1902] RISE OF THE TRANSPIRATION STREAM 175 



recently it has been redemonstrated by Schwendener (1886 : 563), 

 Tschapohvitsch, and Strasburger (1891 : 6yy), The air content 

 of the heart is rather fixed, but that of the sap wood decidedly 

 variable. With only the same class of exceptions as to the pres- 

 ence of air at all, the air in the tracheae is under a lower tension 

 than that of the atmosphere outside. This lower tension is one 

 of the reasons for the presence of air in the gaseous form. The 

 water absorbed by roots is saturated with air under the atmos- 

 pheric pressure and at ground temperature. In the trunk the 

 water is usually warmed and a part of the pressure is removed, 

 both tending to cause the air to come out of solution. It has 

 not been so easy to tell why there is not more air in the wood. 

 But Noll's recent discovery that there is an active osmotic move- 

 ment of gas between the wood and the air outside throws an 

 altogether new light on this question. At least for vessels in or 

 near the leaves, Noll shows that this osmosis w^ill be very largely 

 outward. Before this discovery, the only known immediate 

 cause of the diminished tension of the tracheal air, aside from 

 the relatively inconsiderable influence of occasional cooling, 

 was the removal of water from the lumina to replace that evapo- 

 rated from the leaves. 



The less air the tracheae contain, the greater will be the 

 change in its tension with a given variation in the amount of 

 water present. Evidently the proportion of water present gives 

 an approximate, though not an exact measure of the tension of 

 the air. This proportion of water is in all cases least in summer^ 

 the maximum being in winter in gymnosperms, and in spring in 

 dicotyledons (Th. Hartig, 1868 ; Ge'leznow, 1876. R. Hartig. Bot. 

 Zeit. 1883, disputes this). Th. Hartig (/. c, 20) tested the water 

 content of tw^enty-six kinds of trees, four feet from the ground, 

 before sunrise, and at 2:00 p. M.,and found the loss of water vary- 

 ing from 2 to 38 per cent. G. Kraus' most accurate work on 

 this subject comes to the same conclusions, epitomized by Kraus 

 himself in Sachs' Pflanzcfiphy stole gie, second edition p. 601, as 

 follows: -Die Pflanzentheile (Stamme, Blatter, Friichte, Knol- 

 len, Knospen) . . . . sind am Tage kleiner, bei Nacht grosser. 



» y 



