36 
AMERICAN JOURNAL OF BOTANY 
[Vol. 8 
the afternoon, as was indicated by the data secured at Ithaca, but the 
difference between the osmotic concentration of the guard cells and that 
of the cells of the epidermis is much greater. Expressed in atmospheres, 
the difference is 8.72 as compared with 3.00 at Ithaca. 
The difference between the curves of January i and January 2 may be 
explained by the difference in the amount of sunshine. On January i the 
sun did not shine during the entire day, while on January 2 the sunshine was 
continuous in the afternoon although very limited in the forenoon. 
Figure 4 shows a check series run with KNO,, instead of CaCU. The 
IpA.'V*. %a.^- /Oavi. !Z. m. Zp.ca. "^P.t^. (oF.m. 
Fig. 4. Cyclamen. Observations made at Columbia. 
same general curve is secured with these solutions as with solutions of 
CaCL; however, the molecular concentration in all cases was considerably 
higher. This is probably shown best by a comparison of the threshold 
concentrations of the epidermal cells in the two series. 
Figure 5 shows the results of observations on cyclamen made at Columbia 
on January 3. The sun shone brightly until 1:45 p.m., after which there 
was no sunlight. These results accord with the results of January 2, with 
the exception that there was a greater difference between the osmotic con- 
centrations of the epidermal and of the guard cells. Expressed in atmos- 
pheres, the greatest difference was 19.4. 
