Margaret Watts Hughes and the Shape of the Human Voice
Megan Watts Hughes (1842–1907), generally known as Margaret, was a passionate Welsh singer and inventor, who became one of the first women to present an invention at the Royal Society in London in 1887, when she unveiled her eidophone. It was a device that made possible to visualize the human voice. Before the arrival of sound recorders (the first recordings on wax cylinders were taking their first steps), the study of the human voice and singing was limited to the expertise of the trained ears of the teachers. It was known since the end of the 18th century that sound created geometric patterns in the sand, but it was considered necessary to improve the technique to find the right method with which to visually record a vocalized note. That was Margaret’s effort.
Mrs. Watts Hughes was also there. She is the lady who sings flower forms on to flat surfaces arranged for the purpose, by emitting notes through a sort of pipe, a performance at which I assisted one-day last season, and which may very possibly lead to considerable developments.
This quotation, taken from Notes from a diary, 1889–1891, by M. E. Grant Duff, sums up the fascination that this invention aroused in its time. Between 1891 and 1904 Margaret held various public demonstrations, as well as published articles and a book about her invention or, as she also defined it: “voice figures” or “visible sounds”, although it was also known as the phenomenon of “voice flowers”, since many of these shapes remembered the figures of various kinds of flowers. The eidophone caused a sensation in its time and was presented, besides in the Royal Society, in many other scientific institutions and musical associations.
Margaret had accidentally discovered the principle on which her invention worked in 1885. The inventor herself described it this way in 1875 (Proceedings of the Musical Association, UK):
It is now about a century since acoustical figures were first brought to notice by Chladni, and they have been since then the subject of many interesting experiments by Savart, Faraday, and others. (…) Considering the delicacy of the forces by which voice figures are produced, the impulses given to the vibrating disc by air thrills set in motion by the vocal organs, it is clear that great flexibility is a most essential quality in any vibrating medium to be used, and after many trials, the best I have found is sheet india-rubber of the most flexible description. This material combines also the advantages of being waterproof, very durable, and allowing a wide range in the size of discs that can be formed from it. Employing as the vibrating medium very flexible sheet rubber, formed into discs by being tightly and evenly stretched over the mouths of receivers of various shapes and sizes, I have been able to produce, by means of the voice, a large variety of figures of several distinct classes, all, so far as I have been able as certain, new to science, and seems to me suggestive of openings for further investigation. (…) The apparatus I have generally used is quite simple: consists of a receiver, open at the top, and with an orifice below. A piece of sheet rubber stretched across the mouth of the receiver. A tube, conveniently curved to allow vocal sounds to be introduced through it into the receiver by the orifice below the latter. Finally, sand, lycopodium, powdered colors, and liquids, such as water and milk, are among the substances which have placed upon the discs to produce, under the influence of the voice, figures of various descriptions.
The bases on which the eidophone was based were very simple, but it was necessary to be very observant to understand the phenomenon and to devise with it a means of registration: Someone sings in a room, while in a certain place we find a flat surface on which a handful of sand has been deposited, which comes to life by reacting by resonance to the sound, taking geometric shapes with beautiful symmetries. Sometimes these shapes will look like spirals, almost shells. In others, they will be as a leafy fern. And, from an observation like this, the eidophone was born. Margaret’s ingenuity gave, over the years, a way to record these forms created by the human voice as if they were drawings. It was not enough just to observe the phenomenon, she wanted it to become something that would be permanently preserved in physical form.
That device for “drawing” the human voice was truly ingenious as well as simple. With a little skill, anyone can build one and emulate the singer, in an experiment that is not only educational but also fun-filled. The eidophone consisted of a long tube, mostly made of brass or wood, with an open mouthpiece. This cylinder slightly enlarges its diameter until it ends in an elbow on which several bottomless resonance boxes were fitted, which, in their upper part, were endowed with an elastic rubber membrane that embraced its entire contour. The cups had different sizes, depending on the figure to be registered. The membranes served as a base on which were placed very fine spores of lycopodium*, whose behavior to sound seems to be much more adequate than that of sand, heavier and unpredictable. With the arrival of a sound to that kind of trumpet, the lycopodium began to dance to form certain patterns.
At the end of the 18th century the German physicist Ernst Chladni, considered one of the fathers of acoustics, had already described in great detail the geometric patterns that appear when a thin layer of sand is deposited on a flat surface and vibrated in response to certain frequencies. This illustrious predecessor of Watts Hughes as far as the visual “recording” of sound is concerned, was also one of the first to propose outer space as the place of origin of the meteorites, an idea that in his time was rejected.
Chladni’s experiments were later popularized thanks to a description by the Irish physicist who pioneered the study of colloids, John Tyndall. At that time French mathematician Sophie Germain had also experimented with string bows and sanded surfaces for her work on vibrations. It is not known to what extent Margaret came to know these previous experiences, but concerned to improve her method of teaching singing, decided to investigate the vocalization and tones of the human voice but did not find any suitable apparatus for it. That did not stop her restlessness, designing her own contraption to register the voice. Her intention was to capture various musical notes that originated in the singing by means of a simple device. She tried different methods, membranes and, the most complicated thing, several means of registration. The sand, inherited from Chladni’s experiments, worked, but it was not very fine. Margaret experimented with all kinds of powdery, granular or even gelatinous substances until she reached what became her medium of choice: lycopodium spores. The result was wonderful because those spores, unlike other materials, did not tend to escape from the membrane, but quickly organized themselves into complex geometric patterns as they were subjected to sound. He also experimented with some success using talc.
The final procedure consisted of singing a note tuned to the end of the mouthpiece of the eidophone. The sound reached the container on which was placed the membrane in tension, which in its center contained a pile of spores of lycopodium. Shortly after being exposed to the sound, the marvelous geometric patterns of that particular musical note appeared. Naturally, if the emitter did not tune correctly or they sang the note forcibly, the result was different. As a curious fact, it is worth mentioning that if the note was maintained in time but in a decreasing form, the granules were blurring the shape created to be grouped again in the center of the membrane.
The process had its complexity, especially on the part of the singer object of the experiment, because the tuning had to be precise. A determined and very clean note was what was necessary to create the desired patterns. In this way, the tone, intensity, and quality of tuning could be studied in various situations. It could be said that the eidophone was a detector of tuning problems, in direct competition with the best-trained ears. A slight change of nuance, an overtone or any change in the “pure” emission of the note, and the lycopodium pattern showed variations. It was an ideal method for the study of voice and singing, but it did not go far beyond what Chladni and other later experimenters had already achieved. Where Margaret made the difference was in the systematics used and in her method for the permanent registration of images. Seeing the granules of talc or lycopodium jump on the membrane was fun, but at the end of the experiment, the figures blurred. Photography could be used to remedy it, capturing the figures in their fullness, but Mrs. Watts Hugues wanted to go further.
Experimenting with mixtures of powdery substances and liquid media, she managed to obtain a kind of water emulsion with color pigments which, spread finely on a sheet of glass subjected to the membrane, made it possible to fix the image as if it were a painting. Later she used colored glycerin as a medium to create better images. Although these captures are colorful and spectacular, they are barely known, since in their time the use of color photography had not been extended and only in recent times have the originals been rescued. Margaret Watts Hughes’ final interest in the forms generated by the sound with her device seems to have been mostly aesthetic and yet they are a treasure only recently beginning to be recovered. Be that as it may, Margaret’s contemporaries welcomed with great expectation her presentations and those very similar devices were divulged that was used more as an amusement than for research, as it was known as “tonograph”, referred to in Scientific American on May 29, 1897.
*[Note] These were spores of certain plant species, such as Lycopodium clavatum, used at the time for their volatile nature in fireworks or explosives.