Tousey-Sinclair: Primer

Project: Sinclair Tousey (born 1864, died about 25 Sept 1937), Medical Electricity, Rontgen Rays and Radium, W. B. Saunders, third edition, April 1921 (first edition 1910, second 1915); 1361 PDF pages, one library-copy scan [read: PDF p.7-8; dates of life from the NYT obituary of 26 Sept 1937, [web] via the Investigator, research/2026-10-08-nytimes.md].

What kind of project this is. A physician's textbook, not a free-energy tract. Tousey was a New York surgeon who became an x-ray and electrotherapy specialist, held seven US patents on x-ray tubes, filters and film carriers (1911-1925) [web: Google Patents, via the Investigator, research/2026-10-08-google-patents.md], and compiled the electrical medicine of his day into one book. Most of the book reports other workers; sentences credited to "the author" are his own. The book sits in an Energy folder because static machines, Tesla coils and Oudin resonators interest apparatus builders (Energetic Forum threads use it as a reference [web: research/2026-10-08-google-search.md]). It makes no over-unity claim: it has two "output greater than input" passages, and Tousey explains both himself. On PDF p.156 he measured an "apparent power of 6000 watts produced by the application of about 600 watts" and put it down to the voltage collapsing once an arc is running [read: PDF p.156]; on PDF p.741 he writes "Of course this would be an impossibility" [read]. These passages are the worked example below, because they are exactly the measurement trap an energy researcher has to recognise.

Revision note (2026-10-09). This primer was updated after the quality check (qe.md) and the arbiter's "QE revision" (arbiter.md, R1-R12). The first version credited Tousey with two positions the book actually rejects: the condenser theory of x-ray burns, and germicidal ultraviolet passing through the glass of a vacuum electrode. It also missed his own p.156 experiment, the book's own dose conversions, Apostoli's 500 diabetes cases and Gay's measured cases. Sections 1, 4, 5, 5b, 6 and 7 are revised, and the changes are listed under "Corrections to earlier documents" at the end.

Source limits (repeated from Step 1 and the arbiter). - Every page is OCR text from a scan. The OCR prints "x-ray" as "z-ray". Where tables or formulas were garbled, the deep readers read page images. - The folder's scan lacks printed pp.282-283, 286-287 and 290-295. The Investigator found a second scan of the same edition (Internet Archive, medicalelectrici00tousiala, UCLA copy). Its text holds pp.282-283 (electrotonus) and p.287 (the membrane "concentration cell" passage). The reviewers read those pages only as noisy OCR; pp.290-295 are still unread. - Read only as digests (1-3 sentence summaries per section): static electricity, most of dynamic electricity, and the radiographic technique chapters. The QE remediation added deep reads of the coil and meter pages (PDF 144-217), physiologic effects (297-349), high-frequency currents and diathermy (544-647), violet and ultraviolet light (557-681), the Cabot and Shaw generator (749-773), x-ray effects, burns and dosage (1130-1180) and radium (1252-1297) [read: deep/...42fec2.md, as weighed in arbiter.md R1-R8]. The high-frequency evidence is now read in full. It remains low confidence as evidence of efficacy, because none of it is controlled. - Everything about Tousey's own patients comes from his own text. No case records survive in the project.

Tags: [read] = from the book or the Investigator's saved pages; [web] = Internet source with URL; [memory] = general knowledge, with confidence; [calc] = arithmetic done for this primer (reproduced in example.py). Page numbers are PDF pages unless "printed" is stated; the printed page is 2 to 17 lower depending on the chapter.

---

1. The idea in one paragraph

Around 1900-1920, doctors had three new tools: electric currents of every kind (steady, pulsed, and very fast alternating "high-frequency" currents from Tesla and Oudin coils), x-rays, and radium. Tousey's book is a working manual for all three, written by a practitioner who also designed apparatus. Its own ideas are practical ones: a standard x-ray dose that he says never burns the skin; a contact x-ray tube (he reports a rival's theory that x-ray burns are an electrical "condenser" effect, cites the experiments against it, and rejects it as the cause while allowing it "a certain element of truth"); high-frequency currents combined with x-rays as a treatment for many diseases; dental x-rays to find hidden tooth abscesses, which he believed caused disease elsewhere in the body, and a formula that predicts the size of a child's adult dental arch from one unerupted tooth; a careful account of scattered x-rays; and an explanation of why his meters seemed to show his coils putting out more power than they took in. Some of this was right and became standard (dental radiography, anti-scatter devices, the constant-voltage x-ray supply, the explanation of the meter paradox). Some was period belief that later evidence overturned (focal infection as a cause of ulcer and cancer; Geyser's condenser theory of burns, which the book itself reports as contradicted by experiment). Much of the clinical material was never tested by a controlled trial and now cannot be.

---

2. What you need to know first

Electric power: peak, average and rms. The power delivered to a device is P = v(t) i(t) at each instant, where v is the voltage across it and i the current through it. What matters for energy is the time average, written ⟨v i⟩ (watts, W). For a steady current P = V I. For pulses or alternating current, "the current" can mean three different numbers: the peak (largest value), the average (area under the curve divided by time), and the rms (root-mean-square: square it, average, take the square root; this is the value that gives the right heating in a resistor, P = I_rms² R). For a sine wave, rms = peak/√2 ≈ 0.707 × peak. For short pulses the three differ a lot. Read: any introductory circuits text on AC power; Horowitz and Hill, The Art of Electronics, chapter 1.

Reactive power and power factor. In a capacitor or inductor the current is 90 degrees out of step with the voltage. Energy flows in during one quarter-cycle and back out in the next, so the time-average power is zero even though both V and I are large. For sine waves, P = V_rms I_rms cos φ, where φ is the phase angle between voltage and current and cos φ is the power factor. The product V_rms I_rms (with no cos φ) is the apparent power, measured in volt-amperes (VA), not watts. Read: Griffiths, Introduction to Electrodynamics (circuits are only touched there); better, any electrical engineering text on AC circuits (e.g. Hughes, Electrical and Electronic Technology).

Meters of 1910. A moving-coil galvanometer (d'Arsonval type) deflects in proportion to the average current, and with current in both directions it reads the net average. A hot-wire meter measures the heating of a thin wire, so it reads rms current whatever the direction. A spark gap measures voltage by the gap length at which air breaks down, so it gives the peak voltage. [memory, high] Tousey describes these meters himself: the D'Arsonval meter "indicates in a general way the average strength and direction of the secondary current" and may read zero while the tube makes a good picture (PDF p.155); the hot-wire meter is "not influenced by the amount of inverse discharge" (PDF p.156) and on alternating current reads "somewhat more than the true" average (PDF p.214) [read; also PDF p.741].

How x-rays are made. An x-ray tube accelerates electrons through a voltage V (tens of kilovolts, kV) into a metal target. The electrons slow down in many collisions and radiate a continuous spectrum of photons called bremsstrahlung ("braking radiation"), plus a few sharp lines characteristic of the target metal. No photon can carry more energy than one electron brings, so the spectrum ends sharply at E_max = eV (e = electron charge); this is the Duane-Hunt limit (1915). The x-ray output rises roughly as (tube current) × V². Only about 0.5-1 percent of the electron energy becomes x-rays; the rest is heat [memory, high]. Read: Bushberg et al., The Essential Physics of Medical Imaging (chapters on x-ray production and x-ray tubes); Johns and Cunningham, The Physics of Radiology.

Radiation dose and the inverse-square law. The absorbed dose is the energy absorbed per kilogram of tissue; its unit is the gray (Gy = 1 J/kg). From a small source the intensity falls as 1/d², d being the distance, because the same energy spreads over a sphere of area 4πd². A tube's output is often quoted per mAs (milliampere-second: tube current in mA times exposure time in s, i.e. the charge passed through the tube). Skin reddening (erythema), the first sign of radiation dermatitis, appears above about 2 Gy in one exposure [web: https://iaea.org/resources/rpop/health-professionals/radiology/erythema]. Before about 1928 doses were measured with pastilles that change colour under x-rays: the Holzknecht unit "H" and the Sabouraud-Noiré "Tint B" (about one erythema dose). Observers disagreed on these by a large factor, and no reliable conversion to gray exists [web: https://pubs.rsna.org/doi/full/10.1148/6.3.236, via the arbiter]. Read: Hall and Giaccia, Radiobiology for the Radiologist (skin effects); Bushberg et al. (dosimetry).

Scattered x-rays. Inside the body, many x-ray photons are not absorbed but bounced off electrons in a new direction (Compton scattering). Scattered photons reach the film from all directions and add a uniform fog; the useful image is the "shadow" made by the different absorption of the straight-through (primary) beam. A grid of thin lead strips (the Bucky grid, 1913) passes the primary beam and blocks most scatter, at the cost of a longer exposure [memory, high]. Read: Bushberg et al. (chapter on scatter and grids).

Ions in water, Faraday and Nernst. Body fluids are salt water; current in them is carried by ions (Na⁺, K⁺, Cl⁻). An ion in an electric field E drifts at speed v = μE, where μ is its mobility (for Na⁺ in water, μ = 5.2 × 10⁻⁸ m²/(V·s) [web: https://en.wikipedia.org/wiki/Electrical_mobility]). Faraday's law of electrolysis: a charge Q decomposes Q/F gram-equivalents of substance at an electrode, where F = 96,485 C/mol is the Faraday constant. A membrane that lets one ion through but not another develops a voltage given by the Nernst equation, E = (RT/zF) ln(c_out/c_in), with R = 8.314 J/(mol·K) the gas constant, T the absolute temperature, z the ion's charge number, and c_out, c_in the concentrations outside and inside. This is the origin of the nerve cell's resting voltage of about −70 to −90 mV [memory, high]. Read: Hille, Ion Channels of Excitable Membranes (early chapters); Nelson, Biological Physics.

Radio-frequency heating. Above roughly 10-100 kHz, alternating current no longer excites nerves and muscles (the membrane cannot charge and discharge fast enough), but it still heats tissue by resistance (Joule heating). This is the basis of diathermy and electrosurgery [memory, high]. The heating per kilogram is called the specific absorption rate (SAR), in W/kg.

---

3. The textbook view

3.1 Power in pulsed and alternating circuits

The real power is always

P = ⟨v(t) i(t)⟩ = (1/T) ∫₀ᵀ v(t) i(t) dt,

where T is one period of the waveform. Two short cuts are commonly misused:

• P ≠ V_peak × I_reading. If v is measured at its peak (spark gap) and i with any meter, the product overstates the power whenever the voltage is lower while the current flows.
• P ≠ V_rms × I_rms unless voltage and current are in step. In general P = V_rms I_rms cos φ.

For a train of current pulses that each look like half a sine wave of peak I_p and occupy a fraction D of the time (the duty cycle), integration gives [calc, example.py Part A]:

I_avg = (2/π) D I_p, I_rms = I_p √(D/2), I_rms / I_avg = π / (2√(2D)).

At D = 0.05 (current flowing 5 percent of the time) the peak is 31 times the average and the rms is 5 times the average. A hot-wire meter on such a current reads five times the average current.

3.2 X-ray output, dose and distance

Approximate rules (modern, tungsten target) [memory, high]: - x-ray output ∝ I_tube × V², where I_tube is the tube current and V the tube voltage. Tousey states the same rule: intensity "varies directly in proportion to the milliamperage and to the square of the voltage" [read: PDF p.741]. - Output per mAs for a modern tube near 70 kV with 2-3 mm aluminium filter: about 0.023-0.030 mGy per mAs at 1 m [web: https://www.rug.nl/education/courses/other-education/radiation-protection/edu/ts-mt/ts-mt-data-examen/mcts-mtfiguura?lang=en, via the arbiter; treat as ± factor 2]. - Dose at distance d: D(d) = D(1 m) × (1 m / d)², plus a small extra loss from absorption in air (about 1-2.4 percent over 25 cm for the soft part of the beam [calc in arbiter D3 and R9, coefficients from NIST XCOM: https://physics.nist.gov/PhysRefData/XrayMassCoef/ComTab/air.html]). - At the skin, scattered radiation from the tissue below adds about 30 percent ("backscatter") [memory, medium]. - The spectrum is continuous up to E_max = eV whatever the supply; a constant voltage fixes the end point but does not make the beam one wavelength [memory, high].

3.3 What causes x-ray skin damage

Biological damage follows absorbed dose. Kienböck's experiments in 1900 showed that "X-ray-dose, not electric phenomena, was the active agent" in skin reactions [web: https://pubmed.ncbi.nlm.nih.gov/25012643/, abstract of Widder 2014, via the Investigator]. Tousey's book reports the same conclusion: "The experiments of Strater, Kienböck, Oudin, and Scholtz seem to leave no room for doubt that radiodermatitis is due to the x-rays and not to electric discharges and other physical phenomena which take place around the tube" [read: PDF p.1158]. Damage accumulates: repeated exposure gives chronic changes (keratoses) and can lead to skin cancer [memory, high].

3.4 Electricity in tissue

• Bulk conduction is ionic. Ions drift slowly: Na⁺ at 1 V/cm (100 V/m) moves v = 5.2 × 10⁻⁸ × 100 = 5.2 × 10⁻⁶ m/s, about 1.9 cm per hour [calc].
• But the field in a conductor is set up almost instantly. Any excess charge spreads out in the charge relaxation time τ = ε/σ, where ε is the permittivity and σ the conductivity. For tissue (relative permittivity about 80, σ about 0.5 S/m [memory, medium]), τ = 80 × 8.85 × 10⁻¹² / 0.5 ≈ 1.4 × 10⁻⁹ s [calc]. Current starts everywhere in the body within nanoseconds; the ions each move only a little.
• Nerves and muscles are excited by charging the cell membrane, which acts as a capacitor: the membrane voltage changes until channels open (Hodgkin and Huxley, 1952) [memory, high]. Electrolysis (chemical decomposition) happens at the electrodes, mainly with direct current.
• Nernst example: K⁺ at 5 mmol/L outside and 140 mmol/L inside, T = 310 K: RT/F = 26.7 mV; ln(5/140) = −3.33; E = −89 mV [calc, from the physicist and arbiter].

3.5 Radio-frequency currents in the body

At frequencies of hundreds of kHz there is no nerve stimulation, only heating. The physicist's estimate for a patient inside Tousey's "autoconduction cage" (a large coil carrying high-frequency current), assuming a field of 1 mT (10 gauss) at 500 kHz and a body radius of 0.15 m: induced field E = ωBr/2 = 236 V/m, SAR = σE²/(2ρ) ≈ 14 W/kg with σ = 0.5 S/m and density ρ = 1000 kg/m³ [calc, physicist, confirmed by the arbiter; the inputs are assumptions]. That is strong whole-body heating (modern whole-body exposure limits for workers are about 0.4 W/kg [memory, medium]).

---

4. Where Tousey departs

Tousey is not an outsider attacking the physics of his day. He mostly reports it, often accurately, and he hedges his own theories in print. The departures are in a few specific places. In each row below, the left column is what he wrote, the right column is the textbook position as the arbiter weighed it.

The common thread. Tousey's starting points are good physics: power is voltage times current, x-ray output grows with current and voltage squared, intensity falls with distance squared, tissue is a salt solution, ions move slowly. Where he goes wrong, it is mostly by letting a correct statement in one chapter sit beside a looser one in another without reconciling them (180,000 mi/s on p.289, a conditional 39 m/s on p.296; the D'Arsonval meter reads the average on p.155, "the meters indicate the maximum" on p.741; ultraviolet "hardly" acts on p.578, a germicidal ultraviolet effect in acne on p.631; the condenser theory "not ... probably correct" on p.1162, "a certain element of truth" on p.1231), or by treating his clinical impression as a measurement ("never", "always") with no count of cases. Several positions a quick reader would pin on him (the condenser theory, ultraviolet through glass, the 39 m/s premise) are ones the book reports from others and argues against.

---

5. Worked example: the power ledger of an x-ray coil

The question. Tousey writes [read: PDF p.740-741]:

"Based upon the calculation of 100,000 volts and 2 milliamperes the power generated by an x-ray coil would be 500 watts, and with 100,000 volts and 15 milliamperes the apparent result would be 1500 watts. But the power of the primary or exciting current ... may be only about 500 watts and 1500 watts respectively, and since we know that nearly half of the power in the secondary is wasted in the form of inverse discharge which must be suppressed, there is apparently greater output than intake of power. Of course this would be an impossibility, and the explanation is found in the fact that the meters indicate the maximum of the successive currents rather than their average strength."

(The OCR is clipped at the left margin; the words are restored from the context, as the deep reader did from the page image.) Is his explanation enough to balance the books? This is the same check anyone should do when a meter says a device puts out more than it takes in. The script example.py reproduces every number.

Step 1. Check the arithmetic first. [calc, example.py Part B] - 100,000 V × 0.002 A = 200 W, not 500 W as printed. - 100,000 V × 0.015 A = 1500 W (correct). - On the same page, a transformer delivering "approximately 1 horse-power or 746 watts" to a Coolidge tube at 56,000 V and 10 mA: 56,000 × 0.010 = 560 W.

Both slips make the apparent output look larger. Lesson: before explaining a paradox, multiply the numbers again.

Step 2. What each instrument actually reported. The voltage, 100 kV, comes from a spark-gap reading, i.e. the voltage at which the air breaks down: a peak. (The book's own transformer table puts a 6-inch spark at 70 kV [read: PDF p.740], so "100 kV" is an even longer spark or a different convention.) The current, 15 mA, comes from a meter, and Tousey says "the meter shows approximately the maximum amperage without very much regard to the average strength" and names two kinds in use: "Hot-wire meters indicate the current strength regardless of its alternating polarity, while electromagnetic meters (galvanometers) are of greatest value when ... the impulses in one direction are practically suppressed" [read: PDF p.741]. The book does not say which meter gave "15 mA". Note that the p.741 sentence ("the meters indicate the maximum") is looser than Tousey's own meter chapter, where he writes that a D'Arsonval meter "indicates in a general way the average strength and direction of the secondary current" [read: PDF p.155]. So he knew an averaging meter would not read the peak; the explanation that works for every meter is the one in Step 3b.

Step 3. Correct the voltage. Tousey measured, with a spark gap on the Cabot direct-current converter, that "in every case there was a drop to less than one-half the voltage during the passage of the spark" [read: PDF p.740]. Apply that by analogy to the coil (an assumption: it was measured on a different machine): while current flows, the tube sits at no more than 50 kV.

Step 3b. Tousey's own experiment: the arc voltage collapses. Nearly six hundred pages earlier the book reports an experiment of Tousey's own [read: PDF p.156]. A 12-inch induction coil with a Wehnelt interrupter, 6 A at 110 V in the primary (660 W, "about 600 watts" after interrupter losses), drew a continuous 4-inch flame between its terminals, with a hot-wire milliammeter (rated to 650 mA) on one pole:

"a voltage of 40,000 and the milliamperemeter indicated an intensity of 150 milliamperes. Here there was an apparent power of 6000 watts produced by the application of about 600 watts ... The apparent discrepancy is due to the fact that the voltage is greatly reduced during the passage of the continuous flame, just as 100 volts will maintain an electric arc 1 inch long which it would take 10,000 volts to originate."

Check it with energy conservation [calc, example.py Part B]: - Apparent output: 40,000 V × 0.150 A = 6000 W, ten times the input. The 40 kV is not measured; it is the voltage to strike a 4-inch spark (about 10,000 V per inch, PDF p.217). - To deliver no more than 600 W at 150 mA, the mean voltage across the running flame must be at most 600 / 0.150 = 4000 V, one-tenth of the striking voltage (less still if the meter reads high or the coil loses power). - His own arc analogy (100 V sustains what 10,000 V strikes) implies a collapse by a factor of up to 100. A tenfold collapse is all that is needed.

This is the right physics. Once an arc or spark is conducting, the gap is filled with ionised gas and the voltage across it falls to a small fraction of the voltage that was needed to break the air down [memory, high]. So a spark-gap reading measures the striking voltage, which the circuit does not hold while current flows. Applied to the x-ray coil of p.741: if the voltage during conduction falls to 1/10-1/100 of the 100 kV spark reading (10 kV to 1 kV), then even with an averaging meter reading 15 mA the power is 150 W to 15 W [calc, example.py Part B], inside the 500 W input. The voltage factor alone can close the ledger. (An x-ray tube is not an open arc, so how far the tube voltage falls is not measured in the book; the p.740 measurement on the Cabot converter gave "less than one-half". The p.156 factor shows how large the collapse can be in a running gas discharge; it is not a measurement on the tube.)

Step 4. Correct the current. Model the secondary current as half-sine pulses occupying a fraction D of the time. Take D = 0.05 as an illustrative value (the book gives no pulse length; this is an assumption). From Section 3.1: - I_avg = (2/π)(0.05) I_p = 0.0318 I_p - I_rms = I_p √(0.05/2) = 0.158 I_p

This step is the hot-wire case: it asks how much the current reading can overstate the average, keeping the voltage at the Step 3 value. Now suppose the meter reads 15 mA, and work out the true average current for each kind of meter:

[calc, example.py Part B]

Step 5. Use conservation of energy as a measuring instrument. If the meter really was an averaging galvanometer, energy conservation tells us something the book does not: with 15 mA average and at most 500 W in, the mean tube voltage during conduction must be at most

V_mean ≤ η × 500 W / 0.015 A = 33 kV × η,

where η (0 < η ≤ 1) is the coil's efficiency. At η = 0.7 that is 23 kV, a quarter of the 100 kV spark reading [calc]. And if, as Tousey says, nearly half of the secondary power goes into "inverse discharge" (current in the wrong direction through the tube), the useful power is at most about 250 W × η. So with an averaging meter, the "less than one-half" voltage drop measured on the Cabot converter (p.740) is not enough on its own; the voltage must fall to a third or less, or less than 15 mA was flowing. The arbiter reached the same bound (arbiter D2). But the book does supply a larger factor: Tousey's own p.156 experiment and arc analogy (Step 3b) describe a collapse of tenfold to a hundredfold, which meets the bound with room to spare. The arbiter's QE revision accepts this: "with a voltage collapse of tenfold to a hundredfold, as in p.156, the ledger closes for any meter" (arbiter R5). What the book does not give is a measured tube voltage for the 100 kV, 15 mA case.

Step 6. The Oudin resonator. Tousey's second case: "10,000 volts × 0.250 amperes = 2500 watts; while the power of the primary current ... may be only 100 volts × 5 amperes = 500 watts" [read: PDF p.744]. The 10,000 V comes from a one-inch spark (a peak); the 0.25 A from a hot-wire meter (an rms value).

• Apparent power, done properly: V_rms = 10,000/√2 = 7071 V; S = 7071 × 0.25 = 1768 VA, not watts [calc, Part C].
• Why it is not real power: the resonator drives current through a capacitance (electrode, glass wall, patient). For a sine-wave voltage v = V_p cos ωt and current i = I_p cos(ωt + φ), the average power over one cycle is (V_p I_p / 2) cos φ. For a pure capacitor φ = 90°, so the average power is zero [calc, SymPy in Part C].
• How much capacitance is that? At 500 kHz (the book quotes frequencies of 300,000 to 600,000 per second), I = ωCV gives C = 0.25 / (2π × 500,000 × 7071) = 11 pF [calc]. That is about the capacitance of a hand-sized electrode near a body, so the reading makes physical sense as a mostly capacitive current.
• The real power is only what the resistance in the path absorbs, P = I_rms² R. With R = 200 to 1000 Ω (tissue, skin and spark; assumed): 12 to 62 W, a power factor of 0.007 to 0.035 [calc]. Add the resonator coil's own losses; the arbiter's assumed power factor of 0.1 [memory, medium] gives 177 W. Every case fits inside 500 W.

So here Tousey's "maximum of each successive discharge" is the smaller part of the story. The larger part is that volts times amps is not watts when current and voltage are out of step.

What the example shows. 1. Tousey reasoned correctly from conservation of energy: an apparent excess is a measurement error, not new energy. That is the right instinct, and it is the instinct that many later "over-unity" claims lack. 2. He gives two explanations. The p.741 one ("meters read the maximum") works for a hot-wire or peak-reading meter on short pulses, not for an averaging galvanometer, and his own p.155 says the galvanometer reads the average. The p.156 one (the voltage collapses once the arc runs, as 100 V sustains an arc that 10,000 V strikes) works whatever the meter, because it corrects the voltage, not the current. The lesson: a spark gap measures the voltage to start a discharge, not the voltage while it runs. 3. For the resonator, the missing idea is the power factor. A modern reader should never multiply a peak voltage by an rms current and call it watts. 4. Energy conservation is a tool: when the ledger refuses to balance, it tells you what the instrument must really have been measuring.

5b. A second, shorter check: the dose of the "Author's dose"

[calc, reproducing arbiter D1; example.py Part D] - Charge through the tube: 3 mA × 600 s = 1800 mAs. - Distance factor from 1 m to 10 in (0.254 m): (1/0.254)² = 15.5. - Free-in-air dose: (0.023 to 0.030 mGy/mAs) × 1800 × 15.5 = 0.64 to 0.84 Gy. - Backscatter at the skin, ×1.3 [memory, medium]: 0.83 to 1.09 Gy. - A coil-driven gas tube runs at an effective voltage below its peak. If 55 kV instead of 70 kV, output (∝ V²) falls by (55/70)² = 0.62: 0.51 to 0.67 Gy. - The arbiter's range, allowing a factor of 2-3 for gas-tube drift: about 0.3 to 1.1 Gy per session, central value near 0.7 Gy. The modern threshold for transient erythema is about 2 Gy, i.e. 2 to 4 times higher than the central range.

The book's own conversions. The first version of this primer (and of the arbiter) said the book holds no period calibration of the "(5) H" label. That was wrong: the book converts the same settings into erythema doses in at least eight places (arbiter R1, from the deep read of PDF 749-773 and 1130-1180). The author's dose is 6-inch spark (about 70 kV), 3 mA, 10 in, either 5 min unfiltered (900 mA·s) or 10 min through 3 mm aluminium (1800 mA·s) [read: PDF p.1184 note, p.1181 table]. The erythema dose E is taken as 5 H where the book gives H. [read: pages as listed; calc: example.py Part E]

(The QE's table placed Cabot and Dodd's figure "at about 90 kV" from p.759; the arbiter's deep read found that p.759 uses 90 kV only as an example of the rule "10,000 volts = 1 Benoist", and only p.763 names the experimenters and the measured 4 Benoist. Hence the scaled range 0.16-0.81 for row 7.)

How to read the table: - For one set of settings, the book's own scales give anything from about 0.16 to 1.5 erythema doses: they disagree with each other by a factor of about 6 (unscaled, 0.26 to 1.5) to 9 (scaled). This is why the arbiter now calls the "(5) H" label "Internally inconsistent" in the book's own units, no longer "minor". - The two period calibrations from other workers' measured settings (rows 7 and 8: Cabot and Dodd; the 57 kV figure) put the unfiltered dose at roughly a quarter to a half of an erythema dose. They agree with the modern estimate above (0.3-1.1 Gy against a 2 Gy threshold, i.e. 0.15-0.55). - The conversions that put the dose at or above an erythema (rows 1 and 4) are Tousey's own table and directions. Row 4 contradicts "never causes dermatitis" in his own terms more strongly than the "(5) H" label does. - The book warns against all of these measures itself: "milliampere seconds do not form a reliable measurement of x-ray dosage" (p.763); the Tousey unit "is no longer required" (p.1173); on a gas tube "If there is inverse discharge the milliamperes are dangerously misleading" (p.1181) [read].

So the physical dose was probably below the reddening threshold, which supports "usually no visible reaction after one sitting", and now with two period calibrations behind it as well as the modern estimate. It cannot support "never": a gas tube's vacuum drifts, there is no case series, and two of the book's own conversions place the same settings at or above an erythema dose. Which of the book's figures, if any, was measured on his own apparatus is unknown.

---

6. Evidence table

Verdicts are the arbiter's (arbiter.md, as revised in its "QE revision" section R1-R9), on the Step 2 scale. Rows marked with a prime (2a', 2b', 3c') were added in the revision. "Tested?" means a neutral or controlled test on record, not Tousey's own cases.

Material that stays in the record whatever the verdict (from the Investigator and the reviews): - Tousey's seven US patents, including US1100037 (x-ray tube with a lead-glass wall and a window, filed 1911-02-23, granted 1914), US1372110 (cellular diaphragm) and US1535359 (compensating filter); five probable Canadian and French family members; firearm-aiming and fire-control patents filed 1917 [web: research/2026-10-08-google-patents.md]. The lead-glass idea (shield everything but the useful beam) is how modern tube housings work [memory, high]. - His own case report of x-ray keratoses on his hands, cured by radium, Archives of Radiology and Electrotherapy 20(10):344 (1916) [web: https://academic.oup.com/bjrbir/article-pdf/20/10/344/57486712/are.1916.0084.pdf; snippet only, not opened], and his 1922 statement "3600 units of soft x-rays will produce severe dermatitis" [web: Google Books snippet, research/2026-10-08-google-books.md]. These concern operator exposure and a different dose; they show he knew the hazard was dose. - The 1913 generator article, NY Medical Journal 97:648-649: Cabot and Shaw, sent to x-rays by "Doctor Williams, of Boston", found "three or four pages which suggested the present discovery" in his book [web: https://archive.org/details/newyorkmedicaljo9719unse]. Documented in one published source by the claimant, who has a stake in the credit; Cabot's 1912 converter patents (US1112436, US1191759) do not mention x-rays or Tousey, which is silence, not contradiction (patents rarely credit consultants). - Geyser's side of the contact-tube dispute (about 5000 contact applications without a burn) is known only as Tousey reports it (PDF p.1230). Geyser is also an interested party; his later Tricho hair-removal system was criticised by the American Medical Association in 1925 [web: https://en.wikipedia.org/wiki/Albert_C._Geyser]. That bears on how much weight his 1908 record deserves; it does not refute it. - A contemporary review of the 1910 edition (American Journal of the Medical Sciences, July 1911) found that "many statements are made of cures ... which are not borne out by facts, but which the author quotes in good faith" [web: https://archive.org/details/paper-doi-10_1097_00000441-191107000-00023]. A 1916 orthodontics review called his focal-infection reports "more convincing" [web: IA sim_american-journal-of-orthodontics-and-dentofacial_1916-10_2_10]. The second shows the medical climate of belief, not a test. - Hartzell, Henrici and Leonard's counts (endocarditis in 23 of 31 cardiac cases), quoted in his 1916 dental monograph [web: https://archive.org/details/roentgenographic00tousiala]; not read by the reviewers. - The Tousey unit was published (Archives of the Roentgen Ray 17:427, 1913), abstracted in JAMA 60:1664 (1913), used in 1916, and noted by the historian R. F. Mould [web: research/2026-10-08-google-books.md]. His radium assay through the sealed tube (NYT 7 Jan 1914) uses the standard gamma-ray method [web: https://www.nytimes.com/1914/01/07/archives/suggests-testing-tube-of-radium-strength-can-be-determined-without.html]. - Where the record goes quiet (arbiter, "Absence-of-evidence analysis"): no reply by Cabot, Shaw or Williams on the generator priority; the Tousey unit after 1922; Geyser's own account; the commercial fate of the tube patents; the 1917 war patents. Competing explanations for each (ordinary neglect, the unit replaced by the roentgen, lost company papers, an overstated role, a patent that went nowhere) are set out there with the records that would decide. No sign of suppression was found in any of them; these are leads, not conclusions. - Tousey's pregnant-cat experiment, 1904 [read: PDF p.1142-1143, with the settings table; arbiter R1]. He radiographed a cat that had borne several healthy litters, seven exposures from 6 Oct to 17 Dec 1904 (3-8 in spark, 15-19 in to the plate, 10-65 s each, 276 s in all; no tube current given), holding her with his own hands in the beam ("this was a dangerous experiment"). Three "mature and perfectly formed kittens" were born dead on 30 Dec; with exposures stopped, four healthy kittens followed on 22 Mar 1905. One animal, one litter, no control, dose not recoverable; cat gestation of about 63-65 days [memory, high] means the October exposures probably preceded conception. It is a primary observation by the inventor, in line with the greater radiosensitivity of embryonic tissue (Bergonié and Tribondeau's rat-testis work, cited by the book p.1145; Established). As evidence on its own: Never properly tested. - Other experiments of his own in the book (arbiter R8): his radium forearm self-test (20 mg applicator, redness and desquamation after 2-5 min, about two weeks' incubation; PDF p.1270, Plate 15); the 1904 radiated-saline experiment (radium sealed in glass in saline for fourteen days, no radioactivity found, no effect in animals; PDF p.1282-1283; a null result the sealed tube would give anyway, a reader's inference); a shoe-conductivity measurement of 12 May 1919 (PDF p.360); oyster-pearl radiography, 121 oysters per plate, a method used in Ceylon (PDF p.1141). Also reported from others: Millener's 22 power-house workers with health effects attributed to stray fields (PDF p.302-303). - Operator injuries recorded in the book (arbiter R1): chronic dermatitis "produced by frequently repeated exposures which individually would produce no effect" (pp.1154-1155); his own hand keratoses, which he blames on holding the film in the patient's mouth for dental radiographs (p.1159); deaths of named x-ray workers from x-ray cancer (p.1155). "Never causes dermatitis" refers to one sitting on a patient; it does not cover cumulative exposure. - Thiosinamin. Tousey introduced this drug in 1894 for scars and keloids, and the book claims results with it (PDF p.634, 1189-1190); the Investigator found his 1896 monograph on it [web: Internet Archive, sites.md]. It is his own pharmaceutical contribution and has not been assessed by the reviewers. - Tousey's papers in archives [web, via the Investigator, sites.md]: the New-York Historical Society holds a small collection, including a 1911 typescript, "Protection of Patient and Operator in Radiography and Radiotherapy" (https://www.nyhistory.org/blogs/the-x-rays-of-melville-e-stone-jr); the University of Pennsylvania holds "Tousey, Sinclair, 1906-1920" in Milo Hellman's reprints collection, box 29, folder 40 (https://archives.upenn.edu/collections/finding-aid/upc4_11/). Neither has been examined. They are the nearest records that could decide the contact-tube and dental-arch priority questions. - Name confusion: the publisher Sinclair Tousey (1815/1818-1887), probably his grandfather, and a Sinclair Tousey found dead in Yonkers in 1915 are different men; The Scar Book conflates the publisher with the physician [web: via the Investigator].

---

7. What a learner should take away

What is solid. - The energy-conservation reasoning in the two "output greater than input" passages. Tousey refused to believe his meters over the law, and he was right; and on p.156 he gave the correct physical reason (a running arc needs only a small fraction of its striking voltage). - Several correct positions that a quick reading pins the other way: he cites the experiments that rule out the condenser theory as the cause of burns (p.1158); he says ultraviolet does not pass mica or glass and comes from sparks outside the electrode (p.557, 678); he states that the slow drift of the low-potential wave is not the speed of the current (p.289). - Output ∝ I V², the case for a constant-potential x-ray supply, and the warning that a milliammeter without a fixed voltage is "dangerous guesswork" [read: PDF p.741]. - Scatter is a large part of what reaches the film; shielding and grids help. His patented lead-glass tube applies the same idea at the source. - Tissue conducts by ions; ionic medication acts near the surface; the resting voltage comes from a concentration difference across a selective membrane (the passage he quotes, PDF p.287 of the second scan). - Dental radiography of hidden abscesses and unerupted teeth, and the inverse-square reasoning about contact treatment (Tint B in contact = four times the usual dose).

What is wrong, and why the mistake is instructive. - 39 m/s and "almost all effects are electrolysis". The premise that the body "conducts only by electrolysis" is one Tousey reports from "many observers" (p.296); his own p.289 gives the correct distinction (current at "possibly 180,000 miles a second"). But seven pages later he draws the observers' conclusion himself ("This would mean ... 39 meters a second") without reconciling it, and his p.297 sentence on electrolysis is contradicted by his own pages on condenser discharges and the d'Arsonval method (pp.306, 340). The lesson is general: the speed of the carriers is not the speed of the signal. Electrons in a copper wire also drift at fractions of a millimetre per second, yet a lamp lights at once. A second lesson: a long compilation can state the right answer and the wrong one a few pages apart, so quote the page, not the book. - The condenser theory of burns. This was Geyser's theory, not Tousey's. The book reports the experiments against it (Strater, Kienböck, Oudin, Scholtz, p.1158) and rejects it as the cause (p.1162), while elsewhere allowing it "a certain element of truth" (p.1231). Tousey's own inverse-square remark about contact dose was the simpler explanation all along. The instructive point is how an interested party's theory (Geyser's contact tube competed with Tousey's) can travel under another author's name when a summary drops who held it. - Ultraviolet from the vacuum electrode. The first versions of this review quoted the book as crediting ultraviolet "absorbed by the body"; the full sentence says it is "hardly to be supposed" that it plays a part (p.578). A quote cut from its sentence can invert its meaning. What remains untested is his claim, in other chapters, of a germicidal effect of ultraviolet and ozone from outside sparks in acne and skin disease (p.631, 635). - Dose units that disagree. The book's own conversions put one set of settings anywhere from about 0.16 to 1.5 erythema doses (Section 5b). Period dosimetry had no agreed physical unit, and the book itself warns that milliampere-seconds "do not form a reliable measurement". - "Never" and "always". "Never causes dermatitis" and "has always found dental infection" in cancer are strong universal statements with no case counts. When the condition is common (dental infection in adults), "always" proves nothing; when the outcome is rare (a burn at sub-erythema dose), "never" needs many patients to show. - Focal infection. This is the instructive case of the medical establishment being wrong first: it was the consensus of leading physicians, praised in journals, and then dropped when controlled follow-up arrived. Consensus is not proof, in either direction; controlled comparison is. - Multiplying peak volts by meter amps. The worked example shows how a 1500 W "output" turns into 15-150 W of real power once the striking voltage and the meters are understood.

What is genuinely open. - Whether Cabot and Shaw's constant-potential generator came from Tousey's pages: the deciding records are Cabot-Shaw company papers, the correspondence of "Doctor Williams of Boston", and the 1910 edition's relevant pages (on the Internet Archive, not yet read). - Contact-tube priority: Dental Cosmos 1906, the New York Odontological Society minutes of 17 Oct 1905, and Geyser's JAMA 1908 paper. - The actual dose of his standard treatment: a measured spectrum and output of a period coil-and-gas-tube set at 10 in would settle it. - Whether his arch formula predicts adult arches: needs a follow-up series. - Whether high-frequency treatment has any effect beyond heating in a defined condition: needs a sham-controlled trial, and the original reports of Apostoli (500 cases), Moutier, Gidon, Guilleminot, Gay, and Delherm and Laquerrière (how many patients) to see what was measured. - Which meter gave his "15 mA" on p.741, and how far the tube voltage fell during conduction (Section 5): not stated in the book. The p.156 arc experiment shows a collapse large enough to close the ledger for any meter, but it was measured on an open flame, not on a tube. - Which of the book's eight dose conversions, if any, was measured on his own apparatus, and at what voltage Cabot and Dodd's 3400 mA·s was taken (Section 5b): the Cabot and Dodd paper (Massachusetts General Hospital) and Tousey's own papers would decide. - Whether early burns come from tube ultraviolet (p.681): his cited source (Bierry, Henri and Ranc, 1911) and a measurement of the UV transmission of a period tube envelope. - Tousey's archival papers (New-York Historical Society; University of Pennsylvania, Hellman reprints, box 29, folder 40) have not been examined; they may bear on the contact-tube and dental-arch priority questions.

---

8. Check yourself

1. A hot-wire meter in series with an x-ray tube reads 10 mA. The current flows in half-sine pulses for 2 percent of the time. What is the average current, and what power does the tube take if its voltage during conduction is 40 kV?
2. The Oudin resonator: why is "10,000 volts × 0.25 amperes" not 2500 watts? Give two separate reasons.
3. Tousey's contact tube puts the skin 2.5 in from the target instead of 10 in. By what factor does the dose rate at the skin rise, and what does this imply for comparing Geyser's "5000 applications without a burn" with treatments at a distance?
4. A sodium ion drifts in a field of 1 V/cm. How far does it move in one minute? Why does this not mean that an electric current takes hours to cross the body?
5. Tousey writes that he "has always found dental infection" in every cancer patient he radiographed. Suppose 80 percent of adults of that age have some dental infection on radiographs. What would you need to collect to test his claim?

---

9. Further reading

Mainstream (textbooks; chapter numbers given only where certain) - Bushberg, Seibert, Leidholdt and Boone, The Essential Physics of Medical Imaging (x-ray production, tubes and generators, scatter and grids, dosimetry). - Johns and Cunningham, The Physics of Radiology (x-ray spectra, output, dose). - Hall and Giaccia, Radiobiology for the Radiologist (skin reactions and dose). - Griffiths, Introduction to Electrodynamics, ch. 7 (electromotive force, Faraday's law, inductance); any AC circuits text for rms, reactive power and power factor. - Horowitz and Hill, The Art of Electronics, ch. 1 (signals, rms, capacitors in AC circuits). - Hille, Ion Channels of Excitable Membranes, and Nelson, Biological Physics (Nernst potential, membranes, ionic conduction). - Widder 2014 on Kienböck and the dose theory of x-ray skin effects [web: https://pubmed.ncbi.nlm.nih.gov/25012643/]. - R. F. Mould, A Century of X-Rays and Radioactivity in Medicine (history of early dosimetry, mentions the Tousey unit [web: via the Investigator]).

Tousey's own key works - Medical Electricity, Rontgen Rays and Radium, 3rd ed. 1921 (this project); 1st ed. 1910 and 2nd ed. 1915 on the Internet Archive; second 1921 scan: https://archive.org/details/medicalelectrici00tousiala. - Roentgenographic Diagnosis of Dental Infection in Systemic Diseases (Hoeber, 1916): https://archive.org/details/roentgenographic00tousiala. - "An Improved X Ray Generator", New York Medical Journal 97:648-649 (29 Mar 1913): https://archive.org/details/newyorkmedicaljo9719unse. - "A New Unit of X-ray Power", Archives of the Roentgen Ray 17:427 (1913). - "X-Ray Keratoses on Hand of Radiologist. Cured by Radium", Archives of Radiology and Electrotherapy 20(10):344 (1916). - US patent 1,100,037 (x-ray tube, filed 1911): https://patents.google.com/patent/US1100037A/en. - His 1896 monograph on thiosinamin (Internet Archive, found by the Investigator; see sites.md). - Unexamined archival holdings: New-York Historical Society, Tousey papers (1911 typescript "Protection of Patient and Operator in Radiography and Radiotherapy"); University of Pennsylvania, Milo Hellman reprints collection, box 29, folder 40, "Tousey, Sinclair, 1906-1920": https://archives.upenn.edu/collections/finding-aid/upc4_11/. - In the book itself, the passages this primer leans on most: PDF p.155-156 (meters and the 6000 W experiment), p.289 and 296 (signal speed), p.557-558, 578, 631, 678 (ultraviolet), p.603-604 (diabetes, Apostoli, his caveat), p.740-744 (coil and Oudin ledgers), p.755-764 (Cabot and Shaw generator, GE test), p.1158 and 1162 (causes of dermatitis), p.1167-1184 (dosage).

Code: example.py in this folder reproduces Sections 3.1, 5 (including Step 3b) and 5b (Part D, the modern estimate; Part E, the book's own conversions). Run: .venv/bin/python review/Tousey-Sinclair/example.py

---

Corrections to earlier documents

1. arbiter.md, D2, point (i), and mathematician.md, Corrections to Step 1, item 1 ("A moving-coil milliammeter reads the average current ... so the current factor is not inflated"). This holds only if the meter was a moving-coil galvanometer. The book names hot-wire meters as the other type in use and says they "indicate the current strength regardless of its alternating polarity" [read: PDF p.741]; Tousey says the hot-wire meter gave the Oudin reading [read: PDF p.744]. A hot-wire meter reads rms current, which for half-sine pulses exceeds the average by π/(2√(2D)), i.e. 5 times at a duty cycle of 5 percent [calc, example.py Part A]. So Tousey's "the meters indicate the maximum ... rather than their average strength" is roughly right for a hot-wire meter on short pulses, and the coil ledger then closes (Section 5, Step 4). Which meter gave his 15 mA figure is not stated. The arbiter's conclusion (no over-unity, quantitative reconciliation not in the book) stands; the statement that the current factor is not inflated should read "not inflated if the meter was a moving-coil galvanometer".
2. arbiter.md, D2, point (ii), cites the "less than one-half" voltage drop of PDF p.740 as evidence about the x-ray coil. The page says those observations were made "with the Cabot direct current, high potential converter" [read: PDF p.740]. Applying it to the induction coil is an analogy, not a measurement on the coil. This does not change any verdict.
3. Page citations for the 39 m/s passage. (Superseded in part by item 7 below.) The reviewers cite "p.296" (PDF) and the arbiter cites "p.281" for 180,000 mi/s. The 39 m/s sentence is on PDF p.296, which carries the running head of printed p.300 [read]; and it is framed as the consequence of a view held "according to many observers", not stated by Tousey as his own measurement. The verdict (Internally inconsistent; Refuted by calculation) is unchanged, but later documents should attribute the premise to the observers he reports.

Added in the QE revision (2026-10-09), applying qe.md and arbiter.md R1-R12. Each item corrects this primer's own first version.

4. Section 1 and section 4 (claim 2), and section 7, "The condenser theory of burns. A plausible mechanism was held after a 1900 experiment had already set it aside." Wrong attribution. The theory is Geyser's. The book cites "Strater, Kienböck, Oudin, and Scholtz" against it [read: PDF p.1158] and says "The author does not regard this as probably correct" [read: PDF p.1162]; only p.1231 allows "a certain element of truth". Verdict split accordingly (table rows 2a, 2a').
5. Section 6, row 3c, "Germicidal UV from a soda-lime glass vacuum electrode: Refuted by calculation". This targeted a claim the book disowns. The book says ultraviolet "will not penetrate the thinnest piece of mica" and comes from sparks outside the electrode (PDF p.557-558) and that its direct part is "hardly to be supposed" (PDF p.578). The arbiter's first-pass quote ("ultraviolet rays absorbed by the body" matter, cited to "PDF p.585") was from p.578 and inverted the sentence (qe.md section 4, row 3; arbiter R3). Row replaced by 3c (Established, as the book states it) and 3c' (benefits of outside-spark UV and ozone: Never properly tested).
6. Section 5, Step 5, and "What the example shows", item 2 ("does not by itself work for an averaging galvanometer"). Incomplete: it omitted Tousey's own p.156 experiment (apparent 6000 W from about 600 W) and his arc explanation, which implies a voltage collapse of tenfold or more and closes the ledger for any meter [read: PDF p.156; arbiter R5]. Step 3b added; the duty-cycle table is kept as the hot-wire case. Also, p.155 shows Tousey knew the D'Arsonval meter reads the average.
7. Section 4 (claim 7) and section 7, "Both take a true premise ... one step too far", presented as Tousey's error. The premise "conducts only by electrolysis" is reported from "many observers" (PDF p.296); Tousey's own p.289 gives the correct distinction. The conditional "this would mean 39 m/s" is still his, drawn without reconciling it with p.289, so the verdict stays Internally inconsistent; Refuted by calculation as a signal speed (arbiter R7). Body text changed to match.
8. Section 5b, "his own table labels the same dose '(5) H' ... The two disagree" and the implied absence of a period calibration. The book converts the author's settings into erythema doses in at least eight places, giving about 0.16-1.5 erythema doses (Section 5b table; qe.md 3.1; arbiter R1). The "(5) H" inconsistency is therefore not "minor".
9. Section 6, row 3b, and section 4 (claim 3). One-sided: the contrary authors were listed but Apostoli's "success in over 500 cases" (PDF p.603), Gay's 12 measured cases (PDF p.595), the other small positive series, and Tousey's own caveat that his diabetic patients kept their medicines (PDF p.604) were left out. Added (qe.md section 5; arbiter R4).
10. Section 6, rows 1d and 4d. Scale lapses: the TB line scored x-ray as a TB treatment rather than Tousey's statement, and "Geometry consistent" was not a scale verdict. Restated (arbiter R9).
11. Material that stays in the record. The pregnant-cat experiment, his other experiments, the operator injuries, thiosinamin and the NYHS and University of Pennsylvania archival holdings were missing. Added.

---

Answers

1. For half-sine pulses, I_rms = I_p √(D/2) and I_avg = (2/π) D I_p. With D = 0.02: I_p = 10 mA / √0.01 = 100 mA; I_avg = (2/π)(0.02)(100 mA) = 1.27 mA. Power = 40 kV × 1.27 mA = 51 W (not 40 kV × 10 mA = 400 W). The hot-wire reading overstates the average current by a factor of 7.9.
2. (i) The 10,000 V is a peak value from a spark length; the rms voltage is 7071 V, so even the apparent power is 1768 VA. (ii) The current is mostly capacitive (through about 11 pF at 500 kHz), 90 degrees out of step with the voltage, so the average power ⟨v i⟩ is close to zero; real power is only I_rms² R in the resistance of the path, tens of watts.
3. (10/2.5)² = 16 times the dose rate at the skin (6.25 times if the book's other figure, "10 cm" or about 4 in, is right). A contact treatment and a distance treatment can only be compared at equal measured skin dose. Geyser's record says nothing about an electrical effect unless the doses were matched, and Tousey himself notes no comparative experiments with equal doses were made (PDF p.1230).
4. v = μE = 5.2 × 10⁻⁸ m²/(V·s) × 100 V/m = 5.2 × 10⁻⁶ m/s; in 60 s that is 0.31 mm. The current does not wait for one ion to cross the body: the electric field is set up throughout the conductor in about ε/σ ≈ 10⁻⁹ s, and every ion everywhere starts to drift at once, so current flows at once along the whole path.
5. Dental radiographs of a group of cancer patients and of a matched control group without cancer (same age, sex, social class), read blind by someone who does not know which group each film belongs to, with the infection rate in each group counted. If 80 percent of controls also show infection, finding it in every cancer patient tells you little; what matters is whether the rate (or severity, which Tousey also claimed) differs between the groups by more than chance.