CIL IV] INTERCELLULAR SPACES. 107 



We have only made the experiment in winter, when 

 we found that the leaves on branches of Pr units lusitanica 

 which had been in 0'5 p.c. NaCl for 8 days, had distinctly 

 less power of reddening cobalt paper placed on their 

 lower surfaces than had the leaves of the control 

 branches. 



(119) Stomata: connection with intercellular spaces 1 . 



For this experiment the choice of a suitable leaf 

 is important. The following answer well : Arum macu- 

 latum, Ranunculus ficaria, Eranthis hie?nalis, Caltha 

 palustriSj Primula sinensis, Limnanthemum sp. 



The leaf stalk is fixed air-tight in a rubber cork which 

 fits a bottle filled with water. The leaf can be fixed by 

 piercing the cork with a hole too small for the stalk, and 

 dividing the cork longitudinally down one side till the 

 hole is laid open. Or an undivided hole may be used 

 if made air-tight with wax-mixture. Through a second 

 hole in the cork passes a glass tube (it must not dip into 

 the water) connected with the air-pump. When the pump 

 is set in action a stream of bubbles emerges from the cut 

 end of the stalk, which is below the surface of the water. 

 The reverse experiment may also be made by placing the 

 lamina in the water and the cut end in the air. 



(120) Injection with water. 



Many leaves, e.g. Ranunculus ficaria, Eranthis 

 hiemalis, Arum maculatum, Caltha palustris, Limnan- 

 themum, Hydrocharis, can be injected by sucking the stalk 



1 See the figures in Pfeffer's Physiologic, Vol. i. p. 96. 



