The Neurocalometer

The claim

The first instrument built to read a nervous system rather than a skeleton

The neurocalometer, leased by the Palmer School of Chiropractic in Davenport, Iowa from 1924, was the first commercial instrument built to read the state of a nervous system rather than the position of a bone (Senzon 2018). It was designed by Dossa Dixon Evins, an electrical engineer and Palmer graduate. It was promoted by B.J. Palmer in a lyceum address published that year as The Hour Has Struck. Mechanically it did one thing. It compared skin temperature on the left and right sides of the spine and moved a galvanometer needle when the two sides disagreed. That is a modest engineering claim and an enormous conceptual one.

Ask what it commits you to. If the thing worth measuring is a difference between two sides rather than the location of a single part, then the object of care has quietly moved from structure to state. Palmer never used the word tone in the sense this library uses it. He used pressure, interference, transmission. But the instrument he leased encoded a premise he did not fully name: that a nervous system has a measurable condition, that the condition is expressed in tissue, and that tissue reports it as an asymmetry. The commerce around 1924 was ugly. The physics underneath it was serious. Both are true, and this page holds both.

The inventor

Dossa Dixon Evins built it, and B.J. Palmer sold it

Dossa Dixon Evins invented the neurocalometer. He was born on February 13, 1886 in Blodgett, Missouri, and was dead by 1932, aged around forty six. Before he built anything for chiropractors he performed in vaudeville with his wife Billie, worked as an electrical engineer, and served as a radio operator for the Secret Service during the First World War. He then graduated from the Palmer School (Nash 1995). The Omaha Bee of September 14, 1924, reprinted in Palmer’s own Fountain Head News, dates the laboratory work to 1920 and says the invention was brought to Palmer’s attention in April of the previous year, placing that meeting in April 1923.

Palmer’s account roughly agrees. In the lyceum address he describes an involvement of about eighteen months, ten of them spent convincing himself the instrument was right and roughly nine more spent building the commercial program around it. He credited Evins in public and in print, calling the device the work of a man who had done what physics said could not be done. When a neurocalometer organization was formed at the 1924 lyceum and merged with the National Spinographic Society, Palmer was elected first honorary president and Evins second. The invention was Evins’s. The exclusive right to manufacture and sell it, in the United States and sixteen other countries, belonged to the Palmer School.

The physics

A thermocouple pair, a galvanometer, and nothing else

The neurocalometer was a thermocouple pair read on a galvanometer, and nothing more. The contemporary description is unusually clear on this point. Although the instrument had a Westinghouse appearance, the Omaha Bee reported, it was not an electrical device and was not operated by electricity; it was a self-generating heat-recording device indicating heat differences on a galvanometer. That is the Seebeck effect. Two dissimilar metals joined at a junction generate a small voltage proportional to temperature. Wire two junctions against each other, place them on either side of a spinous process, and the needle deflects only when the sides are unequal. Left deflection, left side warmer. Right deflection, right side warmer. A reading of zero means symmetry.

The reported sensitivity deserves both the number and the caveat. The same article states that tests by the United States Bureau of Standards found the device registered thermal differences of one hundredth of a degree centigrade, and that six patent applications were pending across seventeen countries. That claim reaches us through Palmer’s own weekly reprinting a newspaper item, not through an independent laboratory record, so it should be read as reported rather than confirmed. The rest of the description is easier to credit and more revealing. The needle would register a hand held a foot away, breath at eight inches, and a lit cigar three feet off. An instrument that responsive to ambient heat is an instrument that lives or dies on protocol.

The reframe

The hot box: heat as the signature of resistance

Palmer’s model of the instrument rested on a railroad metaphor, and the metaphor is better than its reputation. A hot box is an overheated axle bearing, a place where friction has replaced smooth transmission and announced itself as temperature. Palmer applied that image to nerves. Where transmission meets resistance, energy that should have moved instead accumulates, and accumulation shows as heat. Whether the mechanism is right in detail is a separate question from whether the reasoning is sound in form. The reasoning is systems reasoning. It treats temperature as a readout of a process rather than as a complaint in its own right.

Look at what that paragraph from 1924 actually does. It defines a lesion as a place of resistance. It predicts a physical consequence of that resistance. It names an instrument to detect the consequence. It sets a target value of zero and calls a spine with all readings at zero a normal one. That is a testable chain, and testable chains are rare in the therapeutic literature of the period. Palmer’s failure was not in constructing the chain. It was in declaring the chain proven before anyone outside his own building had tested a single link of it, and then attaching a price to the announcement.

The evidence

Five hundred cases, ten months, and no published protocol

Palmer’s entire public evidence for the neurocalometer was one sentence describing five hundred cases over ten months. He told the lyceum he had studied the contrast between cases with the instrument and without it, thoroughly, for ten months on five hundred cases, before anybody outside of three or four people knew what he was doing. Take the number seriously first. Five hundred cases is not a trivial series for 1924, and a ten month observation window shows genuine patience in a man not otherwise known for it. He was, on his own account, trying to convince himself before he tried to convince anyone else.

Now ask the harder question. What was recorded, by whom, against what comparison, and where can any of it be read today? None of it was published. There is no protocol, no control condition, no blinding, no independent replication, and no surviving data set. Worse followed. A sample letter for lessees printed in the Fountain Head News of November 8, 1924 instructed practitioners to tell patients that experimental work on approximately a thousand cases showed results obtainable in one quarter to one half the time previously required. Detecting an asymmetry is one class of claim. Halving recovery time is another entirely, and nothing in the record supports it.

The price

The lease that broke the profession

The neurocalometer was never sold, only leased, and the price climbed in public from five hundred dollars to two thousand two hundred within a few months. Palmer traced the escalation himself in the Fountain Head News of August 2, 1924. First five hundred dollars, which people thought stiff. Then a rise to six hundred, and some resentment. Then a short lived seven hundred and fifty, and then, in his words, the storm burst loose. By early September the standing figure was fifteen hundred dollars with a deadline attached. After midnight on September 14, 1924 the price became two thousand two hundred dollars on a ten year contract, terms of two hundred down and one hundred a month at six per cent.

The conditions ran further than the price. The contract obliged lessees to charge patients a minimum of ten dollars for a neurocalometer reading, pushing the price floor all the way to the public. No instrument was delivered until the lessee had completed a two day technique course at the school. Leasing was restricted to chiropractors, and future access was promised to Palmer graduates. By October 18, 1924 Palmer reported over a thousand instruments in actual use within four months. A week later he wrote that a further rise, from two thousand two hundred to three thousand, was quite certain before the new year. He was pricing against a profession he had just told the room he mostly disdained.

The address

The Hour Has Struck is a sales pitch that insults its audience

The Hour Has Struck spends most of its length insulting the people it is selling to. Palmer told the room that about fifty per cent of chiropractors had no ability to be chiropractors and ought not to be, and that a further twenty five per cent could not run an office commercially. He proposed that of twenty thousand practitioners, ten thousand could be discounted outright as unworthy. He described the lease plan explicitly as a filter, a mechanism for culling roughly five thousand acceptable people out of twenty thousand. He said rules are not made for the honest or the worthy, and that the other seventeen thousand had to be governed, instructed and led.

He was equally candid about the money, which reads strangely beside the contempt. He said outright that the school would make money on the instrument and promised to put every dollar back into the profession. He itemized the costs he was covering: a traveling school of ten men at five thousand dollars a month for ten years, a national publicity campaign, a crew of lawyers to defend the patents, and an upgrade of radio station WOC from five hundred watts to five thousand. He compared himself to Lincoln. He forgave his critics twice in one speech. Then he closed by handing his audience what he called the most valuable invention of history, on the grounds that it picks, proves and locates the cause of all dis-eases of the human race.

The rupture

What the announcement actually cost him

The 1924 announcement cost B.J. Palmer control of the profession his father had founded. The American Chiropractic Association, organized in 1922 in opposition to Palmer’s Universal Chiropractors’ Association, swelled as long-standing Palmer loyalists left him (Keating 1996). Palmer answered by persuading the UCA to require a neurocalometer lease as a condition of membership, which turned a professional body into a distribution channel and hardened the objection rather than answering it. In 1925 he resigned as secretary of the UCA and then failed in his bid for re-election. He founded the Chiropractic Health Bureau, forerunner of the International Chiropractors Association, and led it until his death in 1961. He never again held the support of a majority of the profession.

The school paid alongside him. In the first half of the 1920s the Palmer School boasted an enrollment of more than three thousand students, at a time when more than eighty chiropractic schools were operating in the United States. By the end of the decade the Palmer student body had declined by as much as ninety per cent. More than one force drove that collapse, including the expiry of veterans’ benefits that had inflated enrollment across every school in the country. But 1924 cost Palmer the one asset an institution cannot repurchase quickly, which is the confidence of its own graduates. Twenty five instructors were named in print as having bought contracts, a list published precisely because loyalty had become a question.

The record

Evins, not Palmer, invented the neurocalometer presented in 1924, Black Friday is a later name, and the most quoted Palmer line is a compression

Three claims about 1924 circulate widely and do not survive contact with the primary sources. First, Palmer did not invent the neurocalometer. Evins did, and Palmer said so repeatedly in print. The Palmer School held manufacture and sale rights under exclusive license from the inventor, which is a commercial arrangement and not an authorship claim. Second, the label Black Friday is retrospective. It does not appear in the Fountain Head News across 1924, nor in The Hour Has Struck, nor in the standard scholarly chronologies of the episode (Moore 1995). It is a name later memory attached to the rupture. It is useful shorthand and misleading if read as a date.

Third, the most quoted Palmer line about the device is a compression of something he stated differently. In a letter dated March 24, 1941 he complained that one statement had been misrepresented for years, innocently or deliberately. He acknowledged having said that the day would come when a chiropractor could not remain in business unless he used a neurocalometer. He insisted his meaning was that a practitioner has no reliable way of knowing when to adjust and when not to without one, rather than a threat to drive anyone out of business. Whether that distinction rescues him is a judgment each reader can make. That he was still drawing it seventeen years later, and still stinging, is a documented fact.

The verdict

The measurement was real; the interpretation was the weak link

The signal the neurocalometer detected is real; the interpretation placed on it was the weak link, and to a degree still is. Paraspinal skin temperature is under sympathetic vasomotor control, so a stable left to right difference along the spine is a genuine autonomic readout rather than an artefact. Evins built a device that detected it, and detected it well enough to respond to a hand at a foot’s distance. The open question is not whether the asymmetry exists. It is what a given pattern means, and whether two trained examiners looking at the same pattern will report the same thing. That is a reliability question, and reliability questions have answers.

Part of the answer arrived in 2015. Mansholt, Vining, Long and Goertz asked seventeen doctors of chiropractic to interpret paraspinal thermographic scans and published the results in the Journal of the Canadian Chiropractic Association (Mansholt 2015). Overall inter-examiner agreement was fair, at a kappa of 0.43. Where a scan matched a reference pattern completely, agreement was good at 0.73. Where the match was only partial, agreement fell to a range of 0.05 to 0.22, at or near chance. That is the shape of the instrument’s legacy. Clear signals are read consistently. Ambiguous ones are not. It is precisely the study Palmer should have commissioned in 1924 instead of a price list and a deadline.

The inheritance

1924 is the year the state of a nervous system became a number

Nineteen twenty four is the year the state of a nervous system stopped being something a practitioner asserted and became something a needle could contradict. That is the gain, and it survives the commerce completely. Evins was pointing a thermocouple at a spine in the same decade that Otto Loewi was proving that nerves speak in chemistry, and the two experiments belong to one movement of thought: the nervous system as a system with measurable running conditions. Every later instrument that reads it as a condition rather than a picture sits in the same lineage. Surface electromyography, heart rate variability, infrared paraspinal scanning, pupillometry. Each samples a tissue, extracts an asymmetry or a variability, and treats it as a report on regulation.

This library reads that inheritance as tone, the living state of the nervous system expressed as tissue tension at every scale. Palmer did not use that word and did not hold that model; the reading is ours, and we state it as ours rather than putting it in his mouth. What belongs to him and to Evins is harder and more durable than any doctrine. They built a machine capable of telling its owner he was wrong. Set that against the practice it replaced, where a joint that cracked was assumed to have moved and a case that failed had no explanation available, and the scale of the shift is clear. The 1924 answer was crude and oversold. The 1924 question was the right one, and it is still open.

The neurocalometer and the model

What a real tone measure has to do, and where the neurocalometer stopped short

The instinct to instrument the spine was right, and the instrument was partial. The Unified Model of Tone treats the living state of the nervous system as a quantity that can be read rather than asserted. It also holds that every window onto that state is partial and carries its own error, which is why the discipline is triangulation across several of them rather than confidence in any one. A thermocouple pair reading left to right skin heat is one window. Evins built it well. Palmer priced it as though it were the whole view.

The model puts accuracy earlier in the encounter than the professions usually look for it. Specificity is not primarily a property of the hands, the needle, the dose or the instrument. It is a property of the reading that precedes them, because an input can only correspond to a pattern that has first been found. Assessment, and not delivery, is the true seat of accuracy. By that measure the neurocalometer was aimed at the right target. A profession that had been inferring nerve interference from a joint that cracked was handed a needle that could report a difference, and could equally report none.

The model sets two conditions on any measure that claims to read tone. It has to be recordable before the clinical outcome is known, so that the reading is not a label applied after the fact. It also has to carry information about how differently the same event will land on two different people. The windows the model names meet both conditions. They are the variability of a rhythm rather than its mean, the coupling between a slow rhythm and a fast one, the responsiveness of a reflex, and the recovery time after a demand.

The neurocalometer met the first condition and never tested the second. Its needle moved before anyone knew how the case would turn out. Whether the reading predicted how differently two patients would respond is the question the five hundred cases were never published to answer. The model makes that claim testable in one specific way. It predicts that variability structure, cross-frequency coupling, reflex responsiveness and recovery time will share a common underlying factor when recorded together in the same people. They will not vary independently. Loading together on that common factor is the finding that establishes tone as one variable and confirms the claim.

The 2015 kappa figures are where a successor instrument has to start. Agreement was good where a scan matched a reference pattern completely and no better than chance where the match was partial. The model reads that split as the cost of asking one window to carry a whole reading, and it predicts that agreement rises when several windows are read together. Evins built the measurement. The standard it failed is the model’s.

What the record shows

The neurocalometer in figures, from the lyceum floor to the 2015 reliability data

• 1924. Dossa Dixon Evins demonstrated a pair of thermocouples wired against one another and read on a galvanometer, so the needle moved only when the two sides of the spine were unequal (Nash 1995).
• Five hundred cases, ten months. That one sentence was Palmer’s entire public evidence at the 1924 lyceum. No protocol, control condition, blinding or data set was ever published.
• September 14, 1924. After midnight the lease price rose to two thousand two hundred dollars on a ten year contract, terms of two hundred down and one hundred a month at six per cent. The figure had been five hundred dollars months earlier.
• Ten dollars. The contract obliged lessees to charge patients that minimum for a single neurocalometer reading, carrying the price floor all the way to the public.
• October 18, 1924. Palmer reported over a thousand instruments in actual use within four months, and a week later called a further rise to three thousand dollars quite certain before the new year.
• Black Friday is not a 1924 term. The label appears nowhere in the Fountain Head News across that year, nor in The Hour Has Struck, nor in the standard scholarly chronologies (Moore 1995). It is a name later memory attached to the rupture.
• 2015. Mansholt, Vining, Long and Goertz asked seventeen doctors of chiropractic to interpret paraspinal thermographic scans. Overall agreement was fair at a kappa of 0.43, good at 0.73 where the pattern matched completely, and poor at 0.05 to 0.22 where it matched in part (Mansholt 2015).