230 BOTANICAL GAZETTE [apeil 



three 



see that as we go west from the middle outer portion of the salt marsh 

 into the garden, the rate decreases by a fourth. As we go south to the 

 edge of the salt marsh the rate is decreased nearly a half, and when 

 the middle of the fresh-water marsh is reached the rate has decreased 



to one-third 



GENERAL CONCLUSIONS 



From the foregoing data, meager as they are, it is evident that we 



have in the porous-cup vaporimeter an instrument that will be of the 



greatest importance in the study of habitat conditions. Unlike so many 



of the recently exploited forms of instruments, it furnishes data that 



may be directly related to the plant. It has the great advantage over 



readings made from open water surfaces, in that the surface is constant 



and is continually exposed in the same way. Comparative readings 



may be obtained from different habitats by first standardizing the in- 

 struments. 



The principal objection to the porous cup is its inability to with- 

 stand frost; this makes it useless in winter and early spring; it also 

 limits its use at high elevations. Its usefulness might be greatly 

 increased by making it self-recording, so that the diurnal variations 

 coidd be accurately obtained. 



The data which my observations furnish are of course character- 

 istic only of the stratum within a meter of the soil surface. In the 

 forest as we go from soil to tree -top this relative evaporation must 

 mcrease; but it is in the lowest stratum that the seedlings, which are 

 to determine the future of the area, have their struggle with the environ- 

 ment. With these data in hand it is not difficult to see why seedlings 



ire siirrps;sfnl in the swaniD forest 



eratnim 



with Its lo per cent, relative evaporation ; why they fail in the opec 

 hiUside forest with its 50 per cent, rate ; and why they are never seen 

 on the near-by gravel sUde with its relative rate of 100 per cent, in 



unstable 



The foUowing diagram (fig. p) shows in a more graphic way the 

 nparative rates of evaporation in the lowest stratum of the common 

 aitats about Cold Spring Harbor. 



In the reforestation of the gravel slides in this locality it wU be seen 

 It the greatest decrease in th*. ri^r^^r^A. f^,- t^or,c.rl,-rati'f^n nn the 



