heat immediately on the appearance of the cloudiness above referred to, distinct and separate crystals can be observed under the microscope, which appear like leaden snowflakes on a ground of ruby red.

Upon removing the heat when crystallization is further advanced, we perceive under the microscope, masses of these crystals arranged like basaltic columns, standing detached from one another -and at a still higher temperature the distinct columns are no longer traceable, but the whole mass resembles metallic puddingstone with here and there a separate snowflake, like a fossil on the surface. Selenium crystals, formed during slow cooling after fusion, present an entirely different appearance, showing distinct facets.

I will now endeavor to explain the means by which a beam of light can be controlled by the voice of a speaker.

Photophonic Transmitters.

We have devised upwards of fifty forms of apparatus for varying a beam of light in the manner required, but only a few typical varieties need be described.

(1) The source of light may be controlled, or (2) a steady beam may be modified at any point in its path.

In illustration of the first method we have devised several forms of apparatus founded upon Koenig's manometric capsule, operating to cause variations in the pressure of gas supplied to a burner, so that the light can be vibrated by the voice.

In illustration of the second method, I have already shown one form of apparatus by which the light is obstructed in a greater or less degree, in its passage through perforated plates. But the beam may be controlled in many other ways. For instance, it may be polarized, and then affected by electrical or magnetical influences in the manner discovered by Faraday and Dr. Ker.

Let a polarized beam of light be passed through a solution of bisulphide of carbon contained in a vessel inside a helix of insulated wire, through which is passed an undulatory current of electricity from a microphone or telephonic transmitter operated by the voice of a speaker.

The passage of the polarized beam should be normally partially obstructed by a Nicols prism, and the varying rotation of the plane of polarization would allow more or less of the light to pass

through the prism, thus causing an undulatory beam of light capable of producing speech.

The beam of polarized light, instead of being passed through a liquid, could be reflected from the polished pole of an electromagnet in circuit with a telephonic transmitter.

Another method of affecting a beam of light is to pass it through a lens of variable focus 21 formed of two sheets of thin glass or mica containing between them a transparent liquid or gas. The vibrations of the voice are communicated to the gas or liquid, thus causing a vibratory change in the convexity of the glass surfaces and a corresponding change in the intensity of the light received upon the sensitive selenium. We have found that the simplest form of apparatus for producing the effect consists of a plane mirror of flexible material, such as silvered mica or microscope-glass, against the back of which the speaker's voice is directed, as shown in the diagram (fig. 5).

Light reflected from such a mirror is thrown into vibrations

When crystalline selenium is exposed to the undulatory beam, reflected from such an apparatus, the telephone connected with the selenium audibly reproduces the articulation of the person speaking to the mirror.

In arranging the apparatus for the purpose of reproducing sound at a distance, any powerful source of light may be used, but we have experimented chiefly with sunlight.

For this purpose, a large beam is concentrated by means of a lens upon the diaphragm mirror and after reflection is again

21 I observe that a lens of similar construction has been invented in France by Dr. Cusco, and is described in a recent paper in "La Nature," June 19, 1880. See, also, Scien. Amer., Aug. 28, 1880, Vol. XLIII, p. 131. Mr. Tainter and I have usel such a lens in our experiments for months past.

rendered parallel by means of another lens. The beam is received at a distant station upon a parabolic reflector, in the focus of which is placed a sensitive selenium cell, connected in a local circuit with a battery and telephone. We have found it advisable to protect the mirror by placing it out of the focal point, and by passing the beam through an alum cell, as shown in Fig. 6.

A large number of trials of this apparatus have been made with the transmitting and receiving instruments so far apart that sounds could not be heard directly through the air. In illustration, I shall describe one of the most recent of these experiments. Mr. Tainter operated the transmitting instrument, which was placed on the top of the Franklin School House in Washington, D. C., and the sensitive receiver was arranged in one of the

windows of my laboratory, 1325 L Street, at a distance of 213 metres.

Upon placing the telephone to my ear, I heard distinctly from the illuminated receiver the words:-"Mr. Bell, if you hear what I say, come to the window and wave your hat."

In laboratory experiments the transmitting and receiving instruments are necessarily within earshot of one another, and we have therefore been accustomed to prolong the electric circuit connected with the selenium receiver, so as to place the telephones in another room.

By such experiments we have found that articulate speech can be reproduced by the oxyhydrogen light, and even by the light of a kerosene lamp. The loudest effect obtained from light is produced by rapidly interrupting the beam.

A suitable apparatus for doing this is a perforated disk which can be rapidly rotated. The great advantage of this form of apparatus for experimental work is the noiselessness of its operation, admitting of the close approach of the receiver without interfering with the audibility of the effect heard from the latterfor it will be understood that musical tones are emitted from the receiver when no sound has been made at the transmitter. A

silent motion thus produces a sound. In this way musical tones have been heard even from the light of a candle.

When distant effects are sought the apparatus can be arranged as shown in fig. 7.

By placing an opaque screen near the rotating disk, the beam can be entirely cut off by a slight motion of the hand, and musical signals, like the dots and dashes of the Morse telegraph code, can thus be produced at the distant receiving station. Such a screen operated by a key like. a Morse telegraph key

FIG. 8.

is shown in fig. 8, and has been operated very successfully.

Experiments to ascer

tain the nature of the rays that affect selenium.

We have made experiments with the object of ascertaining the

nature of the rays that affect selenium. For this purpose we have

placed in the path of an intermittent beam various absorbing substances.

Prof. Cross has been kind enough to give his assistance in conducting these experiments.

When a solution of alum, or bisulphide of carbon, is employed, the loudness of the sound produced by the intermittent beam is very slightly diminished, but a solution of iodine in bisulphide of carbon cuts off most, but not all, of the audible effect. Even an apparently opaque sheet of hard rubber does not entirely do this. This observation, which was first made in Washington, D. C., by Mr. Tainter and myself, is so curious and suggestive that I give in full the arrangement for studying the effect.

When a sheet of hard rubber, A, was held as shown in the diagram (fig. 9) the rotation of the disk or wheel B interrupted what was then an invisible beam, which passed over a space of several metres before it reached the lens C, which finally concentrated it upon the selenium cell, D.

A faint but perfectly perceptible musical tone was heard from the telephone connected with the selenium that could be interrupted at will by placing the hand in the path of the invisible beam.

It would be premature without further experiments to speculate too much concerning the nature of these invisible rays; but it is difficult to believe that they can be heat rays, as the effect is produced through two sheets of hard rubber having between them a saturated solution of alum.

Although effects are produced as above shown by forms of radiant energy which are invisible, we have named the apparatus for the production and reproduction of sounds in this way "the