Febeuaet 23, 1917] 



SCIENCE 



179 



more vigorous when the soil containing the 

 roots was artificially aerated. 



The other series of experiments was con- 

 ducted by one of us (Free) under the direction 

 of Professor B. E. Livingston in the labora- 

 tory of plant physiology of Johns Hopkins 

 University. A technique has been devised by 

 which plants of mature size can be sealed into 

 tin cans of about two liters capacity, the shoot 

 projecting through the seal into the open 

 atmosphere. The root system, with the soil in 

 which it grows, is inside the sealed space and 

 the soil atmosphere can be replaced at will by 

 an atmosphere of any desired composition. 

 Water is supplied to the plant by the Living- 

 ston auto-irrigator. The controlled atmos- 

 phere inside the can is kept automatically at 

 a pressure slightly (about 3 centimeters of 

 water) greater than the general atmospheric 

 pressure, changes of volume due to variable 

 temperatures being compensated. This assures 

 that any slight leakage will be outward and 

 without effect on the experiment. 



With this technique experiments have been 

 made on four species : Coleus hlumei, Eelio- 

 tropium peruvianum, Nerium oleander and 

 Salix sp. (probably nigra). With Coleus it is 

 found that even a very small decrease of 

 oxygen below that normal to the atmosphere 

 is injurious to the plant. Thus a plant, the 

 roots of which were supplied with gas consist- 

 ing of 75 per cent, air and 25 per cent, nitro- 

 gen, was injured within three days and killed 

 within 45 days. With lower oxygen content 

 in the soil atmosphere injury and death are 

 still more prompt. Heliotropium behaves sub- 

 stantially like Coleus, except that the period 

 between initial injury and death is shorter. 

 iNeriiun is much more resistant to oxygen 

 deprivation. With a soil atmosphere of pure 

 nitrogen, injury was first apparent in the 

 shoot after 26 days. An atmosphere of 50 per 

 cent, air and 50 per cent, nitrogen had pro- 

 duced no perceptible injury in 45 days when 

 the experiment was stopped. Dilution of the 

 soil atmosphere with carbon dioxide instead of 

 nitrogen appeared to have a like effect. No 

 evidence was observed of any specific toxic 



effect of carbon dioxide, though such an effect 

 is not excluded by the experimental results. 



The most interesting result was with Salix. 

 With this plant entire deprivation of oxygen 

 appeared to be without injurious effect. In an 

 experiment three times repeated the plant 

 grew normally with a soil atmosphere of pure 

 nitrogen, one of the experiments lasting for 

 ten weeks. Replacement of the nitrogen with 

 carbon dioxide and the use of various mix- 

 tures of carbon dioxide and air were also with- 

 out perceptible effect. It appears that this 

 species of Salix is quite independent of the 

 content of oxygen in the soil atmosphere. 

 That the respiration of the root cells can be 

 anaerobic is less certain, but is strongly sug- 

 gested by the data. 



The two series of experiments outlined are 

 sufficient to show that different species of 

 plants may differ markedly in their response to 

 variations in the composition of the soil 

 atmosphere, and hence to changes in soil 

 aeration. The effects of diminution of oxygen 

 are manifest and the results with Opuntia 

 indicate a direct and specific effect of carbon 

 dioxide in addition to the effect of the dilu- 

 tion of the oxygen. 



Though many details are lacking it is known 

 that the composition of the soil atmosphere is 

 neither the same as, nor so constant as, the 

 composition of the general atmosphere. The 

 presence of living matter in the soil, including 

 bacteria, fungi and protozoa, as well as the 

 roots of higher plants, tends to decrease the 

 oxygen of the soil atmosphere and to increase 

 its content of carbon dioxide. Doubtless there 

 are chemical reactions associated with the 

 decay of dead organic matter and which have 

 the same or similar results. This tendency to- 

 ward impoverishment in oxygen and enrich- 

 ment in carbon dioxide must be counteracted 

 by diffusion between the soil atmosphere and 

 the general atmosphere, assisted, no doubt, by 

 changes in barometric pressure and in tem- 

 perature of soil and air. The importance of 

 these assisting agencies is difficult to estimate 

 but the effect of diffusion alone has been shown 

 by Buckingham^ to be extremely slow. Doubt- 



iBuUetin 25, TJ. S. Bureau of Soils (1904). 



