ter of fact is not appreciably affected thereby, but solely because of the experimental uncertainties involved in work upon either exceedingly slow or exceedingly fast drops. When the velocities are very small residual convection currents and Brownian movements introduce errors, and when they are very large the time determination becomes unreliable, so that it is scarcely legitimate to include such observations in the final mean. However, for the sake of showing how completely formula (9) fits our experimental results throughout the whole range of the observations of Table XIII, figure 3 has been introduced. The smooth curve in this figure is computed from (7) under the assumption of e=4.891X10-10 and the experimentally determined values of e, are plotted about this curve, every observation contained in Table XIII being shown in the figure.

The probable error in the final mean value 4.891 X 10-10, computed by least squares from the numbers in the last column, is four hundredths of 1 per cent. If there is an error of as much as 3 per cent in the determination of A the final value of e would be affected thereby by only about 0.2 per cent. Since, however, the coefficient of viscosity of air is involved in the formula, the accuracy with which e is known is limited by that which has been attained in the measurement of this constant. There is no other factor involved in this work which has not been measured with an accuracy at least as great as 0.2 per cent.

The value of 15 which has been used in the computation of all of the preceding tables, viz, 0.00017856, is in my judgment the most probable value which can be obtained from a study of all of the large mass of data which has been accumulated within the past 40 years upon this constant. It represents not only the result of what seems to me to be the most reliable single determination of μ which has thus far been made, viz, that of Stokes and Tomlinson 1 who deduced it from the damping of oscillating cylinders and spheres, but it is exactly the mean of the three most recent and very concordant values obtained by the outflow method (Table XV), and it is furthermore the mean of all of the most reliable determinations which have ever been made. These determinations are as follows:

[The discussion of the determinations of the coefficient of viscosity of air is here omitted.]

We have devised two modifications of this method of determining e which do not involve the value μ. It is scarcely likely, however, that the necessary experimental error in these methods can be reduced below the error in μ It is probable, therefore, that any increased accuracy in our knowledge of e is to be looked for in increased accuracy in the determination of μ.

1 Stokes, Math, and Phys. Papers, v. 5, p. 181.

EXPERIMENTS UPON SUBSTANCES OTHER THAN OIL.

All of the preceding experiments except those recorded in Table I were made with the use of a specially cleaned gas-engine oil of density 0.9041 at 25° C. Those in Table I were made with the use of a similar, though more volatile, mineral oil (machine oil) of density 0.8960. The reason that we worked so continuously upon a single substance was that it was found that in order to maintain a drop of constant size it was necessary, even with these very nonvolatile substances, to have the drop in equilibrium with its saturated vapor. This is shown by the following observations. The inner surfaces of the condenser plates had been covered with a very thin coat of machine oil in order that they might catch dust particles. Drops blown from a considerable number of nonvolatile substances were introduced between the plates and were found in the main to evaporate too rapidly to make accurate observing possible. This was true even of so nonvolatile substances as glycerine and castor oil, as the following observations show:

In order to get rid of this continuous increase in G, the drops were next blown from the least volatile liquid at hand, viz, gas-engine oil, and the behavior of a given drop showed immediately that it was growing in size instead of evaporating. This can be seen from the following readings:

This behavior was shown consistently by all the drops experimented upon (six or eight in number) throughout a period of two days. Imagining that the vapor from the more volatile machine oil upon the plates was condensing into the less volatile but similar oil of the drop I took down the apparatus, cleaned the plates carefully, and oiled them again, this time with the gas-engine oil. Every gas-engine oil drop tried thereafter showed the sort of constancy which is seen in Tables III to XII. Series of observations similar to that made upon gas-engine oil and tabulated in Tables XIII and XIV will ultimately be made upon other substances. Thus far the aim has been to take enough observations upon other substances to make sure that the results obtained from these substances are substantially in agreement with those obtained from gas-engine oil and to concentrate attention upon an accurate series of observations upon one substance. As a matter of fact, we have a fairly complete series upon machine oil and a number of observations upon watch oil, castor oil, and glycerine, all of which are in agreement within the limits of observational error, in some cases as much as 2 or 3 per cent, with the observations upon gas-engine oil.

*

*

*

*

The conclusion to be drawn from all of the work thus far done on substances other than oil is merely that there is nothing in it to cast a doubt upon the correctness of the value of e obtained from the much more extended and much more accurate work upon gas-engine oil.

COMPARISONS WITH OTHER DETERMINATIONS.

The value of e herewith obtained is in perfect agreement with the result reached by Regener1 in his remarkably careful and consistent work in the counting of the number of scintillations produced by the particles emitted by a known amount of polonium and measuring the total charge carried by these same particles. His final value of this charge is 9.58 × 10-10, and upon the assumption that this is twice the elementary charge-an assumption which seems to be justified by Rutherford's experiments 2-he finds for e 4.79 × 10-10, with a probable error of 3 per cent. Since the difference between this value and 4.89 × 10-10 is but 2 per cent the two results obviously agree within the limits of observational error.

* * *

[The author then discusses several other determinations of e, and explains some discrepancies which appear.]

In conclusion there is presented a summary of the most important of the molecular magnitudes, accurate values of which are made

1 E. Regener, Sitz. Ber. d. k. Preuss. Acad. d. Wiss., 37, p. 948, 1909.

2 Rutherford, Phil. Mag., 17, p. 281, 1909.

possible by an accurate determination of e. The Faraday constant is taken as Ne = 9,655 absolute electromagnetic units.

the smallest quantity of electricity capable of separate existence.

the number of molecules in one gram molecule of any substance.

the number of molecules in 1 cubic centimeter of any gas at 0° C. and 76 centimeters.

the constant of molecular energy. Molecular en-
ergy € = aT.

the kinetic energy of agitation of a single molecule
at 0° C. and 76 centimeters. ε=273α.
the weight of the hydrogen atom.

My thanks are due to Profs. Crew, Carman, and Guthe for loaning to me tubes of radium when my own supply met with an accident. I wish also to acknowledge my great indebtedness to Mr. Harvey Fletcher who has most ably assisted me throughout the whole of this investigation.

1 These diameters have been obtained from the above value of n and the viscosity equation

Sutherland's correction for cohesional force (Phil. Mag., 17, p. 320, 1909) and Jean's correction for persistence of velocities being added. This procedure is thought to yield more reliable results than applying the above corrections to means of D obtained from viscosity, diffusion, heat conduction, and departures from Boyle's law, since computations based on the last three phenomena involve both theoretical and experimental uncertainties of large magnitude,

THE TELEGRAPHY OF PHOTOGRAPHS, WIRELESS AND

BY WIRE.1

[With 2 plates.]

By T. THORNE Baker, Esq., F. C. S., A. I. E. E.

It frequently happens that when two alternate processes are available for certain work, and one of them is considerably less practical than the other, the less practical one is possessed of much higher scientific interest. This may certainly be said of the telegraphy of pictures and photographs. The whole of the methods of transmission can be classed as either purely mechanical, or dependent on the physical properties of some substance which, like selenium, is sensitive to light.

The latter methods are of no little scientific interest, and, although very delicate and for the moment obsolete, there is every likelihood of their coming into more extended use later on.

The telegraphy of pictures differs only from the transmission of ordinary messages in that the telegraphed signals, recorded by a marker on paper, must essentially occupy a fixed position. In the case of an ordinary telegram it matters little whether the received message occupy two, three, or more lines when written out on paper, but when a picture is telegraphed every component part of it must be recorded in a definite position on the paper.

Suppose you greatly enlarge a portrait, and divide it up by ruled lines into a thousand square parts. Suppose also that the photograph is printed on celluloid, so that it is transparent. If, now, the portrait be held in front of some even source of illumination, it will be seen that each square-each thousandth part-is of different density. The light parts of the photograph will consist of squares of little density, the dark parts, of squares of greater density, and so on. In this way the photograph is analyzed into composite sections, each section corresponding precisely to a letter in a message; letters and

1 Lecture before the Royal Institution of Great Britain, at the weekly evening meeting, Friday, Apr. 22, 1910. His Grace the Duke of Northumberland, K.G., P.C., D.C.L., LL.D., F.R.S., president, in the chair. Reprinted by permission from author's copy published by the Royal Institution. Printed also in Nature, No. 2129, Aug. 18, 1910.

97578°-SM 1910-17