The Permeability of Cells for Acids. 153 
explanation of how substances pass membranes and how they enter 
cells, but unfortunately we know that the nature of the membrane 
makes a great deal of difference in determining which substances can 
pass it. Dead cells allow quite different substances to pass from those 
which living cells do. Traube now admits that the Haftdruck of the 
substance in the membrane is also essential. Surface tension is a 
measure of the Haftdruck of a solution, but we have no method of 
determining the Haftdruck of the substance in the membrane, so that 
the simplicity of the Haftdruck theory is lost. Traube now states 
that with certain substances even surface tension is no certain measure 
of the Haftdruck of a solution, a condition which still further compli- 
cates the theory (see Pfluger’s Archiv, 153, p. 304, 1913). 
According to Traube the lowering of surface tension (capillary 
activity), lipoid solubility, and osmotic velocity all run more or less 
parallel, but osmotic velocity is dependent primarily on capillary 
activity. Traube claims that capillary activity is a better measure of 
narcotizing power than is lipoid solubility, but our data on this point - 
are rather scanty. The error in determining narcotic strength is large 
and few accurate measurements of partition coefficients have been made. 
The Haftdruck of KOH is high. It slightly increases the surface 
tension of water, whereas the surface tension of aqueous propylamine 
is low. A normal solution of KOH has a surface tension (y15°) of 
7.46 mg. mm. while n-propylamine = 4.63 and water=7.3.! The pene- 
trability of these two alkalies accords very nicely with the Haftdruck 
theory. On the other hand NH,OH lowers the surface tension of 
water very little despite the fact that it enters cells with great rapidity. 
Thus in a capillary tube in which water will rise to a height of 91.5 mm. 
n/4 NH,OH rises to 91 mm. while diethylamine rises to only 68.8 
and piperidin to 67.5 mm. Yet NH,OH, piperidin, and diethylamine 
all penetrate cells equally readily. 
My results with acids give no clue as to whether lipoid solubility or 
capillary activity is the determining factor in penetration, since the 
penetration seriesis notin exact agreement with either lipoid solubility or 
capillary activity series. Wasteneys’sresultson the partition coefficients 
between olive oil and water show that the “‘lipoid’’ studied does not 
make very much difference. In any case butyric, propionic, and acetic 
acids are most out of place in the penetration series. I had fully 
expected them to penetrate the cells rather readily and was not a little 
surprised to note their position. I at first thought that they were so 
weak (only about 4 per cent dissociated) that the indicator (antedonin) 
would not detect them. This assumption can not be true, however, 
for valeric acid is also only 4 per cent dissociated, yet the indicator can 
detect it readily enough. The only other possibility is that the indi- 
cator is contained in some lipoid combination, similar in properties to 
'Surface tension figures from Traube. J. Ber. d. d. Physik ges., 10, p. 889, 1908. 
