96 



EXPERIMENTS WITH PLANTS 



Fig. 78. Do the roots, on emerging from the sawdust, 



turn back to it? The air now contains much more 



moisture than before, but in other 



respects there is no essential 



change. It would seem that in 



the first instance they turned back 



because the air was so dry, or, in 



other words, were attracted'by the 



water in the sawdust. 



As a matter of fact, we know 



that tree roots find (and fill up 



more or less completely) cisterns 



and drains which are three hun- 



dred feet or more distant from the 



tree; in many towns there, exist 



ordinances prohibiting the main- 



tenance of certain trees within a 



hundred feet of a sewer. 



In the above experiments the roots are not under 



perfectly natural conditions; we may approach these 



more nearly by the ar- 

 rangement shown in Fig. 

 79. In the middle of a 

 box of earth place a 

 flower- pot, the hole of 



79. Arrangement for ascertaining whether wMch haS be611 



78. Arrangement to test the 

 behavior of roots in moist 

 air. (Sectional view). 



Closed by a rubber or 

 walls int cork stopper or by seal- 



