PERMEABILITY BY THE METHOD OF TISSUE TENSION 563 
a secondary increase in the curvature of strips from the peduncles 
of Centranthus ruber and Rudheckia triloba. 
Among such tissues, strips of peduncles of the dandelion (Taraxa- 
cum officinale Weber) are well known for their large and rapid response 
to changes in the concentration of the solution in which they are 
immersed. Such strips also showed themselves to be excellent 
material for the study of the rate of penetration of salts. Upon being 
cut, they are forced by the existing tension to bend rather strongly 
outward around an axis tangential to the peduncle. If they are then 
placed in a slightly hypertonic solution, their curvature decreases 
during a period varying from a few seconds up to several minutes^ 
remains constant a moment and then slowly increases, sooner or later 
exceeding the original curvature. The last phase is analogous to the 
recovery of plasmolyzed cells, and will, for convenience, be also 
termed "recovery." 
By means of observations on the rate at which this recovery oc- 
curred in various salt solutions it was possible to determine the per- 
meability of dandelion protoplasm to inorganic salts, and the pro- 
gressive changes in permeability produced by such salts. 
Method 
The salts used were the same or of the same grade of purity as 
those used in the experiments on exosmosis from dandelion tissue, as 
described by the writer in a recent paper (2). They were dissolved in 
distilled water which had a specific conductivity of less than 2 X io~^ 
ohms. Molecular stock solutions were made up and carefully stan- 
dardized by titration against a o.l ikf solution of silver nitrate with 
potassium monochromate as an indicator. The desired concen- 
trations were secured by dilution of these stock solutions, and were 
accurate to 0.5 percent. This accuracy was sufficient for the purposes 
of the experiment. 
Strips of dandelion peduncle, each about 2.5 cm. long and 3 mm. 
wide, or similar strips from the midrib of the leaf were used in these 
experiments.^ There was practically no difference in the data furnished 
by the two types of tissue. 
Each strip was firmly gripped at one end by the two halves of a 
partially split rubber stopper, which in turn was secured by means of 
2 The midrib, like the peduncle, is hollow and for the sake of uniformity strips 
from the upper (ventral) half of this tissue were always used. 
