420 



THE AMERICAN NATURALIST [Vol. XL 



weight independent of the experiment, still the abrupt change 

 between Nos. 1 and 2, 2 and 3, 5 and 6, 8 and 9 shows very well the 

 increase in weight when cooled and a corresponding decrease when 



Similar results were obtained whether or not the wood was com- 

 pletely saturated, although all of Sach's experiments seem to have 

 been with material rather near the point of saturation. 



In the above experiment the weight of the twig dry was 22.8 

 gms. which subtracted each time gave the figures in the last column 

 as the amount of water in the twig at each weighing. It will be 

 seen that between 4° and 30° there was a loss of 1.6 gms. of water 

 from the 31.4 gms., which would equal a loss of 5.09 gms. for 

 €ach 100 gms. of water. Pure water, however, will expand only 

 1 gm. for every 100 gms. between the above temperatures.^ Con- 

 sequently water expansion alone will account for only about one 

 fifth of the water excretion. Similar results were obtained for 

 several other woods, including Corylus avellana, Abies cxcelsa, 

 birch, beech, and oak, except that in the beech and some others 

 the excretion was as much as seven times greater than the com- 

 puted water expansion or even more. In many cases bubbles of 

 air were extruded on warming, and none of the pieces of wood 

 were saturated. Therefore Sachs concluded that the increased 

 excretion of water must have been due to the expansion of gas, 

 and this without doubt is the only true explanation of the pheno- 

 mena with which he was dealing. 



After recording this and several other similar experiments, 

 Sachs felt warranted in saying that, if a rooted maple tree at a 

 temperature of 0° R. in all its parts, is cut in two in the middle 

 and the lower part with the roots is warmed, water will flow from 

 the cut surface of the stump; and Hkewise, if the upper part with 

 the branches is warmed, sap will flow from the other cut surface. 



