyrum glaucum lost more than a pound, and an equal area covered with 

 a similar carpet of dead grasses lost only one-third of a pound, which was less 

 than that from bare soil. In mixed-prairie (Phillipsburg, Kansas) , Bouteloua 

 gracilis lost 15 pounds in a 15-day period and Andropogon furcatus lost 28 

 pounds. Losses from similar vegetational areas in the true-prairie (Lincoln, 

 Nebraska) were less. Crops of oats, millet, and alfalfa gave equal or greater 

 losses than the native vegetation at the several stations. 



Plant Production as a Measure of Environment, by J. E. Weaver. 



In connection with the studies in experimental vegetation (p. 311), an 

 attempt has been made to determine the precise relation between habitat 

 factors as measured by instruments and plant response as determined by 

 growth. At each of the stations in the three grassland associations, all of 

 the important aerial and edaphic ecological factors, such as rainfall, holard 

 and nutrient content of soil, humidity, evaporating power of air, soil and air 

 temperature, and wind movement, were measured and compared during 

 three growing seasons. The production of native grasses and the smaller 

 cereal crops was determined by means of a large number of cut quadrats. 

 The relative production of certain legumes and maize was also ascertained. 

 The yield of pure or mixed stands of grasses increased progressively as the 

 water-content of soil and inversely as the evaporating power of the air, i.e., 

 proceeding from the western stations eastward. 



