Electrolysis


Nineteenth century physicians knew that hair grew from a ‘pulp’ at the base of the hair follicle, and that eliminating it would permanently remove the hair. However, the methods they had at their disposal to destroy the ‘germinal papilla’ were crude at best and generally accompanied by noticeable scarring. These included, inserting unsterile needles into the hair follicles, hypodermically injecting carbolic acid, twisting barbed needles (Hinkel & Lind, 1986, p. 150), and heating inserted needles with red-hot curling irons (Wagner, Brysk & Tyring, 1997, p. 949).

A new treatment

Ophthalmologists of the time were also interested in removing hair permanently. Ingrown and other aberrant eyelashes could irritate the eye, resulting in chronic inflammation and even blindness. Dr Charles Michel was one such ophthalmologist trying to remove aberrant eyelashes. He tried heated needles, surgery and twisting needles but found that all produced unsuitable body reactions and scarring. Eventually, he modified a process which had been previously used in general surgery – chemical decomposition through electricity – known as electrolysis or electrocoagulation. He connected a gilt needle to the negative electrode of a battery, inserted the needle into the hair follicle of the eyelash, applied a current for a few seconds and then removed the hair with a pair of tweezers. The germinal papilla of the eyelash follicle was destroyed because sodium hydroxide (lye/caustic soda) was produced at the negative electrode.

The agent employed is electricity, (a constant current battery of 8 to 20 medium sized cells is all-sufficient) the form, electrolysis. I simply pass a fine, gilt needle into the hair follicle and allow the current to produce the electrochemical decomposition of it and its papillae.

(Michel, 1897, as reported in Wagner, Brysk & Tyring, 1997, p. 947)

After his success with eyelashes Michel also used the technique to permanently remove eyebrow hair.

Michel published a report detailing his electrochemical decomposition of hair follicles in the St. Louis Clinical Record in 1875. The editor, William Hardaway, was a dermatologist and decided to try the technique in his own practice. Using Michel’s procedure, he successfully treated patients with excess body hair and presented his results at the second meeting of the American Dermatological Association. After this, other dermatologists took up the practice and the treatment spread. For example, in 1889 a Detroit physician reported he had treated over fifteen hundred cases of superfluous hair with electrolysis (Herzig, 2008, p. 869).

We know nothing of “Miss X” save what her physician, W. A. Hardaway, recorded in 1877. Twenty-two years old when she came under Dr. Hardaway’s care, Miss X was “thoroughly feminine” in character and physique, nicely plump, and robustly healthy. She was also the “unfortunate owner of a beard that for strength and luxuriance rivaled the hirsute appendages of any man.” Previous efforts to remove her beard had not gone well: after the application of depilatory powder the beard grew back “thicker and more profuse” than before. … Eager to be of service to the young woman, Hardaway decided to tackle the “luxuriant” but unwanted beard with an experimental procedure: electrolysis. … Moving strand by strand in this fashion, Hardaway and a colleague removed the entirety of Miss X’s “appendage.” The two men worked on the young woman's face for an hour or two at each session, as many as nine sessions per week—more than 350 treatments in total. Despite the procedure’s tedium, Hardaway concluded that this “radical cure of hirsuties” is “brilliant in its results”.

(Herzig, 2008, p. 867)

Early electrolysis machines

The electrolysis machines developed by Michel, Hardaway and others were battery operated, generating what medical practitioners of the day called a ‘galvanic current’, named after Luigi Galvani [1737-1798] – known for using electricity to produce muscle contraction in frog’s legs. Many nineteenth century physicians were familiar with galvanic batteries, galvanic induction coils and electro-therapeutics, so it was relatively easy for them to set up working electrolysis machines. If the necessary parts were not available in their surgery, they could easily obtain them from any one of a number of suppliers.

Eventually, manufacturers combined all the required parts of an electrolysis machine into complete kits and advertised them in their catalogues. These machines were made up of battery cells along with all the necessary cords and electrodes. To make them easier to use, detailed instruction manuals were often included which outlined the process involved to remove hair by electrolysis.

Schall & Son London
No 675. Complete set for epilation, consisting of 9-cell battery with collector inserting the cells one by one bracelet electrode No. 412, needle-holder No. 664, forceps No. 666, a packet of needles and collecting wires £3.0.0
Explicit directions for use are sent with this outfit.

(Schall & Son, 1913)

Commercial operators

Manufacturers were naturally keen to sell as many electrolysis machines as possible; mains electricity was absent from most cities in the early part of the twentieth century but as the equipment was battery powered, and very portable, this was not an issue. In addition, their operation was not regulated so, before long, their use spread outside the medical profession. Attitudes of some physicians helped. Although some could see the distress that excess hair was causing their patients, many saw electrolysis simply as a beautifying practice – as the correction of a ‘cosmetic defect’ rather than a cure for a ‘serious disease’ – and discounted it as a medical procedure, leaving it to others to provide the service.

By the end of the nineteenth century, electrolysis treatments could be readily obtained from non-medical sources including specialist operators, as well some barbers, hairdressers and beauty salons. As physicians discovered other uses for electrolysis, the non-medical operators followed and were soon using their machines to treat other facial blemishes such as moles, warts, spider veins, birthmarks, pimples, blackheads and acne. Newspaper advertising of the time indicates that there were often many operators working in major cities and it is more than likely that some of these branched out and became Beauty Culturists as well.

Superfluous hairs on the face can only be really eradicated by electrolysis, and those who wish to undergo this little process should always be most careful to go to a qualified person. I can most thoroughly recommend Mrs. Pomeroy, 29 Old Bond Street, for the removal of these and other facial blemishes in a skilled and competent manner.

(Browning, 1898, pp. 216-217)

By the 1940s, the medical profession was retreating from removing hair by electrolysis completely, and in second half of the twentieth century it became increasingly a non-medical procedure (Herzig, 2008, p. 878).

Operator skill

Despite the helpful literature supplied by manufacturers, practitioners (both medical and commercial) soon realised that using the machines to produce satisfactory results without scarring or pitting the skin was not a simple matter – skill and experience counted.

In no operation where human life is not involved does experience count for more than it does in this comparatively simple and easily executed procedure.
Whenever it is possible to watch the operation as performed by an experienced cosmetiste, it is advisable that this should be done. There are many little points in the technic of the operation that contribute materially to its success which can be better mastered by observing the operation than reading any description of it.
It is recommended that the beginner attempt the removal of hair on portions of his or her own body rather than the face and that no attempt be made on the face until the student is sure that he understands at least the working principles of electrolysis. Only partial experience can be gained in this manner however.

(W. M. Meyer, 1936, p. 349)

Concern about the expertise of operators came from a number of areas; legislators, manufacturers, the medical profession and commercial operators were all interested in this area. In France, commercial operators were barred and patients were required to seek medical assistance. In the U.S. and Britain the situation was more open. As commercial operators became more common, some authorities began to regulate the practice; however, this varied from country to country and from state to state. The arrival of professional associations and training schools helped set standards and provide supervised training programs. Manufacturers were also an important source of training. For example, when multiple needle electrolysis machines were introduced, the company set up to manufacture them also trained operators in their use.

Limitations of electrolysis

Although electrolysis was a significant improvement over previous treatments it had a number of problems. Some of these were due to the technical limitations of the equipment used, others were inherent to the method. As well as introducing the possibility of infection, the process was very slow, painful, could produce noticeable scars (if rushed or done incorrectly) and/or create pigmentation problems in the treated areas of some clients. Patient forbearance, variation in skin and hair types, operator fatigue, acuity of vision and the cost of treatments were all factors affecting the successful outcome of a treatment regime. The fact that patients endured the drawbacks gives us an idea of the measures they would undergo in order to rid themselves of the hair.

Treatment times: The process was inherently slow. Even skilled operators had to wait for the production of sufficient sodium hydroxide to destroy the papilla. The process could be speeded up with a stronger current but then the risk of scarring increased – an ongoing problem. Removing the needle too early would only lead to regrowth.

An experienced operator could probably average removing about 60 hairs an hour, although this depended on whether hair was being removed from lip, cheeks, chin or neck. Patients with difficult hairs would have lower removal rates.

One way around this problem was suggested by the father of demabrasion, the German dermatologist Ernst Kromayer [1862-1933] in 1908 – use multiple needles.

Usually electrolytic epilation is performed with one needle at a time, but he [Kromayer] sees no reason why several may not be employed simultaneously. He therefore has his needles connected with 15 cm. of the thinnest copper wire, which can be brought in contact with the copper wire of the other needles. Having formed a bundle of such needles,he seizes each one in rotation by artery forceps and inserts it into the hair follicle. When all the needles required are in place the wires are connected with the battery and the current turned on. In applying a number of needles simultaneously it is necessary to adjust the current according to the number of needles acting. It has been found that a current of 5 milliamperes destroys a hair completely in one minute. The complete destruction of the hair can be ascertained by pulling lightly on the hair shaft. The hair will be found to be held tight. On pulling somewhat more strongly, the hair comes away and is found to end sharply, as if cut off. No trace of the hair bulb may be found,as this should have been destroyed by the electric current. Before the process is complete a hardening around the needle is felt, and soon a definite nodule is seen. If this nodule begins to be transparent, it is time to remove the needle, as there is danger of the cutis becoming involved in the necrosis and of the skin “burning through.” If five needles are applied at one time the current will be divided into fifths, and it will therefore be necessary to allow a 25 milliampere-minute current to act. It is wise to apply the strongest current well tolerated, and the author usually begins with 1 milliampere, and increases this up to 5, 10, 20, and 40. The highest he has used was 50 milliamperes. It must be remembered that the stronger the current the shorter will be the time during which it is necessary to apply the current to destroy the hair bulb. … At times he finds it wiser to use only a few needles, for example five, and to apply them continuously. After one has done its work on one hair bulb it is withdrawn, and is inserted into another follicle, and so on. In this way there are always five needles acting.

(An epitome of current medical literature, 1909, pp. 83-84)

This idea was taken up by Professor Paul N. Kree who patented a multiple-needle electrolysis machine in 1918 (US: 1445961) and established a business to manufacture it. However, for individuals with severe problems, treatments could still go on for months or longer.

See also: The Drilling Machine

Scarring: Applying a current for too long could result in the development of a visible scar. Practitioners would sometimes curtail the application of electricity but this could reduce their success rate and lead to regrowth. Ernst Kromayer had a suggestion for this as well and recommended that insulated needles should be used.

Kromayer states that the ordinary method of destroying hairs by electrolysis has the disadvantage of producing a scar which remains visible. In order to limit this unsightly appearance it is usual to interrupt the process before the hair is completely destroyed, and in this way so-called “recurrences” are frequently seen (Deut. med. Woch., December 24th, 1908). Even when the superficial layer of the hair bulb is destroyed the hair may grow again. He therefore considers that it would be better if the whole process could be carried out subcutaneously, so that it would not be necessary to interrupt the action until the hair was completely and finally destroyed. For this purpose he has constructed electrolytic subcutaneous needles, which are coated up to within from 2 to 10 mm. of the point with varnish. This coating is so thin that it does not hinder the introduction of the needle.

(An epitome of current medical literature, 1909, pp. 83-84)

Time: The length of the treatment session was also a limiting factor. Sessions were generally between a half and a full hour. Even if the patient could continue, treatments rarely lasted longer, as the operator would become fatigued. This meant that very few hairs were removed at each sitting. Given that some removed hairs grew back, patients were often discouraged by a perceived lack of progress and it was necessary to reassure them that progress was being made.

As has already been hinted, there occur times of discouragement, and these times are sure to come if you have not forewarned your patient that some hairs will return. You must remember that your patients are morbid on the subject of what they consider a facial blemish; so much so that they are determined to have you remove all of the lanugo hairs that may be found upon the face. Again, they forget their appearance when they first came to you.
I have had patients come into my office and tell me that the work was a failure, that all the hairs removed had returned, and they, to say the least, were very much dissatisfied; almost without exception I have been able to convince them that their condition was much better than when they first came to me, and that it would require but comparatively little work to make the operation a complete success. These patients have subsequently become my warmest supporters.
Were it possible to prevail upon patients to have a photograph taken before the operation of the part of their face where the growth had been, the comparison would aid very materially in overcoming the discouragement. In every case we can absolute promise that eventually every hair will be permanently removed.

(Hayes, 1904, pp. 81-82)

Pain: Patients showed a range of tolerance to pain. The pain generated also varied depending on the area of the body undergoing treatment.

In early machines, the galvanic current was produced by cell batteries and regulated not by a dial (rheostat) but rather by connecting more or fewer cells to the circuit. Voltage regulation was limited and the intensity of current used was higher and therefore more painful. Later machines did show improvements with the addition of regulating devices (rheostats), amperage meters and the use of mains electricity but there was no way to avoid the production of sodium hydroxide and the associated pain.

Needles were also an issue. Unlike today, they were not disposable. Platinum and gold needles were expensive and tended to deteriorate with reuse, so steel was preferred. Some commercial operators used sewing needles to reduce costs but even needles purchased though supply companies were thicker than those currently in use, adding to patient discomfort.

Some physicians tried topical anæsthetics such as cocaine to reduce the pain with mixed results.

Dr. A. D. Rockwell, of New York, said … he had employed cataphoresis extensively in the removal of hair, and had found that by using cocaine cataphorically this ordinarily uncomfortable treatment could be applied without causing any pain. Recently, however, during the application of cocaine in this way the patient suddenly became unconscious, and was at first thought to be dead, although she eventually recovered. This had made him hesitate to use cocaine as frequently as formerly.

(Transactions of the American electro-therapeutic association, 1899, p. 89)

Commercial operators also tried to minimize the pain but generally steered clear of local anaesthetics.

Should the patient be very sensitive to the electrical current, less pain is produced when the needle is introduced before the current is completed. The patient is then directed to put fingers into the solution, and before the needle is removed, is requested to remove the fingers from the solution. By this means, the slight shock on making and breaking the current is much less than when the needle is introduced without the current having been broken … The needle is however more easily inserted if a small amount of current is on while inserting.

(W. M. Meyer, 1936, p. 346)

Cost: In the early part of the twentieth century a commercial operator in the U.S. might earn between six to ten dollars an hour, more than a day’s pay for many other workers (Herzig, 2008, p. 871) making the treatments out of the reach of many. As treatments could extend over many months some women were also forced to abandon them before completion and return to shaving, depliators, tweezing and other methods of removing unwanted hair, simply because they ran out of money.

See also: Chemical depilatories

Cost pressures also had effects on the operators. This was particularly so in the early history of electrolysis, when the technique was slowest. When working on wealthy women, operators could take more time and use lower current levels, thereby reducing the risk of scarring.

“[P]atients whose time and money are limited,” wrote Dr. Sorenson in 1893, “will urge the doctor to take out hairs that are too close together to be operated upon with good results in one day, or even on two successive days … Affluent women were easier to treat than poor women, another physician argued at a gathering of dermatologists, since “We could take our time in treating these cases, using an amperage of low intensity … with less tendency to scarring.”

(Herzig, 2008, p. 876)

Poorer patients tended to pressure operators to use higher current flows and traded off an increased chance of scarring for a larger number of hairs being treated at each session. Fortunately, as the levels of prosperity rose in the twentieth century and new technologies for hair removal arrived this trade off became less of an imperative.

One way to reduce the cost was to treat yourself. Beginning in the 1920s a number of electrolysis units designed for home use began to appear on the market. Devices such as the ‘Beautiderm Midget’ appealed to users who wanted to avoid the cost of a salon treatments or were too far away from a major centre to make regular treatments practicable (Herzig, 2008 p. 877). Home electrolysis treatments have continued right up until today but have a number of problems. First, even if you have a stready hand, good eyesight and the resolve, it is not easy to treat yourself in front of a mirror; results are always better if someone else operates the machine. Secondly, the machines produced for the home market are never as effective as those used commercially. Today, self-treaters recommend the purchase of second-hand commercial machines.

The issues associated with permanent hair removal by electrolysis – cost, pain and time – meant that many women had their hair removed by X-rays, with disastrous consequences.

See also: X-rays and Hair Removal

One other problem was skin discolouration caused by black deposits produced by a chemical reaction between hydrochloric acid and the metal in the needle if, by chance, the operator used the wrong polarity. The problem was easily avoided by using needles made from gold or platinum alloy, or stainless steel, but some electrologists in the past failed to do this and mistakes happened. Fortunately, this is not an issue today.

Enelectrolysis

In 1905, in an article to the Lancet, Balmanno Squire outlined an alternative method to the standard procedure. Rather than inserting the needle, applying current, and then removing the hair, he suggested that the hair should be pulled out first and the needle then inserted down the hole left after the hair was extracted. He called his method enelectrolysis. His observations that it less was likely to puncture the wall of the follicle must be tempered by the fact that he used the blunt end of a No. 12 sewing needle during enelectrolysis, whereas doing electrolysis, he had been using the sharp end. His statements that it was more expeditious must also take into account that he suffered from myopia, which presumably made it easier for him to identify the opening to the hair follicle once the hair had been removed. Needless to say his method did not catch on.

Decline

In the 1940s, reliable high frequency machines came into operation and the practice of electrolysis began to wane. However, it did not die out and some current practitioners still prefer it. The new high frequency machines produced hair loss by heat (thermolysis/diathermy) rather than by chemical decomposition (electrolysis) and removed hairs at a more rapid rate. Unfortunately, this new technique also had its problems.

See also: Thermolysis and the Blend

Updated: 22nd February 2016

Sources

An epitome of current medical literature. (1909) The British Medical Journal. May 22, 1(2525), 81-84.

Browning, H. E. (1898). Beauty culture. London: Hutchinson & Co.

Foan, G. A., & Bari-Woolls, J. (Eds.). (n.d.). The art and craft of hairdressing: The standard and complete guide to the technique of modern hairdressing, manicure, massage and beauty culture (2nd ed.). London: New Era Publishing.

Fox, G. H. (1886). The use of electricity in the removal of superfluous hair and the treatment of various facial blemishes. Detroit, MI: G. S. Davis.

Hayes, P. S. (1904). Electricity and the methods of its employment in removing superfluous hair and other facial blemishes. Chicago, Ill: McIntosh Battery and Optical Co.

Hinkel, A. R., & Lind, R. W. (1968). Electrolysis, thermolysis and the blend: The principles and practice of permanent hair removal. Los Angeles, Ca: Arroway.

Herzig, R. (2008). Subjected to the current: Batteries, bodies, and the early history of electrification in the United States in Journal of Social History. Summer; 41, 4. Academic Research Library.

Moler, A.B. (1906). The manual on barbering, hairdressing, manicuring, facial massage, electrolysis and chiropody as taught in the Moler system of colleges.

Müller, R. W. (1916). Hair. Its nature, growth, and most common afflictions, with hygienic rules for its preservation. New York: William R. Jenkins Company.

Schall & Son. (1913). Electro-medical instruments and their management and illustrated price list of electro-medical apparatus. (15th ed.). Stonebridge: John Wright & Sons Ltd.

Squire, B. (1905). Enelectrolysis: An improved method of operating on superfluous hair. The Lancet. 165(4252). 493-494.

Mould, R. F., Robison, R. F., & Van Tiggelen, R. (2010). Louis-Frédéric Wickham (1861-1913): Father of radium therapy. Journal of Oncology. 60(4), 79e-103e.

Transactions of the American electro-therapeutic association. (1899). Buffalo, NY: Author.

Verni, M. (1946). Modern beauty culture (2nd ed.). London: New Era Publishing.

Wagner, R. F., Brysk, H., & Tyring, S. K. (1997). Revisiting the Michel/Green controversy of 1879: Was Carron du Villards the first to use probe/needle electrolysis for permanent hair destruction? International Journal of Dermatology. 36. 947-951.

W. M. Meyer Co. (1936). The cosmetiste: A textbook on cosmetology with special reference to the employment of electricity in the care of the hair, scalp, face, and hands, also permanent waving and hair curling. Chicago, Ill: Author.