466 BOTANICAL GAZETTE [December 



mirrored surface, (If, at this stage, more water is removed, the 

 minute spot rapidly enlarges, spreading out in a circle around the 

 exposed point.) A very small quantity of water was then added, 

 just enough to obliterate entirely the dark spot. Usually the 

 finger was applied to the point at this stage, a slight rubbing serving 

 to clean the point and to break the surface film of the water over 

 the point. The point was shown to be just beneath the surface, as, 

 on lightly touching the water with the finger, a tiny wave would 

 be sent across the surface, and the point would appear momentarily 

 to be immediately obliterated. This phenomenon was most readily 

 observed when the light was somewhat diminished. At midday the 

 sun on a cloudless day was so intense that the eye was flooded with 

 the light reflected from the water surface, making it difficult to see 

 the dark spot. Repeated observations under varying conditions 

 carried out with a test tank showed that on the average a change 

 in water height produced by the addition or subtraction of 25 cc. 

 of water could be detected. Since the area of water surface in 

 the tank approximated 4,575 sq. cm., this volume would give a 

 theoretical difference in height, or minimum error due to observa- 

 tion, of 0.05 mm. 



Experiments 



main 



summers 



first 



km 



west of Ann Arbor, Michigan. A long, narrow peninsula covered 

 with a scrubby growth of trees and shrubs separated the bay from 

 the lake proper and protected the station from the prevailing 

 westerly winds. It was subsequently found necessary to establish 

 a boom of floating logs on the south and east as a protection against 

 occasional winds from these directions. 



Two tanks at the diagonal corners of the raft contained neither 

 soil nor plants and were used to ascertain the evaporation from the 

 free water surface. The other tanks were supplied with lake mud. 

 the depth depending upon the plants placed therein. Plants from 

 the immediate vicinity were planted in these tanks, the depth of the 

 water being in all cases identical with the depth of the water where 

 the plants were naturally growing. Care was also taken to set the 



