20 The Ohio Naturalist. [Vol. X, No. 2, 



great. The importance of pioneer shade plants as reducers of 

 transpiration is also pointed out. 



The instruments used at Buckeye Lake were a slight mod- 

 ification of those used by the writers just mentioned. The 

 evaporation takes place from the surface of an exposed porous 

 clay cup, about thirteen centimeters in length, two and one-half 

 cm. in diameter, and with a wall of four millimeters thickness. 

 The upper end is closed, and rotmded, the lower end is closed 

 tightly by a perforated rubber stopper, through which passes a 

 glass tube. This tube extends down to the bottle below, which 

 serves as the reservoir of water. Since the instruments were to 

 be left for rather long periods of time, a larger and more stable 

 form of reservoir was required. In place of the "Mason " jar and 

 cork stopper, a bottle of 5U0U cc. capacity was used, with a neck 

 of sufficient slope so that the water level could readily be seen 

 from above. x\t the mouth of the bottle, the glass tube passed 

 through two rubber stoppers, the one a two-holed stopper 

 inserted in the bottle, and the other with its large end down, cov- 

 ering the hole, and preventing the entrance of water, but allow- 

 ing free access of air. A file mark near the top of the bottle 

 indicated the point to which the water level was raised on refilling. 

 Throughout the experiment only distilled water, containing a 

 small per cent of formaldehyde, was used. 



The interior of the cup remains free from air because of the 

 surface tension of the water films closing the pores. The cup 

 thus remains filled with water, and as evaporation takes place at 

 the surface, more water is forced up from below into the vacuum 

 by the air pressure upon the water surface in the reservoir. 



The porous cups used in this work were obtained through 

 Dr. Dachnowski from the Carnegie Institute and were standard- 

 ized at the Desert Laboratory at Tucson. 



When used during the growing period of plants, the principal 

 defect of this instrument is that rain may enter the reservoir 

 through the porous cup, and thus cause an error in the results. 

 If daily readings are taken, the length of time of precipitation 

 can be recorded, and corrections made for the error. But in 

 taking readings at intervals longer than a day, this error must be 

 neglected. 



One instrument was placed in a station of the Maple-Alder 

 zone near the border of the island and was shaded by x\cer 

 rubrum, Alnus rugosa, and Rhus vernix. Osmunda cinna- 

 momia, O. regalis, and Dryopteris cristata were growing nearby. 

 The other station was in the central zone, where the principal 

 plants were Sphagnum, Oxycoccus oxycoccus, Drosera rotundi- 

 folia, Eriophorum virginicum, and Dulichium arundinaceum. 



Readings were begun May 14, and taken weekly until June 11. 

 No data were taken then tuitil July 17, when the evaporation for 

 five weeks was recorded. Weekly readings were then resumed 

 and taken until August 21, when after another break of three 



