DEPARTMENT OF BOTANICAL RESEARCH. 77 



While the amount of phosphorus in the tufa is nearly double what 

 should have been present, based on the 1907 determination, it must 

 be remembered that the determination of so small an amount would be 

 very uncertain under the circumstances. We may consider, therefore, 

 that the phosphorus in the tufas represents practically all of that ele- 

 ment lost from the water. On the contrary, the potassium found in 

 the tufa represents only about 3 per cent of that lost from the water. 

 With regard to the fate of the lost potassium we are forced to conclude 

 that while a small part was fixed in the tufas, the greater part has been 

 removed by the adsorptive action of the clay and silt sediments, 

 probably in the manner discussed by Watson.-^ 



The Vegetation of a Desert Valley, hy Forrest Shreve. 



In 1918 work was begun on the relation of environmental conditions 

 to the distribution of vegetation in the Avra Valley, a poorly drained 

 basin, or semi-bolson, lying southwest of Tucson. The work has been 

 extended during the past year to embrace not only a characteristic sec- 

 tion of this valley, but also a group of low volcanic hills and their out- 

 wash slopes, a portion of the flood-plain of the Santa Cruz River, and 

 the lower ends of two extensive bajadas or ''mesas." 



The charting of the vegetation has now been completed over an area 

 of 240 square miles, in which there are relatively small differences of ele- 

 vation and only minor and inconstant differences in the climatic condi- 

 tions. The fact that only a very small number of species of perennial 

 plants play a dominant part m the vegetation serves to emphasize the 

 marked differences of plant covering that characterize this small area. 

 The object of the investigation is to determine to what extent, and in 

 what manner, these differences result from the character of the soil, 

 and to what extent it may be necessary to seek other causal conditions. 



A map has been nearly completed sho"\\'ing, for this area, the five 

 physiographic types of surface into which it is naturally divisible; 

 (a) hills, with rock in place; (6) coarse detrital slopes; (c) bajadas or 

 outwash slopes of gentle gradient; (d) playas or poorly drained areas 

 subject to inundation; (e) flood-plains. On the basis of these physio- 

 graphic regions a series of mechanical soil analyses has been begun in 

 order to determine the extent to wliich particular soil types may be 

 characteristic of each of them. A general uniformity of soil composi- 

 tion has been found throughout each of the areas except the coarse det- 

 rital slopes and the bajadas. In the former there is a great variation in 

 the percentage of very coarse material. In the latter there is a striking 

 uniformity throughout the bajada of the Sierrita Mountains, with a 

 "wdde variation in the bajada derived from the Rincon and Santa Rita 

 Mountains. An mvestigation of the principal physical constants of 

 these soils, so far as they are of significance to plants, is now in progress. 



^Abstraction of potassiiun by sedimentation (thesis), University of Virginia (1913). 



