The rates of nerve-conduction are nearly normal in all these solu tions; but although the differences in rate are slight, they are such as we expected, the alkaline sea-water and the weakly acid sea-water are both slightly stimulating, while the more acid sea-water is depressant, as shown in table 3, the temperature of the solutions being 29° C. and the neutral point 6.88 P..
The effects are not so marked as those which are observed when the pulsating rings are taken from natural sea-water and placed in sea water diluted with slightly acid distilled water, or in sea-water made alkaline by being concentrated by evaporation at ordinary tempera tures exposed to the air; but they accord with expectation in so far as they go. Possibly the OH' anion is stimulating even after it penetrates the cell membranes, while the hydrogen cation is depressant under these conditions.
The effects upon the rate of nerve-conduction in Cassiopea which resulted from diluting Tortugas sea-water with alkaline distilled water of P. about 8 and with slightly acid distilled water of P. about 6 are given in. table 4, which shows that acid distilled water is slightly stimulating in comparison with alkaline distilled water in weak dilution down to about 75 per cent sea-water plus 25 per cent distilled water, and then becomes relatively depressant in stronger dilutions. There are many examples of this in the author's past experiments published in 1914, 1915, and 1916, wherein the distilled water used for dilution was usually acid. An interesting parallel appears to be that of acidosis in blood, wherein pathological effects appear when the amount of acid is so slight that the P. is hardly affected, the buffer value of the carbonates and phosphates, however, being reduced.
It was due, indeed, to the acidity of the distilled water, and to my not realizing the marked effect of slight changes in temperature, that I was misled into entertaining the false idea that the curve of decline in rate in diluted sea-water resembles that of Freundlich's curve of adsorption, and that therefore adsorption may play a fundamental role in nerve-conduction. These recent experiments, wherein the sea-water was diluted with distilled water of P. about 8, appear to show that adsorption has nothing to do with the matter.
Indeed, Ralph S. Lillie (1916)* has already called attention to the close resemblance between my curve for decline in rate of nerve-con duction and the corresponding decline in the electrical conductivity of the'diluted sea-water, and in this latest and best-established curve of 1916, using distilled water of about P. 8 to dilute the sea-water, the
accordance with Lillie's expectation is almost perfect, as appears from table 5 and figure 8. It should be said, however, that the degree of ionization of the sodium, calcium, and potassium of the sea-water follows nearly the same law. Thus the rate of nerve-conduction, as we shall show, appears to be directly proportional to the concentration of the surrounding cations of sodium, potassium, and calcium, mag nesium taking an almost negative part in the control of rate of nerve conduction in diluted sea-water.
Table 5 (illustrated by fig. 8) shows the rates of nerve-conduction in Cassiopea in Tortugas sea-water diluted with aerated, alkaline distilled water having a hydrogen-ion concentration of 1.17X or 7.93 P.
The close resemblance between the decline in rate of nerve-conduc tion and the corresponding decline in electrical conductivity of the surrounding sea-water suggests but does not prove that the two phe nomena are directly dependent one upon the other; but as we shall show later, this seems doubtful.
Lillie advocates an extension of the theory of Faraday and de la Rive that the transmission of the excitation state from the immediate site of activity to the adjoining resting areas is dependent upon an elec trical local action of the same essential nature as that which is respon sible for the etching or corrosion of non-homogeneous metallic surfaces, such as iron in contact with an electrolyte solution. Lillie calls this theory the "local action theory of conduction," and if, indeed, nerve conduction be such a process, its rate must be proportional to the electrical conductivity of the conducting medium and the surrounding fluid (sea-water).
Lillie's hypothesis is, moreover, indirectly supported by the recent work of Adrian (1916),* who shows that after stimulation the recovery of conductivity is apparently complete at the same instant when the recovery of excitation is also completed and thus the two mechanisms may be identical and conduction may be merely the spread of a local exciting process.