858 



SCIENCE 



[N. S. Vol XLIII. No. 112(> 



increase of colloidal aggregation increases vis- 

 cosity and causes a decrease of permeability: 

 but if this goes beyond a certain point it pro- 

 duces " a decrease in the degree of intimacy 

 between disperse phases and solvent; the fluid- 

 ity is suddenly increased and diffusion across 

 the surface is correspondingly facilitated." 



Some years ago a similar conception was 

 suggested to the writer by the fact that living 

 tissue of Laminaria placed in JSTaCP becomes 

 much softer while in CaCl, it becomes much 

 harder. The changes in viscosity are so great 

 as to suggest that they are fully capable of ex- 

 plaining the faU. of the electrical resistance of 

 the tissue which occurs when it is placed in 

 NaCl and also the rise of resistance which oc- 

 curs in CaClj (which is always followed by a 

 fall of resistance). 



In the hope of throwing some light upon this 

 process sections of the tissue were observed in 

 CaCl, imder the microscope. It was then seen 

 that after a time the protoplasm assumed a 

 coagulated appearance: it seemed obvious that 

 the process which increased the viscosity might 

 produce a coagulation of the protoplasm or 

 some other change in its structure whereby it 

 became more permeable. 



This conception led the writer to expect de- 

 creased resistance in tissues placed in NaCl 

 (because of decreased viscosity) while in CaCl. 

 we should expect to find increased resistance 

 (due to increased viscosity) followed by a fall 

 of resistance (due to coagulation or other 

 structural change in the protoplasm). 



It soon became apparent that there were sev- 

 eral serious objections to this conception. The 

 most important of these may be briefly stated 

 as follows: 



1. If to a solution of NaCl we add CaCL 

 until the increase of viscosity produced by one 

 salt is just balanced by the decrease produced 

 by the other, the resistance should remain sta- 

 tionary. This is not the case, though it seems 

 to be so when the observations are not taken 

 frequently enough (as happened in some early 

 experiments). There is always a fall, or a rise 

 followed by a fall, of resistance. 



2 Throughout this paper NaCl and CaCl. of the 

 same conductivity as sea water are referred to. 



2. If more CaCl, be added there should be a 

 rise of resistance : this should after a while be- 

 come stationary, provided there is not enough 

 CaCl, to produce the coagulation or other 

 structural change which decreases the resist- 

 ance. This does not occur: the tissue never 

 maintains its increased resistance, but shows 

 a fall of resistance which begins soon after the 

 maximum is reached. 



3. If still more CaCl, be added, so as to pro- 

 duce the coagulation or other structural change 

 which decreases resistance, we should expect to 

 flnd in all cases the same viscosity (and con- 

 sequently the same maximum of resistance) 

 just before the fall begins. Still further in- 

 crease of CaCl, would only hasten this process 

 without changing the maximum. This does 

 not correspond with the facts. The maximum 

 steadily rises as the proportion of CaCl^ in- 

 creases, so that the greatest maximum is found 

 in pure CaCl,. 



4. If the fall of resistance in CaCl, is due 

 to coagulation or to some other structural 

 change it might be expected to be irreversible 

 almost from the start; but this is not the case. 

 Only when it has proceeded a good way toward 

 the death point does it become irreversible. 

 On the other hand the fall in NaCl (due to 

 liquefaction) might be expected to be revers- 

 ible at every stage. But it ceases to be wholly 

 reversible after it has proceeded one sixth of 

 the way (or less) to the death point. 



5. The efl^ect of anions on the permeability 

 of Laminaria is completely at variance with 

 their effect on the viscosity of colloids as seen 

 in Hofmeister's series. 



6. Since the changes in viscosity occur in 

 dead as well as in living tissue we should ex- 

 pect to find in both cases similar changes in 

 resistance. It is found that the decrease in 

 viscosity in ISTaCl produces no appreciable 

 effect on resistance. Even when the process 

 proceeds so far that the tissue is reduced to a 

 very soft jelly there is little or no change in 

 resistance.^ The hardening in CaCl, produces 

 some rise in resistance, but it is much too 



3 In a liquid a change of viscosity alters the re- 

 sistance, but this is not necessarily the case in a 



