1914] OTIS—TRANSPIRA 



465 



maintained. The wire netting was also removed from the raft, and 



a netting 1.5m. high was substituted at a distance of 1 . 5-2 m. 



from, but inclosing, the raft, since turtles were not abundant. The 



relations of the various instruments and the pontoons is shown 

 in fig. 3 . 



Methods 



Readings were regularly made early in the morning and late 

 in the afternoon or early evening, since at these times the water 

 was usually quite still and the light conditions best for accurate 

 readings. A boat was necessary at all times, all work being done 

 from it as soon as observations were begun. Measurements made 

 at the station included wet and dry bulb temperatures as indicated 

 by the psychrometer. Temperatures of the water in the tanks 

 and of the water surrounding the tanks were ascertained and 

 recorded both at the surface and at a depth of 76 cm. beneath the 

 surface. Continuous records of air temperature and wind were 

 made by the recording instruments. The amount of water taken 

 up by the plants or evaporated from the water surface was ascer- 

 tained by measuring the quantities of water it was necessary to 

 place in the tanks to bring the water to the standard level, as 

 indicated by the height of the point of the brass rod. Water taken 

 from the lake or river surface adjacent to the tanks was always used, 

 since it was of the same temperature and had other characteristics 

 in common with the water in the tanks. Once a week the tanks 

 were filled to overflowing with several bucketfuls of water rapidly 

 applied to remove any scum or refuse matter which might accumu- 

 late and which would influence the evaporation therefrom. 



The best way to ascertain accurately the amount of water to 

 be added to a tank in each case was found to consist in adding a 



e> 



little more water than was necessary to cover the point. On placing 

 the eye close to the level of the water a point could be found where 

 the entire surface of the water in the tank appeared like a bright 

 mirror. Then, with the eye directed toward the center of the 

 surface, a small quantity of water was removed. As soon as the 

 surface became still, an examination was made for the appearance 

 of the point. If it did not show, more water was removed, until 

 finally a minute black spot was distinguishable in the center of the 



