Photophone," because an ordinary beam of light contains the rays which are operative.

Non-Electric Photophonic Receivers.

It is a well-known fact that the molecular disturbance produced in a mass of iron by the magnetizing influence of an intermittent electrical current can be observed as sound by placing the ear in close contact with the iron, and it occurred to us that the molecular disturbance produced in crystalline selenium by the action of an intermittent beam of light should be audible in a similar manner without the aid of a telephone or battery.

Many experiments were made to verify this theory, but at first without definite results.

The anomalous behavior of the hard rubber screen alluded to above suggested the thought of listening to it also.

This experiment was tried with extraordinary success. I held the sheet in close contact with my ear while a beam of intermittent light was focussed upon it by means of a lens. A distinct musical note was immediately heard. We found the effect intensified by arranging the sheet of hard rubber as a diaphragm, and listening through a hearing tube, as shown in fig. 10.

We then tried crystalline selenium in the form of a thin disk, and obtained a similar but less intense effect.

The other substances, which I enumerated at the commencement of my address, were now successively tried in the form of thin disks, and sounds were obtained from all but carbon and thin glass.22

In our experiments, one interesting and suggestive feature was the different intensities of the sounds produced from different

We found hard rubber to

substances under similar conditions. produce a louder sound than any other substance we tried, excepting antimony, and zinc; and paper and mica to produce the weakest sounds.

On the whole, we feel warranted in announcing as our conclusion that sounds can be produced by the action of a variable light from substances of all kinds when in the form of thin diaphragms. The reason why thin diaphragms of the various materials are more effective than masses of the same substances appears to be that the molecular disturbance produced by light is chiefly a surface action, and that the vibration has to be transmitted through the mass of the substance in order to affect the ear.

On this account we have endeavored to lead to the ear, air that is directly in contact with the illuminated surface, by throwing the beam of light upon the interior of a tube; and very promising results have been obtained. Fig. 11 shows the arrangement we have

tried. We have heard from interrupted sunlight very perceptible musical tones through tubes of ordinary vulcanized rubber, of brass and of wood. These were all the materials at hand in tubular form, and we have had no opportunity since of extending the observations to other substances.23

I am extremely glad that I have the opportunity of making the first publication of these researches before a scientific society, for it is from scientific men that my work of the last six years has received its earliest and kindest recognition. I gratefully remember the encouragement which I received from the late Professor Henry, at a time when the speaking telephone existed only in

23 A musical tone can be heard by throwing the intermittent beam of light into the ear itself. This experiment was at first unsuccessful on accourt of the position in which the ear was held.

theory. Indeed, it is greatly due to the stimulus of his appreciation that the telephone became an accomplished fact.

I cannot state too highly also the advantage I derived in preliminary experiments on sound vibrations in this building from Professor Cross, and near here from my valued friend Dr. Clarence J. Blake. When the public were incredulous of the possibility of electrical speech, the American Academy of Arts and Sciences, the Philosophical Society of Washington, and the Essex Institute of Salem, recognized the reality of the results and honored me by their congratulations. The public interest, I think, was first awakened by the judgment of the very eminent scientific men before whom the telephone was exhibited in Philadelphia, and by the address of Sir William Thomson before the British Association for the Advancement of Science. At a later period, when even practical telegraphers considered the telephone as a mere toy, several scientific gentlemen, Professor John Pierce, Professor Eli W. Blake, Dr. Channing, Mr. Clark, and Mr. Jones of Providence, R. I., devoted themselves to a series of experiments for the purpose of assisting me in making the telephone of practical utility; and they communicated to me, from time to time, the results of their experiments with a kindness and generosity I can never forget. It is not only pleasant to remember these things and to speak of them, but it is a duty to repeat them, as they give a practical refutation to the often repeated stories of the blindness of scientific men to unaccredited novelties, and of their jealousy of unknown inventors who dare to enter the charmed circle of science.

I trust that the scientific favor which was so readily accorded to the Telephone may be extended by you to this new claimant,— "The Photophone."

ON SOME OF THE CONSEQUENCES OF THE HYPOTHESIS, RECENTLY PROPOSED, THAT THE INTRINSIC BRILLIANCY OF THE FIXED STARS IS THE SAME FOR EACH STAR. By EDWARD S. HOLDEN, of Washington, D. C.

I.

In all statistical researches upon the arrangement of the fixed stars in space, it has been found to be necessary to make some fundamental assumption, more or less probable.

Some assumption is forced upon us by our complete ignorance of the nature of the stars themselves and of the real laws according to which they are distributed.

The fundamental assumptions have usually been, either that the stars are all of equal brightness; or else that they are equally scattered, so that within equal portions of space, equal numbers of stars exist. Various modifications of these two hypotheses have been made and their consequences worked out; but some form of one of them has usually been the starting point.

It may be of interest to see the consequences which follow from an assumption less violent than either of the preceding. This is that the brightness of the unit-area of all stars is the same. The simple formulæ which relate to this subject were put in a form which allows the consequences of each of the three fundamental assumptions to be seen, in NEWCOMв and HOLDEN'S Astronomy,2 page 489. The hypothesis that the brightness of the unit-area of all stars is the same has been recently made the basis of computation in a paper by Prof. E. C. PICKERING, "Dimensions of the Fixed Stars, etc., reprinted from the Proceedings of the American Academy, Cambridge, 1880," in which (page 35) it is used to determine the dimensions of a dark satellite to Algol. As we can have no à priori proof of the truth of this hypothesis, it will perhaps be useful to trace its consequences in various directions. Although the data at our command are not sufficient to enable us to come to certain conclusions, yet they may be sufficient to test the value of the fundamental assumption, and it is only for this reason that I bring them together. It may not be improper to say

1 Dimensions of the fixed stars, etc., by Prof. E. C. PICKERING, Cambridge, 1880, p.3. 2 American Science Series, New York, 1879.

that they were deduced in 1877, and would have remained unpublished except for their bearing on the present question.

For

While the first two assumptions in regard to the distribution of stars have been shown to be roughly and in a general way approximations to the truth 3 we know that they are in fact untrue. the hypothesis that all stars are of equal brightness, and thus that stellar magnitude depends on stellar distance alone, is contradicted directly by the determinations of parallax, and more glaringly and in a more general manner by the existence of clusters, in which stars of different brightness are associated at the same distance from the earth.

The hypothesis of equable distribution is negatived by the existence of clusters at all, so that while this supposition is also in a general way true, it needs serious modifications to make it fit special cases. The third hypothesis of equal brightness of the unit-area of the surfaces of all stars is certainly not true in every case, but in any case it is less violent than either of the others à priori. The objections to it we may consider later. It may be mentioned here, that of the 324,000 stars from first to ninth magnitude, we know considerably less than 1,000 highly colored stars ; the vast majority of stars being white. This is in no sense a proof of the assumption. It does not militate against it, however, and is what might be expected if the assumption were indeed true. We may express our conditions in an algebraic form as follows: If S be the surface and R the radius of a star at distance D; i the amount of light emitted per unit of surface, Bm, its brightness in arbitrary units as seen from the earth, m, its stellar magnitude on any scale whose light-ratio is d, then the most general expression for Bm is

=

Xi

if light is not extinguished in space, or

SX if light is extinguished in space.

The unit of D is arbitrary.

If the brightness of an average first magnitude star is unity,

B1 = 1, then

(2) Bm=¿m-1; so that for a star of the mth magnitude

Not only by the results of W. HERSCHEL, but by the later researches of PETERS, GYLDEN, C. S. PEIRCE, GOULD, and others.