zotero-db/storage/H6D6V6CF/.zotero-ft-cache

This is a reproduction of a library book that was digitized by Google as part of an ongoing effort to preserve the information in books and make it universally accessible.
https://books.google.com

muNORTWEHSTE

UNIVE

1851
NORTHWESTERN UNIVERSITY LIBRARY

UNID
1851
NORTHWESTERN UNIVERSITY
LIBRARY

-

James PrescowJoule
Engraved by
London Purushesh Macmilia

‫دننک یم دورو‬

THE
SCIENTIFIC PAPERS
OF
JAMES PRESCOTT JOULE,
D.O.L. (Oxon .), LL.D. ( DUBL. ET Edin .), F.R.S., Hon. F.R.S.E.,
PREB. 80C . LIT. PHIL . NAXC., P.C.S., DOC. NAT . PHIL. LUGD. BAT., 8OC . PHIL . CANTAB ., BOC . PBIL . GLASC ., INST . MACH . ET NAUP. SCOT. ET BOC. ANTIQ . PERTA ,
80C. HONOR. INST. FR . (ACAD. SCI.),
CORRESP. SOC . REG . DAN. HAFN . TAURIN . BOLON ., SOC . PHIL . NAT. BASIL ., BOC . PHYS . TR , PAR . ET HAL .,
ET ACAD , AYER , SCI. ET ARTIB . ADBOC . HONOR.
PUBLISHED BY
THE PHYSICAL SOCIETY OF LONDON .
LONDON : TAYLOR AND FRANCIS , RED LION COURT, FLEET STREET.
1884 .

Yeh ... opet ,
500.17
546 vid
lory
Maid \?. c.5 .
ALERE FLAMMAM .
PRINTED BY TAYLOR AND FRANCIS ,
RED LION COURT, FLEET STREET.
2834 ?
271 1877

ADVERTISEMENT.
In issuing this volume, the Council of the Physical Society of London desire to put on record their cordial appreciation of the kindness with which Dr. Joule not only agreed to their request to allow them to publish a collected edition of his Scientific Papers, but undertook personally the labour of getting together and editing the collection .
They desire also to thank the Council of the Philosophical Society of Manchester and the Editors of the Philosophical Magazine' for the use of woodcuts, and Messrs. Macmillan and Co. for allowing Jeens's excellent portrait of Dr. Joule, originally published with ' Nature,' to appear as a frontispiece to this volume.
A second volume, containing papers published by Dr. Joule in conjunction with other men of science, is in progress, and will be published as soon as practicable.
FEBRUARY 1884 .

530.4
J86
vil

idea

PREFACE.
This book appears in consequence of the flattering proposal of the Physical Society of London to collect and reprint the papers on scientific subjects which have appeared in my own name, and those under my own in association with the Rev. Dr. Scoresby, Sir Lyon Playfair, and Sir Wm. Thomson . In this, the first volume, I have endeavoured to fulfil the former part of this design .
The more important papers are literal transcripts from the original; but in some of the earlier a few alterations have been made in the phraseology, not with the intention of altering the meaning, but of making it more clear.
I feel that many imperfections will be found in the com pilation, and even that some of the papers may be thought superfluous. However, I have yielded to the wish that the whole should be printed in chronological order as nearly as possible.
J. P. JOULE .
12 Wardle Road, Sale, Cheshire,
November 1883.

TABLE OF CONTENTS.
DESCRIPTION OF AN ELECTRO -MAGNETIC ENGINE, 1-3.
Construction of the magnets, 1. - Arrangement of them in circles,
2. - Description of machine, 3.
DESCRIPTION OF AN ELECTRO -MAGNETIC ENGINE, WITH EXPERIMENTS;
4-6 .
Value of wire-magnets, 4. - Construction of engine, 5. - Its per
formance, 6 .
ON THE USE OF ELECTRO -MAGNETS MADE OF IRON WIRE FOR THE ELECTRO -MAGNETIC ENGINE, 6-10 .
Comparison of solid magnet with one of square iron wire, 6. Comparison of magnets of round bar iron with hollow cylindrical magnet, 7. - Characteristics of magnets for lifting and for attracting from a distance, 8.—Advantages of hollow and solid magnets, 8. Comparison of square wire magnets with square solid magnets, 9.
INVRSTIGATIONS IN MAGNETISM AND ELECTRO -MAGNETISM , 10–14 . Description of galvanometer, 10. - Experiment with solid and wire
magnets, 11. – First Table of results, 12. - Second Table of results, ib. -Law of attractive force of two electro-magnets, 13. - Apparent failure of law in cases of lifting, ib . - Substitution of magnetism for steam , 14 .
INVESTIGATIONS IN MAGNETISM AND ELECTRO -MAGNETISM, 16–16 .
Experiments to test the law enunciated on p. 13.

viji

TABLE OF CONTENTS .

DESCRIPTION OF AN ELECTRO -MAGNETIC ENGINE, 16–18. Description of engine, 16. - Of the magnets employed, 18.

ON ELECTRO -MAGNETIC FORCES, 19-26.
Ratio of variation of electro -magnetic attraction , 19. - Law of
electro-magnetic attraction, 21. - Experiments with electro -magnetic
engine 21-25 . – Law of electrical resistance, 25 .

ON ELECTRO -MAGNETIC FORCES, 27-39.
Lifting power of magnets, 27. — Units of measurement, 28. Galvanometer used, 29.- Description of electro -magnets used in the experiments, 30. — Tables of electric force and lifting powers of magnets, 32–34 . – Tables of lifting power per degree of electric force, 36. — Tables of lifting power after current is wholly or partially detached , 37 .

Note on Voltaic Batteries.
Description of a battery of improved construction, 38–39 .
ON ELECTRO -MAGNETIC FORCES, 40-42. Lifting power of magnets, 40. — Construction of new magnet, 1.
Refitting of large magnets previously used, 41. -Confirmation of formulafor lifting power, ib .

DESCRIPTION OF A NEW ELECTRO -MAGNET, 42-46.
So constructed as to attain a large sectional area , 42. - Description
of the magnet, 43–44 . - Its actualpower, 45–46 .

ON A NEW CLASS OF MAGNETIC FORCES, 46–53.
Electricity available for mechanical purposes, 46. - Laws of action of electro -magnetic engines, 47. – Comparison of electro -magnetic with steam -engines, 48. - Electro-magnetic coil increases the length of a bar of iron, 48. - Experiments to ascertain the laws of such increase, 49. - Application to the movement of machinery, 50.
Modification of Ampère's hypothesis, 51. – Of the theory of Æpinus,
62. - Explanation of the influence of heat on the magnetic power of iron, 63.

ON VOLTAIC APPARATUS, 53-59.
Destructive effects of local action, 63. - Means of remedy, 64. Method of furnishing the copper with a stable surface, 55.- Con

TABLE OF CONTENTS .

ix

struction of battery with improved plates, 55.I—ntensities of dif ferent batteries, 57. — Experiment with revolving copper disk, i.
Experiments with nitric acid in batteries, 58. — Value of dia
phragms, ib.

ON THE PRODUCTION OF HEAT BY VOLTAIC ELECTRICITY, 59-60 .
Experiment to show different degrees of facility with which metals are heated by electricity, 69. - Law of heat-production by elec tricity, ib.

ON THE HEAT EVOLVED BY METALLIC CONDUCTORS OF ELECTRICITY, AND IN THE CELLS OF A BATTERY DURING ELECTROLYSIS, 60–81.
CA. I. Heat evolved by Metallic Conductors.-- Ratio in which metals are heated by electric current, 60. - Construction of galvanometer used in the experiments, 61.- Degree of current electricity defined, 17. - Experiment to ascertain heating power of a metallic wire, 62. Experiments on heat evolved by conducting wires, 63. — Law of ovo lution of heat by voltaic current, 64. - Experiment to prove rate of increase as square of current, ib. — Applications of the law to mea surement of electric current, 65. — To the comparison of frictional and voltaic electricities, ib.
CA. II. Heat evolved during Electrolysis . - Correction for differences of temperature, 66. — Correction for specific heat, ib . — Determination of standard of resistance, 66-67 . - Necessity for eliminating all action not electrolytic, 68. — Experiments to show agreement be tween theoretical and actual quantity of voltaic heat, 68–72. Formula for voltaic heat, 72. - Experiments on heat evolved by passage of voltaic electricity through electrolytes, 72–76 . — Tabulated results of experiments, 76. – Further experiments, 16. - Law of evolution of voltaic heat, 77.- Light and heat of combustion pro duced by electricity , 78.
Note on Voltaic Batteries.
Table of intensities of batteries in general use, 79.

ON THE ELECTRIC ORIGIN OF THE HEAT OF COMBUSTION, 81-102.
Reasons for undertaking the experiments, 81. - Abnormal deviation of the needle of the galvanometer, 82.— Traceable to the influence of atmospheric air, 83. - Influence of oxygen on negative elements of battery, 84. - Reason why agitation of the negative element produces an increase of intensity, 85. - Experiments on the transitory currents of first immersion, 85–91 . - Affinity of zinc for oxygen , 91. - Of iron for oxygen , 92. - Of potassium for oxygen, 16. - Experiments to show

TABLE OF CONTENTS.
that electric intensity is expended in decomposing compounds and altering the physical condition of their elements, 93-97.- Experi ments to show the quantities of heat evolved by the combustion of metals, 97. — Heat evolved by the combustion of zinc, 98. - By the combustion of iron, ib. - By the combustion of potassium , 99.- By the combustion of hydrogen, ib . — Heat of combustion of equivalents of bodies proportional to the intensities of their affinities for gaseous
oxygen , 100-101.
ON THE ELECTRICAL ORIGIN OF CHEMICAL HEAT, 102-107.
Agreement of Joule's results with those of Dulong, 102–103.— Mode of obtaining theoretical results for copper, 104. - Corrections to be made for the heat expended in separating the acids from their bases, i6 . — Corrections made for zinc, iron, and copper, 105. -For hydrogen, 106. — Correction on account of light, 106–107.
ON SIR G. C. HAUGHTON'S EXPERIMENTS, 108–109 .
Explanation of the phenomenon observed by Sir G.C. Haughton.
ON THE HEAT EVOLVED DURING THE ELECTROLYSIS OF WATER,
109–123 ,
Object of the experiments, 109.- Description of the apparatus employed, 110. — Mode of ascertaining the resistance to electrolysis, 111. – Table of results of experiments, 112. — Excess of heat produced, 113. — Is not due to chemical change, 114. - Is due to resistance to electrolysis, 115. - Intensity due to separation of elements of water, 116. — Experiments to determine such intensity, 116-118. - Obstacles to the flow of the voltaic current three-fold , 119. - Mechanical and huating powers of currents interchangeable, 120. - Appendix on the
theory of heat, 121.
ON THE CALORIFIC EFFECTS OF MAGNETO -ELECTRICITY, AND ON THE
MECHANICAL VALUE OF HEAT, 123–159.
Introduction : the question for discussion , 123 . Part I. On the Calorific Effects of Magneto - Electricity.
Description of apparatus employed, 124–126. — Mode of experi menting, 126. — Precautions against atmospheric influence, 127. Results of first series of experiments, 128. - Second series of experi ments : apparatus employed, 129. — Mode of experimenting, 130. Results, 131. — Third series : results, 132. — Fourth series : results . 133. — Fifth series : results, 134. — Sixth series : results, 135. — Table of general results of the preceding series, 136. - General laws of

--
-

TABLE OF CONTENTS .

xi

evolution of heat deduced from experiments, 137. - Experiment on the heating-power of magneto -electricity beyond the coil, 138 .
On the Calorific Effects of Magneto- with Poltaic Electricity.
Mode of experimenting, 139. - Seventh series : results, 140. Eighth series : results 141. - Ninth series : results, 142. — Tenth series : results, ib . - Eleventh series : results, 143. — Twelfth series : results, ib . — Table of general results of series seven to twelve, 144.— Conclusions to be deduced therefrom , 145 .
On the Heat evolved by a Bar of Iron rotating under Magnetic Influence. Mode of experimenting, 146.- Thirteenth series : results, ib.
Fourteenth series : results, 147. – Fifteenth series : results, 148. Table of general results, ib. - General law deduced therefrom , 149.
1 Part II. On the Mechanical Value of Heat. Object of experiments, 149.- Description of apparatus, 150.
Results of repetition of previous experiments, 150-153. - Sixteenth series : results, 164. – Table of mechanical power, 155. - General law deduced therefrom , 156. — Mechanical equivalent of heat, ib. - Prac tical conclusions, ib. — Evolution of heat by friction, 157. — Equivalent derived from passage of water through narrow tubes, ib. — Appli cation to physiology, 158.

ON THE INTERMITTENT CHARACTER OF THE VOLTAIC CURRENT IN CERTAIN CASES OF ELECTROLYSIS ; AND ON THE INTENSITIES OF VARIOUS VOLTAIC ARRANGEMENTS, 159–171.
Irregular flow ofelectricity indicated by the unsteadiness of the gal vanometer, 159. — Experiments to show this phenomenon in the case of iron , 160. — The irregularity is accompanied by a change in the electrode, 161. - Further experiments, ib. - Effect of dividing the current between two electrolytic cells, 162. - Explanation of the phenomena, 163. - Table of results, 164. - Experiments with zinc, 165. — Distinction between phenomena of zinc and of iron, ib. Account of Schoenbein's experiments, 166. — Distinction between Schonbein's and Joule's experiments, 167. — List of voltaic inten sities, 168-169 . - Mode of obtaining the intensities, 170.

ON THE CHANGES OF TEMPERATURE PRODUCED BY THE RAREFACTION AND CONDENSATION OF AIR , 171-189.
Application of the principle of the conversion of mechanical force intoheat, 172. — Experiments of Cullen and Darwin, ib . — Dalton's mea surement ofthe change of temperature ofair, 173. — Uncertainty ofthe
measurement removed by placing the condensing apparatus in water,
174. - Description of the apparatus, ib. - Special thermometer used in the experiments, ib . — Experiments on condensation of air under

xii

TABLE OF CONTENTS .

water, 176. — Table of results, 177. — Calculation of the mechanical force expended , 178–179.Second series of experiments : table of
results, 179. - Calculation of the mechanical force, 180. - Apparatus for expansion of air, 180–181. — Table of results of simple expansion of air, 182. - Experiments with inverted receivers, 16. - Results of expansion which produced mechanical force, 184. - Fifth and sixth series : table of results, 185–186 . — Results correspond with theory that heat is a mode of motion, 186-187. - Application to the theory of the steam -engine, 188-189.

ON SPECIFIC HEAT, 189–192.
Importance of the law of specific heat, 189. - Results of previous investigations, 190. — Table of specific heats, 191. - Mode of calcula ting specific heats for the table, 192.

ON A NEW METHOD FOR ASCERTAINING THE SPECIFIC HEAT OF BODIES,
192-200 .
Defects of previous methods, 193. — Principles of the new method, 194. - Its accuracy, 195. – Application of it to the determination of specific heats in general, 196–197 . — Method with liquids, 198. With solids, ib.— With gases, 199.

NOTE ON THE EMPLOYMENT OF ELECTRICAL CURRENTS FOR ASCERTAIN ING THE SPECIFIC HEAT OF BODIES, 201.

ON THE MECHANICAL EQUIVALENT OF HEAT, 202. Abstract of the following paper.

ON THE EXISTENCE OF AN EQUIVALENT RELATION BETWEEN HEAT AND THE ORDINARY FORMS OF MECHANICAL POWER, 202–205.
Description of the apparatus employed, 203. - Experiments per formed, ib. - Equivalents obtained, 26. - Quantity of vis viva in matter, 204.

ON THE HEAT DISENGAGED IN CHEMICAL COMBINATIONS, 205–235 .
Prefatory observations, 205. - Object of the paper, 207. - Data previously established, 208. - Laws which govern the evolution of heat by electricity, ib . — New proofs of the laws, 211. — Description of the apparatus employed, ib. - Experiments on the heat evolved by currents traversing metallic wires, 214. – Table of results, 217.- Ex periments varied with currents of feebler tension, ib . — Table of results, 218. - Experiments on the effects produced by a change in
the resistance of the wire, ib.—Table of results, 219. - Repetition of

TABLE OF CONTENTS .
the experiments in roverse order, 220.- Table of results, 16. - Appli cation of the laws, 221. - Experiments to show the heat evolved by the combustion of copper, 222. — Tables of results, 224-226 . - Deter mination of the quantity of heat evolved, 227.--Combustion of zinc, 227. — Tables of results, 228-230 . — Determination of the quantity of beat evolved, 231.- Combustion of hydrogen gas, 231. – Tables of results, 232-233 . - Determination of quantity of heat evolved, 283– 234.- Summary of results, 234-235.
ON THE EFFECTS OF MAGNETISM UPON THE DIMENSIONS OF IRON AND STEEL BARS, 235-264 .
1. Suggestion that a magnetized bar of iron might be made a source of motive power, 235. - Researches by other investigators,
236. - Apparatus for showing the invariability of the bulk of an iron bar under magnetic influence, ib. — Experiments to ascertain the cause of the increase of length, 237. - Results from the first bar, 239.—
From the second bar, 240. – From the third bar, 241. - Elongation shown to be in the duplicate ratio of the magnetic intensity of the
bar, 242.- Experiments with unannealed bars, ib . - Results from the fourth bar, 243. - Results from soft steel wires, 244–245 . — Results from hardened steel bar, 246 .
II. Anomalous results from fine wire, 246. - Shown to be due to tension, 247. - Construction of apparatus to show the effects of tension and pressure, 248. — Experiment with annealed iron pillar, 249. — Results, 250. - Higher pressure on the same bar, 250–251. Experiment with soft iron pillar, 251-252. — Pressure hardly affects elongation, 252. - Experiments on tension with soft iron wire, 253 255. - Law resulting from the experiments, 256. - Shortening effects are probably proportional to the square root of the force of tension, 256-257 . - Experiments on tension with cast iron wire, 257. — Ex periments on pressure with soft steel wire, 258. — On tension, 250. Experiments on tension with hardened steel wire, 261.- Summary of results, 262. - Experiments on change of dimensions of iron at right angles to its polarity, 263. - Results from a piece of gas-tubing,
284.
ON MATTER, LIVING FORCE, AND HEAT, 265-276.
Definition of matter, 265. — The attraction ofgravitation one of its mostimportantproperties, 16. — Repulsion : inertia, 266. — Livingforce, or vis viva, ib . — How it is communicated to bodies, 287. - Indestructi bility of living force, 268. - Heat the equivalent of living force, 269. -Evidence thereof, 270. — Practical application of the equivalence, 271. - Wind, 272. - Shooting - stars, 1. - Animal heat, 273.- Real nature of heat, 273-276.

xiv

TABLE OF CONTENTS .

ON THE MECHANICAL EQUIVALENT OF HEAT, AS DETERMINED FROM THE HEAT EVOLVED BY THE AGITATION OF LIQUIDS, 276-277. Abstract of the following paper.

ON THE MECHANICAL EQUIVALENT OF HEAT, AS DETERMINED BY THE HEAT EVOLVED BY THE FRICTION OF FLUIDS, 277-281.
Experiments described, 278. — Table of results from friction of distilled water, 279.—Mechanical equivalent, 280. - Results from
friction of sperm -oil, 281. - Mechanical equivalent, ib .
ON THE THEORETICAL VELOCITY OF SOUND, 282-283 . Laplace's formula for the velocity of sound, 282. - Confirmed by
experiments on the specific heat of air, ib.

EXPÉRIENCES SUR L'IDENTITÉ ENTRE LE CALORIQUE ET LA FORCE
MÉCANIQUE. DÉTERMINATION DE L'ÉQUIVALENT PAR LA CHA
LEUR DÉGAGÉE PENDANT LA FRICTION DU MERCURE, 283–285. Summary of results of experiments, with table, 285. - Equivalent
deduced, 286 .
ON SHOOTING -STARS, 286–288 . Sir J. W. Lubbock's theories, 286. - Joule's theory, ib.-- Evidence
in favour of it, 287.
ON THE MECHANICAL EQUIVALENT OF HEAT, AND ON THE CONSTITU TION OF ELASTIC FLUIDS, 288-290. Fresh determination of the mechanical equivalent, 289. - Facts established by the experiments, ib .-- Application of them to calculate specific heat of gases, ib .

SOME REMARKS ON HEAT AND THE CONSTITUTION OF ELASTIC FLUIDS,
290–297 .
Statement of results previously obtained ( a) from the friction of fluids, 291; (6 ) from the rarefaction and condensation of atmospheric
air, 292. - Bearing of the results on the constitution of elastic Auids, 293. - Herapath's hypothesis, 294. - Determination of the velocity of
the particles of hydrogen, 2.- Determination of the specific heats of
several gases, 296.
ON THE MECHANICAL EQUIVALENT OF HEAT, 298–328.
Sketch of the history of the doctrine : Count Rumford's experi

-

1

TABLE OF CONTENTS .

XV

ments, 298. - Sir Humphry Davy's, 300. - Dulong's, 16. - Faraday's, 26. - Grove and Mayer's, ib. - Joule's, 301. – Necessity for further research, ib . - Description of the apparatus employed : thermometers, 302. — Machine for producing friction, 303. — Method ofexperimenting, 305. — Experiments on the friction of water : table of results, 306–309. - Determination of the increase of temperature, 309. - Absolute quantity of heat evolved, 310. - Force required, 311. - Equivalent for water, 312. - Experiments on the friction of mercury : table of results, 312–313 . - Increase of temperature, 314. — Absolute heat evolved, 314-318. – Force employed, ib . - Further experiments with mercury : table of results, 319, 320. — Increase of temperature, 321. - Force em ployed, ib.- Experiments on friction of cast iron, 321. – Table of results, 322. - Increase of temperature, 322.- Absolute heat evolved, 323. – Force employed, 325. — Equivalent for cast iron, 18. – Further experiments on cast iron, 16. — Table of results, 326. — Increase of tem perature, 16. - Force employed, 827. - Summary of the above deter
minations of the equivalent, ib . - Formulæ for mechanical equivalence,
328. - Friction consists in the conversion of mechanical power into heat, ib.

ON A REMARKABLE APPEARANCE OF LIGHTNING, 329-330.

ON BOME AMALGAMS, 331. The mode of their formation, and a list of those obtained .

ON THE AIR - ENGINE, 331-356.
Introductory, 331.D-escription of the air-engine, 332. — Mode of working, ib . — Thomson's formulæ , 334 — Estimates of the performance of the engine under various conditions, 335. — Superiority of air -engine over steam -engine, 337. - Tables of work absorbed, 338. - Examples of engines, 339. - Hints on the construction of air-engines, 340-341.
Note to the foregoing paper, with a New Experimental Determination of
the Specific Heat of Atmospheric Air, 342–349.
Probable inaccuracy of received value of specific heat of air, 842. New experiments to determine value, ib . — Tables of results, 344-345.
-True value of k , 346. - Now tables and examples, 348 349.
Additional Note on the Preceding Paper. By [Sir ] WILLIAM THOMSON,
350-356 .
1. Synthetic investigation of the Duty of a perfect Thermodynamic Engine, founded on the Expansions and Condensations of a Fluid for which the gaseous laws hold, and the ratio (k ) of the specific heat under constant pressure to the specific heat in constant volume is

xvi

TABLE OF CONTENTS .

constant; and modification of the result by the assumption of Mayer's hypothesis, 350-353.
2. Note on the Specific Heats of Air, 353-356.

ACCOUNT OF EXPERIMENTS WITH A POWERFUL ELECTRO -MAGNET,
357-362.
Description of the magnet, 357. – Battery, 358. - Experiments (a) with a magnetic needle, 359 ; (6 ) a bar of bismuth , ib.; ( c) piece of soft iron, 360 ; ( d) flat bar of soft iron , ib.; ( e) another bar, 361 ; ( f ) magnetic needle, 362.

ON THE ECONOMICAL PRODUCTION OF MECHANICAL EFFECT FROM CHEMICAL FORCES, 363-368.
Classes of engines which derive their power from the operations of chemical force, 363. - The animal frame an engine, ib . - Formula for work done by the electro -magnetic engine, 364. - For work done by steam - engines, 16. - Thomson on a perfect thermo-dynamic engine, 365. - Construction of air-engine, 366. - Amount of work done by it, 16. - Comparison of it with steam -engine, 367. - Suggested improve ment in the air- engine, 868.
AN ACCOUNT OF SOME EXPERIMENTS WITH A LARGE ELECTRO MAGNET, 368 369.

INTRODUCTORY RESEARCH ON THE INDUCTION OF MAGNETISM BY
ELECTRICAL CURRENTS, 369-381.
Doubts as to the validity of Jacobi and Lenz's law , 369. Apparatus employed, 370. - Air resistance, 371. - Magnetism in duced by the earth , 371-372. - Magnetism induced by electric currents, 373-381. - Influence of the magnetic set, 374. - Law of mag netic set, 375, 376. - Magnetic breaking -point and elasticity, 379. Attraction of contact, 379_381.

ON THE FUSION OF METALS BY VOLTAIC ELECTRICITY, 381-384. Present methods of welding iron, 381. — Possibility of welding it
by electricity, 882. -Confirmed by experiment, 16.-Estimated cost of the process, 382-384.
NOTE ON DALTON'A DETERMINATION OF THE EXPANSION OF AIR BY HEAT, 384-385 .
Regnault's criticism of Dalton , 384. - Shown to be based on a misapprehension, 385.

1

TABLE OF CONTENTS .

xvi

ON THE UTILIZATION OF THE SEWAGE OF LONDON AND OTHER LARGE Town , 886-999.
Criticism of the system of the Metropolitan Board of Works, 386. -History ofthe Sewage question , i . - Principles and outline of the scheme of the Metropolitan Board, 387. — The scheme ignores the utilization of sewage, 389; and will fail to attain its object, 390. The utilization of sewage is practicable, 302. — Magnitude of the loss sustained by wasting sewage, 393 .--Means of preventing wasto : (i) the exclusion of organic matter, 396 ; ( ü ) mode of dealing with the sewage, 397.

NOTICE OF EXPERIMENTS ON THE HEAT DEVELOPED BY FRICTION IN AIR , 399-402.
Results of whirling a thermometer, 899. - Experiments with lathe, 16.- Results, 400. - Further experiments, i6. - Summary of results, 401. - Velocity required for 1º Centigrade, ib.

ON THE INTENSITY OF LIGHT DURING THE RECENT SOLAR ECLIPSE ,
402 403.
Comparison of photographs taken during and after the eclipse, 402. -Inference as to luminosity of the circumference of the Sun's disk ,
403 .

ON AN IMPROVED GALVANOMETER, 404–405.
Description of the instrument, with plan.
ON THS THERMO -ELECTRICITY OF FERRUGINOUS METALS ; AND ON THE THERMAL EFFECTS OF STRETCHING SOLI BODIES, 406 407. Apparatus, 405. — Results with steel and cast iron, 406. — Effect of stretching on metals, wb. - On gutta -percha and vulcanized india rubber, ib. - Confirmation of Thomson's theory, 407.
ON THE THERMAL EFFECTS OF LONGITUDINAL COMPRESSION OF SOLIDS; WITH AN INVESTIGATION ON THE ALTERATIONS OF TEMPERATURE ACCOMPANYING CHANGES OF PRESSURB IN FLUIDS, 407-412.
Verification of Thomson's theory, 407. — Is the force of elasticity in metals impaired by heat P, i . - Experiment with steel wire, 16. With copper wire, 408. — Thomson on the Alterations of Temperature accompanying changes of Pressure in Fluids, 409-412 . - His Table of thermal effects of a pressure of ten atmospheres on water and
mercury , 412.
b

xvii

TABLE OF CONTENTS .

ON SOME THERMO -DYNAMIC PROPERTIES OF SOLIDS, 413-473.
Introductory.
$ 1. An elastic fluid conceivable which would not give out heat by compression, 413 .- $ 2. Apparatus to verify this hypothesis, 414. $ 3. Results of experiments, 415.—$ 4. Change of temperature inap preciable, ib.- $ 6. Experiments on the heat developed by longitudinal compression, and that absorbed on the application of tensile force, 416 . -$ 6. Description of the thermo-multiplier used in the experiments, ib. — 87. Arrangement of the astatic needles, 417.—$ 8. Action of the earth's magnetism counteracted , 418.- 9. Rarefaction of air diminishes effect of currents, i6.- 10. Air obviates oscillation of the needle, 419.—$ 11. Test of the sensitiveness of the instrument, ib. - 12. Method ofdispensing with air-pump after first exhaustion , ib.

Thermo-electricity of Iron in different States.
$ 13. Thermo -electric relation between hard steel and iron, 420. § 14. Changes of thermo -electric position in metals, ib.-- $ 15. Range of intensities of cast iron , 421.
Thermo -electric Intensities of Metals, Alloys, etc.
$ 16. Mode of observation, 422.- 17. Scale of thermo-electric intensities at 12° Centigrade, ib.
Experiments on the Thermal Effects of Tension on Solids.
§ 18. Methods of thermo-electric junction, 423.- 19. Thermome tric value of the deflections, 424.— $ 20, 21. First experiment with iron , 425. - 22. Valuation of the result, ib.- $ 23. Second experi ment, 426.- 24. Third experiment, 16.— $ 25. Experiments with hard steel, ib.— $ 26, 27. With cast iron, 427.— $ 28, 29. With copper, 16. - 30. With lead, 427. - 31, 32. With gutta-percha, 428.- $ 33. India -rubber : extraordinary physical properties of the substance, 429. -$ 34. Its conditions ofrigidity, ib.— $ 35. Effect of tension of it, 430. - $ 36. Influence of low temperature on the experiments, ib.— $ 37, 38, 39. Tension of india -rubber in water, 431.— $ 40. Agreement of results with Thomson's formula, 432.- $ 41. Effect of tension on rigid india-rubber, 433.~ $ 42. “ Vulcanized ” india-rubber : superior permanency of its elasticity, ib . - S 43. Determination of its specific heat, ib.- $ 44. Determination of its expansion by heat, ib.- $ 45 . Gough's theory of the specific gravity ofindia -rubber corrected, ib. $ 46, 47. Thermal effects of stretching vulcanized rubber, 434 .— $ 48. Inferences from the experiments, 435 .- $ 49. Modified result from using a higher temperature, ib.- $ 50. Effect of increase of tempera

TABLE OF CONTENTS .

xix

ture on india -rubber under tension, ib ., 436.— $ 51, 52, 53. Comparison of theoretical and experimental results, 437. - Comparative table of results, 438.— $ 54. Further experiment, 437.- $ 55. Elimination of
the set of the india -rubber, 439 .— $ 56. Effect of removal of tension, ib. 57. Experiments with greater tension, 16.- $ 58. Results of experiments, b .--- $ 59. Experiments with wood, 440 .— $ 60. Descrip tion of apparatus, ib.— $ 61. Mode of securing even temperature, 441. - 62. Bay wood : table of results, 442.— $ 63. Effect of tension on the expansibility of the wood , ib.— $ 64. Further experiments, ib . $ 65. Similarity of wood to indis -rubber, 443.- $ 66. Relation of
elasticity to increase of temperature, ib.— $ 67. Modification of the
phenomenawhen tension is applied, 444.— $ 68. Evaluation of Young's modulus of elasticity, 445.— Table of results, 445.— $ 69. Effects
of hygrometric condition on expansion of wood, ib.— $ 70. Experi
ments with deal, 446. - 8 71, 72. With deal cut across the grain, 447. - $ 73, 74. Experiments with another rod, 448.- $ 75. Deductions
from the foregoing experiments, 450.- $ 76. Strain in passing from dry to wet state, and vice versd, 16.— $ 77. Specific gravity of wood decreases with increase of temperature, 451.- $ 78. Such decrease is due to capillary attraction, ib.— $ 79. Experiments with Ratan cane, 452 .— $ 80. Experiments on rose, pine, and poplar woods, 453.— $ 81. Cane increases in length when moistened, 454. 82. Experiments with paper, leather, and whalebone, 454-455.- $ 83. Effect of mois ture on the expansibility of paper and leather, 454. – S 84. Qualities of whalebone, 16.— $ 85. Its elasticity, ib.- $ 86. Thermal effect of
tension on wood : experiments on pine wood, 456.— $ 87. Determi nation of its specific heat, 457.- $ 88. Theoretical result, 457.— $ 89. Experiment on bay wood, ib.— $ 90. Theoretical result, ib.- 91. Pine cut across the grain , 457. ~ $ 92. Bay wood saturated with water, 458. — 5 93. Comparison of theoretical results with those of above
experiments, 18 .

On the Thermal Effects of Longitudinal Compression on Solids.
$ 94. Experiment with wrought- iron pillar, 459.-- 95. Second experiment, ib.-- 96. Experiment, using higher pressure, 1.- 97, 98,99. Similar experiments,460.- $ 100–103. Cast iron, 460.- $ 104 106. Copper, 462.- $ 107. Lead, 463.- 108. Glass, 464.- $ 109. Pine, ib.- $ 110–113. Cube of pine, 465 .— $ 114-118. Vulcanized india rubber, 466.- $ 119. Table of results of the above experiments, 468. -$ 120. Discussion of the excess of experiment over theory, 469. $ 121. Sources of error, 470 .-- $ 122, 123. Thomson's theory of springs subjected to experiment, 471.-$ 124. Experiments described, ib. Table of results, 473 .- $ 125, 126. Agreement of Thomson's theory
and experiment, 472.
62

TABLE OF CONTENTS .
ON THE THERMAL EFFECTS OF COMPRESSING FLUIDS, 474-479. Thomson's theory, 474. - Regnault's experiments, 16. - Apparatus
described , 475.---Method of experimenting, ib.Sources of error, 476 . --Table of results of compression of water, 478 ; of sperm -oil, 479.
ON A METHOD OF TESTING THE STRENGTH OF STEAM -BOILERS, 480–
481.
Improved mode of testing, 480. - Defects of the ordinary method, ib . — Table of observations of pressure, 481.
EXPERIMENTS ON THE TOTAL HEAT OF STEAM , 482-484.
Definition of total heat, 482. - Mode of experimenting, ib.- Table
of results, 483.
EXPERIMENTS ON THE PASSAGE OF AIR THROUGH PIPES AND APERTURES
IN THIN PLATES, 485–490. Conduct of fluids on flowing through a hole in a thin plate, 485.
Apparatus and experiments on the phenomena, ib.-- Tables of results, 486-487. - Results disturbed by bur on the edge of the hole, 487. Influence of tubes of various lengths and diameters, 488. — Experiment with hollow cone, ib . — Effect of vibration, 489. – Summary, ib.
ON SOME AMALGAMS, 490-500. Amalgams of iron,490. - Mode ofproduction, 491. - Characteristics,
492. - Amalgam of copper, 493 495. - Amalgam of silver, 496.
Amalgam of platinum , 497.- Amalgam of zinc, 498. - Amalgam of lead , ib.--Amalgam of tin , 499. - Amalgam of hydrogen, 500.
ON THE PROBABLE CAUSE OF ELECTRIC STORMS, 500-502.
ON THE SURFACE -CONDENSATION OF STEAM , 502–531.
Scope of the inquiry, 502. - Description of the apparatus used, 503 506. — Method of experimenting, 06.- Corrections for cooling, etc., 507. - Explanation of the table of results, 508. — Thomson's formula of conductivity, 509-510.-- Its agreement with experimental results, 511. - Effect of the stoppage of condensation, 512.- Experiments on
the relation of conductivity to the quantity of water transmitted through the tubes, 513-514.---Averageof results, 615. - Experiments
on the influence of the waterspace, 615-516 . - Effect of cleanliness of
surface, 16. — Effect of a solid in the axis of the steam - tube, 517. Effect of changing the direction of transmission of the refrigerating

-

TABLE OF CONTENTS.

xxi

water, 517–519 . — Effect of spiral direction of the water, 618_521. Comparative effect of transmitting the refrigerating water with, or against, the steam , 521–523 . — Comparison of plain tubes with tubes furnished with spirals, 623–527 . — Summary of the principal results, 527. — Experiments with air as the refrigerating agent, 528–529. Results, 530. - Suggestions for future research, 531.

NOTICE OF A COMPRESSING AIR -PUMP, 531-532.
NOTE ON A MIRAGE AT DOUGLAS, 532–534 .
Description of the mirage, by Dr. Joule, 532.- Description of a similar phenomenon on Skiddaw , by Dr. Angus Smith, 533.

ON A SENSITIVE BAROMETER , 634 . Brief description of the instrument.
ON A SENSITIVE THERMOMETER, 536–536. Description of the instrument.
NOTE ON THE METEOR OF FEBRUARY BTH, 1818 , 536–539. Account of the ineteor, 536. - Probable cause of meteors, 537.
Data respecting them , ib.--Accordance of hypothesis with the facts,
638–539.
ON A METHOD OF HARDENING STERL WIRES FOR MAGNETIC NEEDLES,
540 .
Description of method .
ON AN INSTRUMENT FOR SHOWING RAPID CHANGES IN MAGNETIO
DECLINATION, 540-542. Description and figure of instrument.
DETERMINATION OF THE DYNAMICAL EQUIVALENT OF HEAT FROM THE
THERMAL EFFECTS OF ELECTRIC CURRENTS, 542-567.
Delay in making experiments, 642. — Apparatus used, 543.-
Difficulty in determining the distance between the fixed coils, ib.
Galvanometer and calorimeter described , 544. — Mode of ascertaining
the resistance of the wire, ib .First series of thermal experiments : results, 646. - Radiation experiments : results, 547. - Evaluation of

xii

TABLE OF CONTENTS .

the dynamical equivalent, 547-548 . — Second series: results of thermal experiments, 549 ; of radiation experiments, 550.-- Evaluation of dynamical equivalent, 550-551.- Suspected sources of error, 551–552. - Third series : results of thermal experiments, 564 ; of radiation ex periments, 555. - Determinations of horizontal magnetic intensity 556. - Evaluation of equivalent, 657. – Result of the inquiry, ib.

OBSERVATIONS ON THE ALTERATION OF THE FREEZING -POINT IN THER MOMETERS, 658–559 . Table of changes of freezing -point observed by Dr. Joule.

ON A NEW BALANCE, 559-561.
Description and figure of the instrument.

ON AN APPARATUS FOR DETERMINING THE HORIZONTAL MAGNETIC INTENSITY IN ABSOLUTE MEASURE , 561-567.
Weber's apparatus, 561.I-mproved form of it, 562.-Description of new apparatus, 563. — Mode of observation, 564. - Determination of the time occupied by the vibrations, 565.D—eterminations of hori zontal magnetic intensity , 566.

NOTE ON THE TANGENT -GALVANOMETER, 567-568.

ADDITIONAL NOTE ON THE GALVANOMETER, 668–570. ON A SELF -ACTING APPARATUS FOR STEERING SHIPS, 570_573.
Description of an electromagnetic apparatus for steering a ship automatically in a given direction.
ON A THERMOMETER UNAFFECTED BY RADIATION, 573–574. Description of the instrument, with figure.

NOTE ON THE RESISTANCE OF FLUIDS, 574–575 .
Coulomb's theory, 574 .-- Apparatus for verifying it, ib . - Results of experiments, 574–575 .

ON A NEW MAGNETIC DIP - CIRCLE, 575–576.
Description of an improved instrument.

NOTES ON THE ABOVE, WITH EXPERIMENTS, 577-583.
Improved shape of the instrument, 577. - Experiments on the strength of silk and spider filaments, 579. - Its increase with rise of temperature, 583.

TABLE OF CONTENTS.

Xxiii

OBSERVATIONS WITH THE NEW DIP -CIRCLE, 583.

NOTICE OF, AND OBSERVATIONS WITE , A NEW DIP - CIBOLE, 584. Brief summaries of papers.

DESCRIPTION OF AN ELECTRIC -CURRENT METER, 584-589.
Description and figures.
ACCOUNT OF EXPERIMENTS ON MAGNETS BEGUN IN 1864 AT OLD
TRAFFORD, 589-604 .
Scoresby's investigations into the magnetical properties of steel, 589. - Construction of magnetic systems with common needles, 590. Summary of observations on their deflecting -powers, 591. - Cause of the gradual deterioration of the magnets, 591. - Loss of deflecting power of needles when combined in a system , 593. — Experiments on the effect of changes of temperature on the intensity of magnets, 594 595. - Experiments on the effect of mechanical violence on the in tensity of magnets, 595-597 . - Experiments on the loss and gain of magnetic intensity by alternations of temperature, 597–603. — Similar experiments with loadstones , 604.

ON SOME PHYSICAL PROPERTIES OF BEES'-WAX, 605–606.

ON SOME PHOTOGRAPHS OF THE SUN, 606-607.
ON SUNSET SEEN AT SOUTHPORT, 607.
ON THE ALLEGED ACTION OF COLD IN RENDERING IRON AND STERL BRITTLE, 607-610.
Comparison of the behaviour of cold and warm needles under tension, 608. — Comparison of the behaviour of cold and warm garden nails under a blow , 609-610.
FURTHER OBSERVATIONS ON THE STRENGTH OF GARDEN NAILS,
611-613.
( a) Behaviour of cast - iron under pressure ; ( b) Effect of hardness on the elasticity of iron.
EXAMPLES OF THE PERFORMANCE OF THE ELECTRO -MAGNETIC ENGINE,
613-619 .
Criticism of earlier results, 613.A-rrangement of the engine, 614. - Performances of the engine under various loads, 614–616 . — Addi
tional power supplied to the engine, 616. - Its performances under

xiv

TABLE OF CONTENTS.

various loads, 616-617 . – Further improvement of the engine, 618. Results, ib .- Application of principles to case of reversed velocity, 618. - Scope for improvements, ib. — Reason why the actual duty is
less than the theoretical, 619.

ON THE MAGNETIC STORM OF FEBRUARY 4 , 1872, 619-620.
Observations on the deflections of the needle during the storm .

ON THE POLARIZATION OF PLATINA PLATES BY FRICTIONAL ELEC . TRICITY, 620-622. Précis of some experiments.
Note added in 1883 .
Brief description of the experiments, 621–622 .

ON THE PREVALENCE OF HYDROPHOBIA, 622-623.
Hydrophobia can be stamped out by isolation, 622. - Another
possible preventive, 623.
ON A MERCURIAL AIR -PUMP, 623-627.
( a) Improvements in the air-pump, 623.- Plan of the pump, 624. (6 ) Further improvements, 625, 627. — Plan of pump, 626 . ( c) Improved glass plug, 627.

ON A GLUE -BATTERY, 628 .
Description of a battery in which glue impregnated with salts of zinc and copper is used.

ON THE UTILIZATION OF THE COMMON KITE, 628 629. Defects of the common kite, 628. — Improved form of kite, 629.
ON A BAROMETER , 629-631. Description of an improved barometer.

METHOD OF CHECKING THE OSCILLATIONS OF A TELESCOPE, 631.
NEW DETERMINATION OF THE MECHANICAL EQUIVALENT OF HEAT,
632-657.
Discrepancy between results previously obtained, 632. - Investi
gations to discover the cause of it, 632. - Hirn's experiments, 632–633. -Apparatus used in the experiments, 634.— Measurement of the

TABLE OF CONTENTS .

XXV

groove of the calorimeter, 685.T-hermometers used, 636-637. Thermal capacity of the calorimeter, 637. — Tabular results of expo riments to ascertain the thermal capacity , 639. — Verification from the sum of the capacities of the materials composing the calorimeter, 840-643. — Mode of experimenting for the mechanical equivalent, 643-844. — First table of ( a) Experiments to ascertain the effect of radiation , 645 ; (6) Experiments with friction of water and brass, 646. - Corrected results, 647.- Second table : ( a ) Experiments to ascertain the effect of radiation, 648 ; (6 ) Experiments with friction
of water and brass, 649.- Variation of method , 650. — Third table : ( a) Experiments to ascertain the effect of radiation, 650 ; (6) Expe riments with the friction of water, 652.F-ourth table : ( a ) Experi
ments to ascertain the effect of radiation, 653 ; (6 ) Experiments with almost solely friction of water, 654. – Fifth table : (a) Experiments to ascertain the effect of radiation, 655 ; (6) Experiments with almost solely friction of water, 656. - Summary of results, 657.

-

LIST OF PLATES AND ILLUSTRATIONS.

Fig .

Page

1. Magnet for a magneto -electric engine.

2. Arrangement of magnets in the engine

2

3. Electro -magnetic engine

3

4. Electro-magnetic engine • • •

5

6. Section of hollow magnet

8

6. Section of solid magnet of same area

in .

7. A now galvanometer ..

10

8. A new magneto -electric engine

17

9. Commutator of above....

ib .

10. Holder of above .....

ib .

11. Electro -magnet of above

2.

12, 13, 14. Views of the attractions of magnetic particles

19

16. Ditto ....

20

16. A new galvanometer .

29

17. An electro-magnet ...

31

18. Section showing three cells of a voltaic battery

88

19. An electro -magnet .

40

20. Perspective view of a now electro-magnet .

43

21. Horizontal section of the same from above ...

ib .

22. The same, seen from below

44

23. Mode of suspending the magnet

45

24. Wire coiled on a ring......

is.

25, 26, 27. Sections of atom of iron surrounded by atmospheres of

electricity

51

28. A copper-zinc battery

55

29. Zinc element of the above ....

56

30. Copper element of the above

2,

31. The same battery with a large number of small plates....

it .

: viii

LIST OF ILLUSTRATIONS.

Fig.

Page

32. A new galvanometer

81

33. Apparatus for ascertaining the heating- power of a metallic wire 62

34. Arrangement of a voltaic pair

68

35. Battery with galvanometer and decomposing apparatus

73

36. Ilustration of the formula for determining the true resistance to

conduction

.

75

37. Apparatus for experiments on the heat of combustion

88

38. Illustrating the relation between chemical affinity and the electric

current

90

39. Battery, galvanometer, and decomposing apparatus.

93

40. Projection of quantities of electricity evolved by battery

94

41, 42. Ditto

95

43. Apparatus for burning metal in oxygen

98

44. A revolving electro -magnet...

125

45. Electro-magnet used in connexion with the preceding

126

46. Iron for an electro -magnet

129

47. The preceding in completed state

2b .

48. Representation ofthedirection andintensity of magneto-electricity 138

49, 50. Representations of magneto - electrical currents.

145

51. Apparatus for revolving the magnet of fig .44 ....

150

52. Apparatus for observing the changes of temperature produced by

the rarefaction and condensation of air

173

53. Representation of force of condensation

178

64. Copper receiver used in experiments on the rarefaction and con

densation of air .

180

55. Section of Ash's stopcock

181

56. Receiver of fig. 54, inverted

183

57. Mode of experimenting with the preceding .

ib .

58_66. Apparatus employed in the Experiments on the Heat disen .

gaged in Chemical Combinations. (Plate I.)

205–235

87. Apparatus for observing the effects of tension and pressure on

iron and steel bars

247

68. Apparatus for observing the effects of magnetizing an iron tube 263

69–77. Apparatus used in the Experiments on the Mechanical

Equivalent of Heat. ( PLATE II.) ...

298-328

77 bis. Remarkable appearance of lightning

329

78-82. Apparatus used in the Experiments on the Air -engine.

(PLATE III.)

331-356

1

LIST OF ILLUSTRATIONS.

xxis

Fig .

Page

83. Iron for a new electro -magnet

857

84. The same, completed ....

358

85. An improved galvanometer

404

86. Apparatus for observing the variations of temperature of a spring 414

87. Thermo-multiplier ...

417

88. Mode of poising astatic needles

418

89. Arrangement to secure permanent vacuum in a thermo

multiplier ....

420

90. Thermo-electric junctions .

424

91. Apparatus for examining tension in india -rubber

431

92. Apparatus for examining tension of wood

440

93. Apparatus for ascertaining the surface-condensation of steam .. 504

94. Instrument for observing changes in magnetic declination .... 541

96. Projection of the changes of freezing -point in thermometers .. 559

96. A new form of balance ...

560

97. A form of thermometer unaffected by radiation .

573

98. Magnetic dip - circle

578

99. Electric - current meter

585

100. Coil for the above ....

586

101, 102. Other coils for the same

587

103, 104. Sections of magnetic systems of one- inch needles

590

105, 106. Sections of magnetic systems of half-inch needles...

ib .

107. Side view of a system of half-inch needles. 108. Sunset at Southport 109. Mercurial air-pump

ib . . 607
624

110. The same, improved

626

111. Glass plug of air -pump

627

112. Improved form of barometer ...

630

113-119. Apparatus used in the Experiments for the new determi

nation of the Mechanical Equivalent of Heat. (PLATE IV .) 632-657

SCIENTIFIC PAPERS.
Description of an Electro-Magnetic Engine.
[ From Sturgeon's ' Annals of Electricity,' vol. ii. p. 122.]
SIR ,
I am now making an electro -magnetic engine ; and as I imagine that I have succeeded in effecting considerable improvement in the construction of the magnets, and the whole arrangement of the instrument, I hope you will allow me to lay it before the numerous readers of your valuable
' Annals .'
In fig. 1, ef represents a side view , and b b the poles, of the
Fig. 1. Full size.
e
с
0 08
d
magnet I propose, the distance a between the poles being about
B

8

2

DESCRIPTION OF AN

the fifth part of an inch, and the thickness of each pole or arm of the magnet the same, or perhaps rather less. If the magnets are required to be of greater power, the breadth cd, or length e f, should be increased , but the thickness left the same. Covered wire is wrapped round until the space between the arms is completely filled .
The advantages obtained by using magnets of the above description may be seen on inspecting fig. 2, where the small arrows indicate the direction that the electricity takes, in passing from P to N.

Fig. 2. Full size.

0000

1. The wire round each arm is kept close to the substance
of the iron .
2. The greater part of each arm receives the magnetizing effect of the wire wound on its neighbours as well as
of that belonging to itself. 3. A great saving of room is effected .
4. The objection which applies to arrangements in which the poles are far distant from one another, viz. that the magnets during a great part of their rotation are almost inactive, is obviated .
I have made several electro -maguets of the above con struction . Their lifting power is very good. The spark on breaking battery contact is, however, remarkably brilliant, which may be considered in some respects disadvantageous.
In the engine the electro -magnets I have described are to be arranged in two compact circles, one of which is repre sented by fig. 3. A stout board a b c supports one circle of electro -magnets, the wires of each being attached to those of
its neighbours, so that the whole may be magnetized at once

ELECTRO -MAGNETIC ENGINE .

8

by a current of electricity from h to i. The opposite, or revolving circle is to be magnetized, and the polarity reversed , by means of the commutator g, which consists of two cogged

Fig. 3. Half size.

m

ho lio

000w0प0ा0w00s000

6

C

circles of bright brass communicating with the battery by the wires kl. The electrodes of the commutator, m n , are springs, which, in their revolution, press gently on the circles of
brass.
It will be seen that by a proper use of the clamps h, i, k, 2 the current may be made to go consecutively through the fixed and movable circles of electro -magnets, or that a distinct current may traverse each .
The axle carrying the movable circle is supported by the bearing S, and may be used to turn any kind of machinery.
I am , yours truly,
J. P. JOULE .
Salford, January 8, 1838.
B2

4

EXPERIMENTS WITH AN

Description of 'an Electro-magnetic Engine, with
Experiments. (In a letter to the Editor of the
• Annals of Electricity' *.)
[ Annals of Electricity,' vol. iii. p. 437.)
DEAR SIR, In vol. ii. p. 122 of your interesting work is a com
munication of mine describing a method of making electro magnetic engines, which I thought might be adopted with advantage. I finished the one I was working at during last summer. It weighs 7} lb.; and the greatest power I have been able to develop with a battery of forty -eight Wollaston
four- inch plates was to raise 15 lb. a foot high per minute, in which estimate the friction of the working parts, which
was very considerable, was reckoned as the load.
The result shows that the advantages of a close arrangement of electro -magnets are not such as I anticipated .
I was desirous, before attempting to make another engine, to satisfy myself by experiment how far it was possible to increase the velocity of rotation, which was only 31 feet per second in the above trial. Now , of the many things which limit the velocity, the resistance which iron opposes to the instantaneous induction of magnetism is of considerable importance. I think I shall be able to show how this may be obviated in some measure .
The current of electricity produced by a magneto -electric machine is much increased by the insertion of a bundle of
iron wires, instead of a solid nucleus of iron, into its coil — a phenomenon evidently occasioned by the peculiar texture of the wires, which allows Mr. Sturgeon's magnetic linest to collapse with greater suddenness.
With a view to determine to what extent the velocity of rotation could be increased by the use of wire magnets, I
• The experiments were made at Broom Hill, Pendlebury, near
Manchester,
+ See Annals of Electricity,' vol. i. pp. 251, 277.

ELECTRO -MAGNETIC ENGINE .

5

constructed an apparatus represented by fig. 4 , where ab , cd are steel magnets ; e, f,brass screws with small holes in their

Fig. 4. Scale . f.

m
9

m
w

ends to receive the fine points of the steel axle on which the electro -magnet m m is fixed ; g , h are mercury - cups to connect the wires of the electro-magnet with the pieces of watch
spring i, k which dip into two semicircles * of mercury consti tuting the commutator ; w, w are wires connecting the com mutator with the battery. I had four electro-magnets, which, with their axles & c., could , with great expedition, be put off or on the machine by means of the screw f.
No. 1 electro -magnet was made of a bar of round iron of 1090 grs. weight; No. 2 was a bundle of nineteen iron wires
of about to inch diameter : no particular pains were taken
to anneal the iron of these magnets. No. 3 and No. 4 were
made of iron bar and wires of the same quality and dimen sions, but annealed to great softness. Each of the above was
first enveloped with a double covering of muslin , and then
wound with eleven yards of copper wire 7 inch in diameter
covered with silk . Care was taken to make the friction of
the pivots equal.
* These semicircles were grooved into the wood. As the convex surface of the mercury stood above the level, the steel springs crossed
edgewise over the partition without obstruction or splash.—Note,
1881.

6

EXPERIMENTS WITH ELECTRO -MAGNETS

The following results, in revolutions per minute, were obtained with the above apparatus

No. 1. No. 2. No. 3. No. 4 .

With a single constant cell.......
With two constant cells arranged for intensity ..
Ditto. With weak charge ...
Average

146 177 196 192 233 274 283 321
196 173 224 209 192 208 234 241

The sparks and shocks, on breaking battery circuit, were

just sensible in No.1, twice as great at least in Nos. 2 and 3.

In No. 2 they were a little greater than in No. 3, but were

by far the most brilliant and powerful in No. 4.

I intend to make another engine presently with magnets

made of rectangular wire.

Yours truly,

Salford , December 1, 1838.

J. P. JOULE .

On the use of Electro-magnets made of Iron Wire for
the Electro-magnetic Engine. (In a letter to the Editor of the · Annals of Electricity '* .)
[ Annals of Electricity,' vol. iv. p. 58.] DEAR SIR ,
In my last letter I gave you an account of some expe riments which were thought to prove that electro -magnets
made of iron wire were the most suitable for the electro
magnetic engine. In those experiments the ordinary round wire was used ; and it was my opinion that these wire magnets were at a disadvantage in consequence of the
interstices between the wires. I therefore arranged fresh experiments as follow :
The experiments were made at Broom Hill, Pendlebury, near
Manchester.

|

CONSTRUCTED OF IRON WIRE .

7

I constructed two electro -magnets. The one consisted of 16 pieces of square iron wire, each I of an inch thick and 7 inches long, bound tightly together so as to form a solid mass whose transverse section was t inch square ; it was then enveloped by a ribbon of cotton, and wound with 16 feet of covered copper wire of to inch diameter. The other was made of a bar of solid iron, but in every other respect was precisely like the first. These electro -magnets were succes
sively fitted to the apparatus used in my last experiments,
care being taken to make the friction of the pivots equal in each . On trial, the means of several experiments gave 162 revolutions per minute for the wire magnet, and 130 for
the bar magnet.
In the further prosecution of my inquiries, I took six pieces of round bar iron of different diameters and lengths, also a
hollow cylinder 1 inch thick in the metal. These were bent in the U - form , so that the shortest distance between the poles
of each was half an inch : each was then wound with 10 feet
of covered copper wire to inch in diameter. Their attractive powers under like currents for a straight steel magnet 1} inch long, suspended horizontally to the beam of a balance, were, at the distance of half an inch, as follow :

No. 1. No. 2. No. 3. No. 4. No. 5. No. 6. No. 7. Hollow . Solid . Solid . Solid . Solid . Solid. Solid .

Length round the

bend, in inches

6

Diameter, in inches .

Attraction for steel
magnet, in grains 7-5

Weight lifted, in

ounces .

36

51 2 57

23

*

1

6:3 5.1 5.0 4.1 4:8 3.6

52 92 36 52 20 28

64
cahtacvtoe

A steel magnet gave an attractive power of 23 grains, while its lifting -power was not greater than 60 ounces .
The above results will not appear surprising if we consider, first, the resistance which iron presents to the induction of

8

EXPERIMENTS WITH ELECTRO -MAGNETS

magnetism , and, second, how very much the induction is exalted by the completion of the ferruginous circuit.
Nothing can be more striking than the difference between the ratios of lifting to attractive power at a distance in the dif
ferent magnets. Whilst the steel magnet attracts with a force
of 23 grains and lifts 60 oz., the electro -magnet No.3 attracts with a force of only 5: 1 grains, but lifts as much as 92 oz .
To make a good electro -magnet for lifting -purposes : - 1st. Its iron , if of considerable bulk , should be compound, of good quality, and well annealed . 2nd. The bulk of the iron should bear a much greater ratio to its length than is generally the case . 3rd . The poles should be ground quite true, and fit flatly and accurately to the armature . 4th. The armature should be equal in thickness to the iron of the magnet.
In studying what form of electro -magnet is best for attrac tion from a distance, two things must be considered , viz . the length of the iron , and its sectional area .
Now I have always found it disadvantageous to increase the length beyond what is needful for the winding of the covered wire. Then as to the sectional area, you have your
self shown that on placing a hollow and a solid cylinder of iron successively within the same electro -magnetic coil, the hollow piece exerts the greatest influence on the needle. I wished to ascertain whether a hollow magnet could not be repre sented by a solid one having the same sectional area and girth, but twice the thickness, as in figs. 5 and 6. Two

Fig. 5 .

Fig. 6 .

Half size.
electro -magnets were constructed, 17 inches long, and of
sections similar to those in the figures, and each wound with 22 feet of covered copper wire to inch in diameter .

1 !

--

CONSTRUCTED OF IRON WIRE .

Their respective attractions, at half an inch distance, for

the end of a straight steel magnet were found to be as follow

in two trials with different strengths of current :

Hollow magnet.

Solid magnet.

1.9 grs.

1 : 7 grs.

4.5

4.0

It is evident from this that the hollow magnet has the
greatest attractive force . But the difference between the
two is, I think, hardly sufficient to counterbalance practical advantages which belong to the solid electro -magnet if used
in the engine. Next I made five straight electro -magnets of square iron
wire 1 inch thick. Each was 7 inches long, and wound with 22 feet of covered copper wire tinch in diameter. No. 1 consisted of nine, No. 2 of sixteen, No. 3 of twenty -five, No. 4 of thirty -six, and No. 5 of forty -nine of the square
iron wires. Each was built up into the form of a prism with square base and section. Five other electro -magnets were
made of solid iron, but otherwise were exactly similar to the
first set. The following attracting -powers (in grains at half
an inch distance) for a straight steel magnet were obtained,
using three different galvanic forces :

No. 1. No No. 3. No. 4. No. 5 .

( 1:5 1st experiment Solid magnet .

1.9 1.6 2.1 20

Wire magnet .. 2 : 1 2: 1 1•7 2.0 1.9

2nd experiment | Solid magnet .. 2 : 0 2.5 2:35 2:45 2.2
Wire magnet .. 2:6 2:8 2:1 2.2 2-05

3rd experiment |Solid magnet .. 2:7 3.6 3-4 3.2 3.1
Wire magnet : . 3:3 3.8 3-0 2.9 2-65

The iron wire was taken at the same degree of temper
as that in which it came from the makers, consequently must
have been harder than the bar iron with which it was com
pared. It will be remarked that while the wire magnets are
more powerful in the first Nos., they are less powerful in the

10

INVESTIGATIONS IN MAGNETISM

last Nos. than the solid magnets. I cannot account for this

circumstance, unless by supposing, according to the hypo

thesis of Dr. Page, that the wires of which the magnets are

composed repel one another's magnetism in such a manner

as to tend to neutralize the general force of the electro magnet, and that this effect increases with the number of

wires used .

Yours truly,

Salford, March 27, 1839 .

J. P. JOULE .

Investigations in Magnetism and Electro-magnetism . (In two letters to the Editor of the “ Annals of Electricity ' * .)
[ 'Annals of Electricity ,' vol. iv. p. 131.] DEAR SIR,
I am now able to send you an account of my further investigations on electro -magnetic attraction . It was a matter of importance to use in the research a galvanometer
the indications of which could be depended upon . Fig. 7 represents the form of my galvanometer. The
lozenge- shaped needle n is 2 inches long, the wire 10 feet
Fig . 7. Scale

a
nt

long and to of an inch in diameter. It is disposed in four rectangles, mercury - cups being placed at a , b , c, d, e. The
• The experiments were made at Broom Hill, Pendlebury, near
Manchester.

1

AND ELECTRO -MAGNETISM .

11

wire crosses at x x , but everywhere else is in the same
plane.
The process of graduation was performed as follows :-A current of a certain intensity was passed from a to b, from a to c, from a to d, and from a to e , taking care to decrease the resistance of the battery -wires as the length of the part of
the galvanometer -wire through which the current passed was
increased : the several deviations of the needle thus obtained
were marked 1, 2, 3, and 4 on the card of the instrument. I then increased the power of the battery until the needle stood at 2 when the current passed from a to b : the former process
was then repeated, and I marked on the card 4 , 6 , and 8 ; and
going on in this manner I obtained the graduations 1, 2 , 3, 4 , 6 , 8 , 9, 12, 16, & c. When the galvanometer is used the current is passed from a to b, and the above numbers repre sent proportional quantities of electrical current.
In order to give adefinite ideaofthequantityof electricity
indicated by this galvanometer, I took a diluted acid, con sisting of 10 parts of water to 1 of sulphuric acid, sp. gr. 1.8, and passed through it by platinum electrodes a current = 1 of the graduation. In seven minutes 0:62 cubic inch of the mixed gases was evolved.
The electro -magnets used in Table I. were those described in my last letter; they are straight and square, 7 inches long, and wound with 22 feet of covered copper wire to inch in dia meter. Five of them were of bar iron ; and five corresponding ones were bundles of square iron wires. The sides of the square sections of Nos. 1 are it of an inch, which dimension is gradually increased by it up to Nos. 5, which are 11 of
an inch .
The bar-iron electro -magnets were successively suspended from the beam of a balance ; and the corresponding wire electro -magnets were brought underneath, so that of an inch intervened , a distance maintained by the interposition
of a piece of wood. Currents of the strengths given in the
table were passed through the wires of the electro -magnets and the coil of the galvanometer. The attraction was
measured in grains.

12

INVESTIGATIONS IN MAGNETISM

TABLE I.

Electrical current.
Nos. 1 .

Electro -magnets.
Nos . 2. Nos . 3 . Nos. 4.

Nog. 5.

6

76

65

88

62

42

8

133

100

180

103

98

12

258

296

300

286

206

16

500

548

530

550

410

24

1080 1280 1190 1210 1050

To vary the experiments, and with the view of ascertaining the effect of an increase of length, I constructed ten more electro -magnets of the same sectional areas, but of 14 inches (or double the former) length, and wrapped with 22 yards (or three times the length) of covered wire. Nos. 1 and Nos. 2 were made of square iron wire, the others of bar iron .

Electrical
current. 8
12
16 24

TABLE II.

Electro -magnets.

Nos. 1 .
410
cor. 675
690
cor. 1080
1000 cor. 1460
1460
cor. 2080

Nos. 2.
667
cor. 990
1170 cor. 1740
1920
cor. 2710
3500
cor. 4750

Nos . 3 . No8. 4 . Nos. 5. 1150 1205 1175
2150 3025 2625 4575 5687 4675 9625 11812 10500

Each of the electro -magnets used above, except Nos. 1 and Nos. 2 of the second table, was wound in two layers by the wire, the larger ones being uncovered at intervals. I must mention, however, that Nos. 1 and Nos. 2 of the second table
had to be wound in some parts to three layers, on account of

AND ELECTRO -MAGNETISM .

13

their small diameter. On calculation I give the probable corrections stated in Table II.
It does not appear from the above results that any con siderable detriment arises from the increase of the length of the electro -magnets. It is plain that, as the magnets in Table II. are wound with three times the length of wire, 24 degrees of electricity in the first table should have the same
effect as 8 in the second table. The difference is to be referred
to the increased length of theiron ; but I do not feel justified in assigning its value, which, however, seems to decrease as the section of the magnets increases.
The experiments, however, appear to indicate an important law , which may be expressed as follows -: The attractive force of two electro -magnets for one another is directly pro portional to the square of the electric force to which the iron is exposed ; or, if E denote the electric current, W the length of wire, and M the magnetic attraction , M = E ? W ?.
The discrepancies from this law may , I think, be owing to magnetic inertia and experimental errors. Perhaps the fairest way to compare theory with experiment, is to take the means
of all the results of the first table and the means of Nos. 3, 4 ,
and 5 in the second, omitting Nos. 1 and 2 because it is clear that these were at last becoming saturated with magnetism .
The result is as follows :

From the first Table .

From the second Table.

Ourrents .

Observed Calculatod.
attraction .

Observed
attraction .

Calculated .

6.
8 12
16 .. .
24

66.4 123
269 508 1163

66 ° 4
118 265 472 1063

1177 2600 4079 10646

1177 2648 4708 10593

Desirous to ascertain whether the law held in the case of lifting as well as in that of distant attraction , I made trial

14

INVESTIGATIONS IN MAGNETISM

with a horse -shoe electro -magnet constructed of a cylinder of iron 7 inches long, of an inch in diameter, and wound with 5 yards of covered copper wire. The law seems in this case to fail, principally because the iron is sooner saturated with magnetism . Hence the propriety of giving considerable bulk , rather than length, to electro -magnets designed for lifting
purposes.

Ourrent.
4 6 8
12

Lifting- power in lb. Calculated .

3.5

3.5

6.5

8

11.5

14

21

31.5

I can hardly doubt that electro -magnetism will ultimately

be substituted for steam to propel machinery. If the power

of the engine is in proportion to the attractive force of its

magnets, and if this attraction is as the square of the electric

force, the economy will be in the direct ratio of the quantity

of electricity, and the cost of working the engine may be

reduced ad infinitum . It is, however, yet to be determined how far the effects of magnetic electricity may disappoint

these expectations.

I find that the plan which I had proposed for a new engine

must yield to the views elicited by the above experiments.
As far as I see at present, I think it will be best to use only

two, and these very large electro -magnets, and to concentrate upon them all the strength of electrical current I can
command .

Yours truly,

Broom Hill, near Manchester, May 28, 1839.

J. P. Joule .

-

AND ELECTRO -MAGNETISM .

15

Investigations in Magnetism and Electro-Magnetism .
Letter 2nd.
[ Annals of Electricity,' vol. iv. p. 185.]
DEAR SIR, The following experiments were made for the purpose
of giving a severe test to the law enunciated in my last
letter.
I constructed two pairs of straight electro -magnets. Each of the first pair was made of a bar 30 inches long and 1 inch square ; each of the second pair was 30 inches long, 2 inches broad, and 1 inch thick . The sharp edges having been ground to avoid cutting the wire, each bar was carefully
insulated and then wound with 88 yards of covered copper
wire It of an inch in diameter.
The attractions were measured in the same manner as
before, excepting that a plate of copper was employed instead of wood in order to keep the electro -magnets asunder at the proper distance. The attraction of the suspended electro magnet for the fixed one was measured in ounces avoir dupois.

Ourrents ....

6.

8.

12. 16. 24. | 32.

at finch { Experiment 18 18

at { inch { Experiment . 7 7 .

at finch { Calculated.

3
3

33 72 32 72
13 28 12:44 28
5 • 25 12
6:33 12

124 260
128 288
47 96 49.7112
18 38 62 21: 3 48 85.3

at finch { Calculated.
at finch Experiment
{
Experiment
Calculated .

14 27 60 14 25 56

6.25 12 25 6.25 11.1 25

2.5 5

9.5

2.5 4.44 10

100 240 100 224
40 96 44 : 4 100
17.5 36 17 : 7 40

.pFaiirrst

.pSaeicrond

The experimental results are quite as near to the calculated values as could be expected. Those belonging to the first

16

DESCRIPTION OF ANOTHER

pair are particularly satisfactory , especially if, with regard to the numbers under currents 16, 24 , and 32, some allowance is made for the approaching saturation of the iron.
The above electro -magnets, 30 inches long, with 88 yards of wire and 6 degrees of current, sustained at $ of an inch distance 7000 grains ; but the mean attractive power of Nos. 3, 4 , and 5 in my last was, with the same electric force, viz. 24 degrees of current traversing 22 yards of wire, 10646 grains. From these figures we may form some idea of the diminution of power arising from the increased length
of the bars.
Yours truly ,
J. P. JOULE .
Broom Hill, July 10, 1839.

Description of an Electro -magnetic Engine. (In a letter to the Editor of the · Annals of Electricity.')
[ ' Annals of Electricity ,'vol. iv. p. 203.)
DEAR SIR,
I beg to forward to you the description of my new
electro -magnetic engine*. It is represented in perspective
by fig. 8, where A B C D is a strong wooden frame, the inside measure of which is 4 } feet long, l } foot high, and I foot broad. F G is the axle of cast steel, three quarters of an inch square, and turned down to half an inch at the journals, which work in the brass bearings 66. HH are holders for the
revolving electro -magnets in m : there is a third holder be
tween these, which is omitted in the figure, but which con
tributes very materially to the stability of the apparatus.
nn are the stationary electro -magnets, firmly secured to the top and bottom of the frame by the brass clamps kk. 38 are
silver springs which enter through the frame and press gently
on the brass semicircles of the commutator c. This com mutator is shown on a larger scale in fig. 9. vv are copper
• Constructed at Broom Hill,

--
1

ELECTRO -MAGNETIC ENGINE.

17

wires soldered to the brass semicircles, and connected with the revolving electro-magnets. xx are wires let into the frame, and connected with the stationary electro -magnets. z z are wires to form the necessary connections at the other
end .
Fig. 8. Scale to

A D
n E
m

H

m

H

B

k

n

Ic

Fig. 9. Scale

Fig. 10. Scale is . 16

口

6
Fig. 11. Full size.
B
In fig. 10 one of the holders is represented on a larger scale. It is constructed of hard wood strengthened with
с

18 DESCRIPTION OF AN ELECTRO -MAGNETIC ENGINE .
brass plates. 66 are brass screws, which act upon wedges, so that I can adjust the electro-magnets at different distances
from the axle . a is a screw which secures the holder to the axle .
The iron of each revolving electro -magnet is 36 inches long, three inches broad , and half an inch thick : that of the stationary electro -magnets is of the same breadth and thickness ; but their length is such that when bent their extremities are 364 inches asunder ; the revolving magnets therefore pass at the distance of ß of an inch.
One of the revolving electro-magnets is made of a solid bar ; the other of rectangular iron wire built up as shown by fig. 11, the two planed iron bars a b serving to hold them secure . The iron of these is in the usual condition of hard ness as supplied by the trade. I intend to replace them ultimately by electro -magnets constructed of thoroughly
annealed iron .
Each electro -magnet is furnished with a strip of thick calico next the iron, and with three separate layers of covered copper wire, each 106 yards long and 44 of an inch in diameter, wound with the utmost regularity.
The engine consists of only four electro -magnets ; but it will at once be seen that the principle admits the use of any greater even number. I do not, however, think that the mere increase of number would be attended by advantage.
As the form of the engine enables me with ease to place the electro -magnets in different positions, and as their several coils are insulated, and I am therefore enabled to use the electric current in quantity and intensity arrangements, it offers facilities for experiment. In my preliminary trials I have been much pleased with its performances.
Yours truly ,
J. P. JOULE.
Broom Hill, near Manchester, 30th August, 1839 .

--
1

ON ELECTRO -MAGNETIC FORCES.

19

On Electro-magnetic Forces. By J. P. JOULE * .
[ Annals of Electricity,' vol. iv. p. 474.)
ABOUT the beginning of last April I made some experi ments in electro -magnetismt. I now return to the subject, desirous of placing some of the effects I then witnessed in a more complete and accurate point of view .
I have shown that when a current of voltaic electricity is transmitted through the coils of' two electro -magnets, their mutual attraction is in the ratio of the squares of the quan tities of electric force ; and also that the lifting - power of the “ horse -shoe” electro -magnet is governed by the same law .
I have recently made experiments which prove that the attraction of the electro -magnet, for a magnet of constant force, varies in the simple direct ratio of the quantity of electricity passing through the coil of the electro -magnet. In order to succeed in the experiment it is necessary to make the effects of induction insensible by a proper arrange ment of the apparatus.
Magnetism appears therefore to be excited in soft iron in the direct ratio of the magnetizing electric force ; and electro ma attraction , as well as the attraction of steel magnets, may be considered as proportionate to the product of the intensities of each magnet, or , which is the same thing, to the number of lines which may be drawn between the several magnetic particles of the attracting bodies.
Fig. 12. Fig. 13. Fig. 14.
ITT
This view is illustrated by figs. 12, 13, and 14 , where the
* The experiments were made at Broom Hill, near Manchester. t'Annals of Electricity,'iv. pp. 131, 135.
c2

20

ON ELECTRO -MAGNETIC FORCES .

several attractions of the magnetic particles, viz . 1 to 1, 2 to 1, and 2 to 2, are represented by the number of lines drawn in each instance, i.e. 1, 2 , and 4. Fig. 15 illustrates

Fig. 16.

A
a 6c

B
the complex action of forces constituting the aggregate attraction existing between two magnets, A and B. The magnetic particles, of which six only, viz. a , b, c, d , e, f, are represented, may be conceived to be an indefinitely large number spread throughout the region of the poles.
If this view be correct, it is obvious that the closer the approximation of the magnetic particles in each system , the greater will be the magnetic attraction ; for in that case, for
instance, the particle a will both be nearer the particle f, and the force exerted between them will be in a less oblique
direction .
It was in consequence of my entertaining a different view ,
that I was led to imagine that I had determined the amount of decrease of power arising from the increase of the length of the electro -magnet ; for in a comparison of the attractive powers of three pairs of electro -magnets, of the several sizes B, ft, and Yr inch square, with those of two pairs whose sec tional areas were respectively one inch square and two inches by one inch, I found the latter long electro -magnets weaker than
the former in the ratio of 7000 to 10646an effect which now
appears to me to be principally owing to diffused polarity.
The loss of attractive power in consequence of length

ON ELECTRO -MAGNETIC FORCES.

21

might be readily ascertained ; but neither this nor the diffu sion of polarity affects the main conclusion at which I have arrived with regard to the laws under which magnetic attrac tion , as applicable to the production of motive force, is de veloped by electricity, viz. that the attraction of two electro magnets towards each other is in every case represented by the formula M = WºE ?, where M denotes the magnetic attraction, W the length of wire, and E the quantity of electricity con veyed by that wire in a given period of time,-a formula modi fied merely by the effects of saturation, inductive power of the iron, and distance of the coils from the surface of the
iron .
I have observed that magnetic and electro-magnetic attrac tion decreases, in certain cases, in the simple ratio of the distances. This was found particularly when the magnets were long and the distances between them small. Mr. Harris has observed a similar effect, which may be accounted for on the principles previously illustrated . It is impossible to
doubt the accuracy of Coulomb's law that magnetic attrac tion obeys the law of the inverse square of the distance.
I will now describe some experiments I have made with the electro-magnetic engine described in page 17. For these I prepared a galvanometer constructed on the plan described in page 10. The coil is rectangular, 12 inches by 6 , and of copper wire of an inch in diameter ; the length of the needle near 4 inches. A table of the absolute value of its
indications was formed in the manner already described. In the following experiments the engine was fitted up with
the unannealed bar and iron -wire electro -magnets. After a few trials it was found undesirable to use the wire united so as to form one length. I therefore soldered the ends of the
three coils of each electro -magnet together, and united the combined wires in such a manner that the electric current passed through 424 yards of threefold conducting wire.
In the Tables the first column indicates relative quantities
of electrical current ; the second, the difference between
those quantities ; the third, the velocity of the revolving

22

ON ELECTRO -MAGNETIC FORCES .

electro-magnets in feet per second ; the fourth, the work in cluding friction ; the fifth, the duty in pounds raised to the height of one foot by the agency of one pound of zinc.
In calculating the amount of work, I found that current 12:4 was just sufficient to keep the machine in motion , the
friction referred to the distance from the axle of the revoly
ing electro -magnets being equal to ten ounces avoirdupois : the same quantity of current was, whatever the velocity might be, always able to overcome exactly the same amount
of friction . I therefore felt justified in using it as a basis on
which to calculate the force due to other quantities of cur rent electricity. The duty, in the fifth column, is calculated
on the basis of the decomposition of water effected by a given
current. I must observe that the friction is estimated as part of the work ; and that, whenever the motive force was
not sufficient by itself to turn the machine, a weight thrown
over a pulley on the axis supplied the requisite assistance.

TABLE I.
80 pairs of Wollaston's 4 - inch plates of amalgamated zinc and double copper.

Current, Difference. Velocity. Work .

24.6 21.6
19-6
180 16.5 150

3
2
••••
1.8
145
1.5

0

0

2

3.8

4

6:25

6

7.89

8.85

9.15

Duty .
0
21960 39740 54800 66950 76140

8 10

--

ON ELECTRO -MAGNETIC FORCES.

23

TABLE II.
40 pairs of Wollaston's plates.

Current. Difference. Velocity. Work .

Duty.

11 8
1.6
10.2
0-8
9.4 0.8
8.6
0-6 80

0

0

0

2

0.85

20700

4

1:44

38300

6

1.8

52320

8

2.08

65000

TABLE IIT.
10 pairs of Wollaston's plates.

Current. Difference. Velocity. Work.

Duty.

5
0-8
4.2 0-6
3.6 0-3
3:3 0:3
3.0

0

0

0

2

0.14

33300

4

0.21

58300

6

0 • 265 80300

8

0.292

97300

TABLE IV .
Grove's Battery ; 10 4 -inch plates.

Ourrent. Difference . Velocity. Work.

Duty.

17.6 3:3
14:3 1.9
12.4
14
11.0
0.8
10.2

0

0

0

2

1.66

116080

4

2.5

201600

6

2.95

268200

8

3:38

331400

24

ON ELECTRO -MAGNETIC FORCES .

I now united the conductors in such a manner that the
fluid was divided between the stationary and revolving electro magnets ; in this case the electricity passed through 212 yards
of sixfold wire .

TABLE V.
80 pairs of Wollaston's plates in series of 40.

Current. Difference. Velocity. Work.

52
9
43
5 38
3.2
34 : 8 2.6
32.2
2:4
29.8

0

0

2

3.76

4

5.87

6

7:38

8

8.42

10

9:02

Duty.
0
21800
38600 53100 65400 75700

TABLE VI .
40 pairs of Wollaston's plates in series of 20.

Current. Difference. Velocity. Work.

Duty.

28.2
5
23 : 2 2.5
20 • 7
1:7
19.0 1.4
17.6

0

0

0

.2

1.1

23700

4

1.74

42000

6

2.205

58000

8

2.52

71600

--

1

ON ELECTRO -MAGNETIC FORCES .

25

TABLE VII.
20 pairs of Wollaston's plates in series of 10.

Ourrent. Difference. Velocity. Work .

Duty .

16.8 3
13.8 1.6
12 :2 1.2
11.0 1-0
10 : 0

0

0

0

2

0 :387

28000

4

0-605

49600

6

0.738

67100

8

0.813

81300

The above examples will show pretty clearly the effects of magnetic electrical resistance. This resistance is the prime obstacle to the perfection of the electro -magnetic engine ; and in proportion as it is overcome will the motive force increase.
It therefore claims our first attention .
On comparing the differences with the velocities and cur rents in each table, the general conclusion is that the mag netic electrical intensity resisting the current is directly pro
portional to the velocity multiplied by the magnetism , or, which is the same thing, by the electricity which induces that
magnetism . It is the latter part of this law which makes the differences decrease in the same ratio with the electrical currents opposite. In forming these conclusions I neglect the first difference, or that which exists between no motion
and velocity 2, because this is much augmented by a very trifling inaccuracy in the construction of the commutator.
It appears, moreover, that this law is unaffected by altering
the battery -intensity ; for on comparing the tables of either system together, it will be seen that the differences are always about one tenth of the currents to which they are opposite.
In the second arrangement the conducting metal was half as long and twice as substantial as in the first : hence it is that half the battery -intensity sufficed to pass twice the quantity of current, and so to produce the same motive power. Com

26

ON ELECTRO -MAGNETIC FORCES.

pare Table I. with Table V. Also a reference to these tables
shows that the differences are twice as great in the second arrangement as in the first, the magnetic force remaining very nearly the same. To understand the reason of this we
must observe that — 1st, the magnetic electrical intensity has nothing to do with the thickness of the wire upon which it is induced, but exists simply in the direct ratio of the length ;
consequently that the intensity was only one half as great in
the second arrangement as it was in the first in like circum
stances ; and 2nd, that, as the resistance of the wire to the current in the second arrangement is only one quarter of that in the first, an additional or extraneous resistance will pro
duce four times the effect in the former as in the latter in
stance. Hence, by compounding the two effects, we have the differences of current due to a given increment of velocity with the same amount of magnetism twice as great in the second as in the first arrangement.

If the intensity of the voltaic battery increases in the ratio of the number of its elements, there will be no variation in economy, whatever the arrangement of the conducting metal may be, or whatever the extent of the battery. For if the battery be doubled in intensity, it must consist of twice the number of pairs in series, which will cause twice the quantity of electricity to pass ; and so four times the weight of battery materials will be consumed in the same time ; whilst the force of the engine is also increased fourfold , according to the square of the current. See the duty under Tables 1, 2, 5,
and 6.
I think my engine might be improved by iạcreasing the conductive power of its coils and the softness of the iron of the electro -magnets ; but the augmentation of the intensity of each element of the battery is very important, as it is attended by a proportional increase of duty.

Broom Hill, near Manchester,
March 10, 1840.

ON ELECTRO -MAGNETIO FORCES .

27

On Electro-magnetic Forces. By J. P. JOULE *.
[ Annals of Electricity,' vol. v. p. 187.]
In resuming the relation of my researches, I shall dismiss for the present the investigation of electro -magnetic forces acting at a distance, and consider the laws which govern that peculiar condition which is assumed on the completion of the
ferruginous circuit — the lifting or sustaining power of the
electro -magnet. Although this wonderful property is known to all, and a
variety of forms have been given to the electro -magnet, both as regards the bulk and shape of its iron, and the length and number of its magnetizing spirals, I am not aware that any general rules have been laid down for its manufacture, which is the more to be regretted, as some have been led to imagine that the different capabilities of various arrangements are the consequence of causes too many and recondite to be un ravelled . I shall attempt in this paper to throw some light on the subject, and to describe a construction giving greater results than have hitherto been attained . It was my desire to make my experiments as exact as possible; and as I wish the relation of them to be clear and definite, I will begin with some observations on the measure of current electricity indi cated by the galvanometer, an instrument not only useful but essential in an inquiry like the following.
The great difficulty, if not the impossibility, of understand ing experiments and comparing them with one another, arises in general from incomplete descriptions of apparatus and from the arbitrary and vague numbers which are used to charac terize electric currents. Such a practice might be tolerated in the infancy of the science ; but in its present state of ad vancement greater precision and propriety are imperatively demanded . I have therefore determined for my own part to
abandon my old quantity numbers, and to express my results
on the basis of a unit which shall be at once scientific and convenient.
* The experiments were made at Broom Hill.

28

ON ELECTRO -MAGNETIC FORCES .

That proposed by Dr. Faraday is, I believe, the only standard of this kind that has been suggested. His dis covery of the definite quantity of electricity associated with the atoms or chemical equivalents of bodies, has induced him to use the voltameter as a measurer, and to propose that the hundredth part of a cubic inch of the mixed gases forming water should constitute a degree *. There can be no doubt that this system would offer great advantages to the experi
menter in some cases, and when the above instrument is employed. However, as I am not aware that it has been
used in the researches of any electrician, not excepting those of Faraday himself, I do not hesitate to advance what I think
more appropriate as well as generally advantageous. It is thus simply stated :
1. A degree of static electricity is that quantity which is just able to decompose nine grains of water . 2. A degree of current electricity is the same amount propagated during each hour of time. 3. Where both time and length of conductor are elements, as in electro -dynamics, a degree ofelectric force, or of electro -momentum , is indicated by that same quantity (a degree of static electricity) propagated through the length of one foot in one hour of time. Whenever in this paper I speak of degrees, I intend those I have just defined .
As 9 is the atomic weight of water, it is obvious how
greatly this degree would facilitate the calculation of electro
chemical decompositions. I may adduce an illustration from electrotype engraving : here, if a galvanometer graduated according to the proposed scale were included in the circuit,
it would only be necessary to multiply the degrees of its in dication by 32, the equivalent of copper, and the product by the time in hours during which the work has been carried on, to obtain the weight in grains of the copper which has been precipitated ; and there would therefore be no occasion to arrest the process until calculation has shown that the proper quantity of copper was cast.
The galvanometer I described in my last paper was con

Experimental Researches, series 7 (736 ).

ON ELECTRO -MAGNETIC FORCES.

29

nected with an apparatus furnished with fine platinum elec trodes. A voltaic current was transmitted through the in struments ; and after a few minutes the circuit was broken and
the hydrogen measured in a graduated tube. The mean of ten trials showed that 0-76 gr. of water was decomposed per hour by the current indicated by each unit of my former quantity numbers : hence 11.8 of these last are equal to one degree of my present scale. In proof of the accuracy of the indications of the galvanometer thus graduated, I trans mitted a current of 0 °.415 through sulphate of copper by copper electrodes for the interval of an hour and a quarter. The copper deposited amounted to 15.6 grains, while the
theoretical result is 0°415 x 32 x 1.25 = 16.6 grs ., the slight deficiency being probably owing to the consumption of a part of the current in the decomposition of water, as a few bubbles of hydrogen ascended from the negative electrode.
The dimensions of the single coil of my galvanometer are 12 by 6 inches ; and the deviation of its needle for one degree of current is 34 ° of the graduated card. From these data it is easy to calculate the value of the indications of any similar instrument, bearing in mind that the electro -dynamic force on the needle produced by a constant current is directly as the number of coils and inversely as their linear dimensions.
The quantities of current electricity used in the subsequent experiments were frequently sufficient to bring the needle of
the above galvanometer to an almost rectangular position to the plane of the coil. I have, for use in these cases, devised
the instrument represented by fig. 16 , where CC is a rod of

Fig. 16. Scales.

C

m

m

C
r
copper , bent and fastened firmly to a wooden frame; mm , a magnetized bar of steel, supported, like an ordinary balance

30

ON ELECTRO -MAGNETIC FORCES.

beam , by knife- edges resting on concave surfaces of hard steel; s a scale, hung from the nearer end of the magnet for the purpose of receiving the weights by which the strength of the current is measured ; and re is a rest, the under sur face of which is just touched by the magnet when at zero.
In using this galvanometer it is merely necessary to adjust the magnet to zero, either by means of screws, weights, or by the attraction or repulsion of a small magnet kept for the purpose. Then, on making the necessary battery communi cations at CC, the scale s will rise with a force estimated by the weight in grains, tenths, & c. which is required to bring the beam again to zero . In my instrument one degree of current is indicated by 0.69 of a grain in the scale.
The value of the new galvanometer (the sensibility of which may be increased at pleasure by multiplying the number of coils), besides its usefulness in measuring copious currents, consists in its perfect independence of the terrestrial as well as of any other ordinary magnetic influence . In every situation, provided that the intensity of the balance -bar remains con stant and no interference is induced after its adjustment to zero , the transmitted current is exactly proportional to the weight lifted by the scale ; so that I should have as much confidence in working with it on an iron steamboat as if every particle of neighbouring iron were entirely removed .
I proceed now to describe my electro -magnets, which I constructed of very different sizes in order to develop any curious circumstance which might present itself. A piece of cylindrical wrought iron, 8 inches long, had a hole, one inch in diameter, bored the whole length of its axis ; one side was then planed until the hole was exposed sufficiently to separate the thus-formed poles one third of an inch. Another
piece of iron, also 8 inches long, was then planed, and being secured with its face in contact with the other planed surface, the whole was turned into a cylinder 8 inches long, 34 inches in exterior, and 1 inch in interior diameter. The larger piece was then covered with calico and wound with
four copper wires covered with silk, each 23 feet long and I of an inch in diameter — a quantity which was just suffi

ON ELECTRO -MAGNETIC FORCES .

31

cient to hide the exterior surface and to fill the interior
opened hole. The electro -magnet will perhaps be better un derstood on reference to fig . 17 , where m is the “ horse-shoe," on which I have drawn some lines to show the position of the conducting wire, a is the armature, and 88 are hooks and eye-holes for the purpose of suspension . The above is designated No. 1 ; and the rest are numbered in the order of their description.

Fig. 17. Scale tt .
Il
w

3

m

f

b
H6
W

I made No. 2 of a bar of half- inch round iron 2.7 inches
long. It was bent into an almost semicircular shape, and then covered with 7 feet of insulated copper wire zu of an inch thick. The poles are half an inch asunder ; and the wire completely fills the space between them .
A third electro -magnet was made of a piece of iron 0.7 inch long, 0:37 inch broad , and 0.15 inch thick . Its edges were reduced to such an extent that the transverse section was elliptical. It was bent into a semicircular shape, and wound with 19 inches of silked copper wire o of an inch in di
ameter.
To procure a still more extensive variety, I coustructed what might, from its extreme minuteness, be termed an elementary electro -magnet. It is the smallest, I believe, ever

32

ON ELECTRO -MAGNETIC FORCES .

made, consisting of a bit of iron wire of an inch long and zł of an inch in diameter. It was bent into the form of a semicircle, and was wound with three turns of uninsulated copper wire of an inch in thickness.
A lever, fig. 17, was employed for measuring the strongest lifting powers of No. 1. 6,6, 6, 6 are beams of ash , 3 inches
square and 10 feet long, and strengthened by iron plates ; they are made into two pairs by boards, not shown in the figure, screwed to their upper surfaces ; i, i are movable iron bearings, and f is the fulcrum , also movable, and armed with iron ; w, w are strong cylinders of wood, which bear upon the levers and carry the hooks which are affixed to the electro
magnet and its armature. In Table I. the first column gives degrees of current elec
tricity as already defined . The second gives the products of the length in feet of wire wound on the magnet into the numbers of the currents ; it therefore contains degrees of electric force. The last gives the weight lifted in pounds avoirdupois.

TABLE I.
Electro-magnet No. 1. Weight of its iron 15 lb.
Length of wire 23 feet.

Electric Ourrent.

Electric Force.

Lifting
Power .

.Estimated .Estimated

0-28
0-64 0.91 1:34
2.85
3.83
4:3 5.7 8.6
14:4
21.6
36.0

8:57
14:6
21.0 30.8 65.5 880 99.3 132.5 1987 331.0 497.0 828.0

2.75
10
23
45
238 540 670 890 1060 1400 1800 2030

!

ON ELECTRO -MAGNETIC FORCES .

33

On one occasion & weight of 2090 lb. was required to detach the armature, which is, I believe, the greatest weight any magnet has hitherto carried, and is certainly vastly superior to the performance of any other of its size.
The first two columns in the upper half of the table were reduced from the actual indications, as I had detected a
certain want of insulation of the coil.

TABLE II .

Electro -magnet No. 2. Weight of its iron 1057 grains. Length of wire 7 feet.

Electrio current.

Electric force .

Weight liftod .

0.51

8:57

20

1.53

10-7

38.5

6.1

42 : 7

49

TABLE III.

Electro -magnet No. 3. Weight of its iron 65.3 grains.
Length of wire 1.58 foot.

Electric current.

Electric force.

Weight liftod.

0.42

8.66

5:5

10

1.58

9

2.0

3:16

11

With great care this small electro -magnet supported in one instance twelve pounds, or 1286 times its own weight.
No. 4 , the weight of which was only half a grain , carried in one instance 1417 grains, or 2834 times its own weight.
It required much patience to work with an arrangement 80 minute as this last ; and it is probable that I might ultimately have obtained a larger figure than the above, - which, however, exhibits a power proportioned to its weight far greater than any on record, and is more than eleven times that of the celebrated steel magnet which belonged
to Sir Isaac Newton .
It is well known that a steel magnet ought to have a much greater length than breadth or thickness ; and Mr. Scoresby has found that when a large number of straight steel magnets
D

84

ON ELECTRO -MAGNETIC FORCES .

are bundled together, the power of each when separated and

examined is greatly deteriorated * . All this is easily under

stood, and finds its cause in the attempt of each part of the

system to induce upon the other part a contrary magnetism to its own . Still there is no reason why the principle should

in all cases be extended from the steel to the electro -magnet,

since in the latter case a great and commanding inductive power is brought into play to sustain what the former has to

support by its own unassisted retentive property. All the

preceding experiments support this position ; and the following table gives proof of the obvious and necessary general con

sequence, that the maximum power of the electro-magnet is directly proportional to its least transverse sectional area .
The second column of Table IV . contains the least sectional

area in square inches of the entire magnetic circuit. The

maximum power in pounds avoirdupois is recorded in the third ; and this, reduced to one inch square of sectional area,

is given in the fourth column under the title of specific

power.

TABLE IV .

Least Maximum Specific
sectional area . power. power.

( No. 1 .. 10

2000

209

No.2 ..
My own electro -magnets .
No. 3 .. No. 4 ..

0.196
0 ·0436
0.0012

49

250

275

0.202 162

Mr. J.C. Nesbit's. Length round

the curve 3 feet. Diameter of

iron core 24 inches. Sectional

4.5

1428

317

area 5•7 inches ; do. of arma

ture 4:5 inches. Weight of iron
about 50 lb.

Prof. Henry's. Length round the

curve 20 inches. Section 2 inches

3.94

750

190

square. Sharp edges rounded off.

Weight 21 lb...

Mr. Sturgeon's original. Length

round the curve about 1 foot.

0.196

50

255

Diameter of the round bar half
half an inch

Scoresby's Magnetical Investigations,' p. 37.

ON ELECTRO -MAGNETIC FORCES.

85

The above examples are, I think , sufficient to prove the rule I have advanced. No. I was probably not fully saturated ; otherwise I have no doubt that its power per square inch would have approached 300 . Also the specific power of No. 4 is small, because of the difficulty of making a good experiment with it. The electric force used on Mr. Nesbit's
electro-magnet was exceedingly great, consisting of 19 of Daniell's two-feet cells with coils of 14 lengths of copper
wire, each 70 feet long and inch thick. On the other hand, Professor Henry used only a pair of voltaic plates,
with a much inferior conductor.
The mean of the specific powers of No. 2, No. 3 , and Mr. Nesbit's may, I think, be fairly taken for the expression of the maximum magnetic force of iron under ordinary circumstances, which therefore is given by the formula x = 280 a , where a is the sectional area of the magnetic circuit in square inches, and is the force of adhesion of both poles.
Since the element of length has no place in the above formula , and has in fact only a secondary influence, playing the part of a resistance which it requires a large additional electric force to overcome, it is obvious that in proportion to its reduction the attractions relatively to the weight of iron will increase : hence the large power relative to weight of my short electro -magnets.
The above corroborates what I have already said with regard to the proper construction of the electro -magnet for lifting weights, the condition being well illustrated by
fig. 15. When A and B come into contact the oblique forces disappear, and the attraction is in the simple ratio of the number of saturated magnetic particles opposed to each
other.
With respect to the magnetizing coils, I may observe that each particle of space through which a certain quantity of electricity is propagated appears to operate in moving the magnetism of the bar with a force proportionate to the inverse square of its distance from the surface of the iron, and that, when the tension of the magnetism is the same,
D2

36

ON ELECTRO -MAGNETIC FORCES .

the thickness of the iron on which that particle of conducting space acts has nothing to do with the whole effect. Now it may be shown that, such being the law , if the particle induce upon a large surface, the resulting magnetic induction will not vary very much with the distance, but be a very constant quantity for any distance which bears a very small ratio to
the dimensions of that surface. Hence it is that a coil
within a hollow piece of iron has no power to magnetize it * ;
for in that case its energy is directed in equal quantities
towards opposite directions, the nearness of one surface
counterbalancing the size of its opposite. Hence also in my electro -magnet No. 1 , where the surface is large, the coil exercises nearly the full extent of its duty even in the places where it does not lie closely to the iron .
Where the magnetization is considerably below the point of saturation, the resistance to induction arising from the
length of the magnet becomes a very sensible quantity, varying probably in the direct ratio of that element. Some
idea of its effect may be formed from the following table, in which I have compared half the maximum powers of three of my electro -magnets with the electric forces which produce them ; and by dividing the former by the latter, I have obtained a fourth column which, under the title power per
degree, contains the lifting-power for half the maximum magnetization due to a unit of electric force.

Longth of magnetic
circuit in inches.

No. 1 ...... 7.46

No. 2

3.7

No. 3

1.05

TABLE V.
Electric force.
200
4.5 0.66

Half maximum power . Ib . 1060
25 5.5

Power per degree of
eleotric forco. lb. 5:3
5.5
9.2

It is well known that, after the galvanic circuit is broken ,
the armature of an electro -magnet is still retained with very considerable force. I was desirous of trying the capability of No. 1 in this respect. The following table contains in its first column the degrees of electric force which were cut off,
• Scientific Memoirs,' part 5, p. 14 .

ON ELECTRO -MAGNETIC FORCES.

37

the second gives the lifting -powers due to those forces, and the third gives the weight required to detach the armature
after the current was broken .

Electric force cut off,
88
29 14.5

TABLE VI.

Lifting - power due Lifting -power remaining
after the electric force to electric forco cut off.
was removed .

lb.

lb.

540

33

40

16

10

10

There was considerable difficulty in making a good expe riment with so powerful an electro -magnet as No. 1 when small forces were in question. Nevertheless the above results afford good evidence that nearly all the lifting -power
due to feeble currents is retained after those currents are
cut off.
When the current is not entirely cut off, but is merely reduced in strength by the interposition of a thin wire, a surprising quantity of lifting -power can be retained by the small electric force left. Having subjected No. 1 to 90 ° of electric force, a quantity adequate to bring up its power to
560 lb., I reduced the current to a lower intensity, and then
found the weight requisite to detach the armature. The following table gives the results of several such experiments.

Electric force remaining
after 900 was cut off.
si
21
14.5
6.2
4.1

TABLE VII .
Lifting -power due
to this force . 1b. 45 23 10 2.6 1.1

Weight required to
detach armature , lb.
294 210
112 63
56

A voltaic cell, the size of a common sewing -thimble, was quite sufficient to produce 31° of electric force in No. 1, and consequently to sustain a magnetic power of about 300 lb.; and it is easy to perceive that, by increasing the size of the

38

NOTE ON VOLTAIC BATTERIES .

electro -magnet and the quantity of conducting -wire, this minute source could support a magnetic virtue of indefinite
amount.

Note on Voltaic Batteries.
Having had occasion, about a year ago , to construct a battery of great intensity, it became an object to devise an arrangement which should be convenient to use and easily refitted . After trying two or three systems, I succeeded in producing one which answered very well ; but as I felt that long experience could be the only strict test of its value, I have hitherto refrained from presenting it to public notice. Now , however, that I have worked with it during nine or ten months, and have found it to possess every quality that could be desired, I hope by describiug it to give the same facilities to others which I enjoy myself.
Fig. 18 represents three elements of my battery. A B is
Fig. 18. Scale .

a

a

a

a

S

S

-
MI

А

B

the common divided Wollaston trough with the front side

----

NOTE ON VOLTAIO BATTERIES.

39

removed in order to show the interior. The black lines
within the cells are rectangular pieces of strong sheet copper, bent on a gauge to the shape seen in the figure. Within these, 2 , 2 , z represent plates of sheet zinc, amalga mated in those parts which are in contact with the dilute sulphuric acid, and fixed in their places by pieces of hard wood furnished with grooves, and extending the whole breadth of the zinc. Lastly , a, a, a, a , a are pieces of wood
with holes in their centres to admit the screw bolt 88, which secures the whole.
When the battery is worn out, empty the trough and replace it therein ; then unscrew the bolt and remove it with the pieces of wood ; change the old zinc plates for new ones, taking care in the mean time to see that those parts of the coppers which touch the zincs are bright; then replace the pieces of wood and screw them tightly together.
Mr. Smee's battery may be fitted up on the above plan. I prefer, however, for continued use sheet iron before either copper or platinized silver . In using sheet iron it is well to
tin that part which is to touch the zinc, in order to ensure perfect contact.
I have lately constructed a large battery on Mr. Sturgeon's plan ; and from my experience with it I am convinced that it presents a very superior arrangement of voltaic elements. It consists of eleven cast-iron cells, each 1 foot square and 14 inch in interior diameter . With eight of the pairs arranged in a series of four, I can raise to a full red heat 18 inches of copper wire to inch thick.

Broom Hill, near Manchester, August 21st, 1840.

40

ON ELECTRO -MAGNETIC FORCES.

On Electro-magnetic Forces. By J. P. JOULE *.
[ Annals of Electricity,' vol. v. p. 170.]
IN last paper I have described a method of constructing the electro-magnet which gave great lifting - power. The fresh experiments will, I hope, be deemed confirmatory of the prin ciples before advanced.
A piece of “ stub ” iron was, as in the manufacture of gun barrels, formed in a spiral, and welded on a mandril into the shape of a thick tube, the process rendering the iron very compact and sound throughout. This, and another piece of iron intended for the armature were planed , turned , and fitted with eyehole screws in the manner I have already described.
In fig . 19, C represents the electro -magnet, D the armature,

Fig. 19. Scale
99

‫ܐ ܐ ܐ ܐ ܐܐܐ‬

A

B

and A B a conductor of copper rod or wire passing along one side, returning by the axis, and then away by the other side, 80 as to go about the iron once only, and in a shape somewhat
similar to that of the letter S. The length of the cylinder formed when the armature is in contact is 2 feet ; its external
diameter is 1.42 inch, and its internal 0.5 inch. The weight of the iron of the magnet along with the screws is 6 lb. 11 oz.,
that of the armature 3 lb. 7 oz. The least sectional area of
its magnetic circuit is 104 square inches : I call it No. 5, to distinguish it from those already described .
A copper rod, f of an inch in diameter, covered with tape, was bent about the electro -magnet as just described . It and its armature respectively were then secured by means of cords passing through the eyeholes to the lever (fig. 17) used in my former experiments. Eight large cast - iron cells, each of which presented an effective surface of 2 square feet, were
• The experiments were made at Salford Brewery.

--

ON ELECTRO -MAGNETIC FORCES.

41

arranged as a single pair, and gave a lifting - power of
1350 lb.
Thinking that a bundle of wire was probably a better con ductor than a rod of the same length and weight, I removed the copper rod, and substituted for it a bundle of 60 wires, each its of an inch thick. With this it was found that sixteen of the cast- iron cells arranged in a series of two produced a lifting -power of 1856 lb., or 183 times the weight of iron employed in the construction of both the electro
magnet and its armature.
By dividing the power thus obtained by the least sectional area, the weight lifted per inch is 181 lb., or about two thirds of that which a comparison with other electro-magnets would lead us to expect. I do not think that the electric force was sufficient to bring the magnetism close to the point of satu ration ; and , moreover , it was difficult to secure the necessary
condition of equal tension along the whole length of the iron . Suspecting that the greatest power of the large electro
magnet No. 1 had not been reached in my last experiments, I refitted it, using a coil consisting of 21 copper wires, each 23 feet long and 3 of an inch in diameter, bound together by cotton tape. Sixteen of the large cast - iron cells were arranged in a series of four; and when connected by sufficiently good conductors to the electro -magnet, a weight of 2775 lb. was found requisite to detach the armature .
Now , by the formula in my last paper, 280 x 101 = 2870 is the estimated value for the power of this electro -magnet on
its approach to saturation . That this was very nearly attained
is manifest from the fact that the electric force used in the
last experiment was four times as great as that which was
competent to give a lifting-power of 2128 lb.
Although the battery I have used for obtaining maximum results is powerful, a very good effect may be obtained by means of a very small voltaic arrangement. For instance, No. 1 can carry eight hundredweight when the current generated by a single pair of four-inch plates of iron and amalgamated zinc is transmitted through its coils; and with a pair of platinized silver and zinc plates exposing only two

42

DESCRIPTION OF A

square inches of surface, the attraction is such as to render it almost impossible by the hand to slide the armature .
Broom Hill, near Manchester, Nov. 23rd , 1840.

Description of a new Electro-magnet. By J.P. Joule. (In a letter to Mr. Sturgeon *. )
[ 'Annals of Electricity,' vol. vi. p. 431.]
MY DEAR SIR , In that part of my researches on Electro -magnetic
Forces which was published in the ' Annals ' for September last, I showed that the maximum lifting -power of the electro magnet is proportional to the least transverse sectional area of magnetic circuit ; and at the same time I pointed out the method whereby a very great magnetic attraction could be produced between masses of iron of inconsiderable mag
nitude .
To illustrate my views I made several electro -magnets, two of which have been a long time on exhibition at the “ Victoria Gallery."
Stimulated by my success , some gentlemen of Manchester have constructed electro -magnets of a variety of forms, em bodying the principle of a large sectional area. They have in consequence carried very heavy weights ; Mr. Roberts's, in particular, has sustained the greatest load on record. Mr. Radford's is also very powerful; and his arrangementt of a spiral groove on the face of the magnet to admit the coil involves a new and important principle of magnetic
action .
The following is a description of a new electro -magnet which I have recently constructed . Fig. 20 gives its appear ance in perspective. B, B are two rings of brass, each 1 foot in exterior diameter, 2 inches broad, and 1 inch thick ; to
• The experiments were made at the Salford Brewery. † 'Annals of Electricity ,' vol. vi. p . 231.

--

!
.2.0

48
ovo
.
-

NEW ELECTRO -MAGNET.

each of these flat pieces of iron are fixed by means of the bolt-headed screws 8, 8 & c.; twenty -four of them , m , m (figs.20
and 21) , have rectangular grooves, and are fixed to the upper

Fig. 20. Scale +

89

B

S

.

B Fig. 21. Scale
m

44

DESCRIPTION OF A

ring; twenty -four, a, a & c. ( figs. 20 and 22 ), are plain , and
are fastened to the lower ring.

Fig . 22. Scale

.
A bundle, w, w ( fig. 20 ), consisting of sixteen copper wires, each 16 feet long and zł of an inch thick, covered with a double fold of cotton tape, was bent to and fro in the grooves of mm , as seen in the figure.
Fig. 23 shows the method I adopted to give the electro magnet a firm and equable suspension : aa are hoops of wrought iron, to each of which four bars of the same metal are riveted , and welded together at the other end into a massive hook . The hoops are bound down to the brass rings by copper wires cc & c.
On connecting the coil of the above electro -magnet with a battery *, consisting of sixteen of your large cast - iron cells arranged in a series of four, I obtained a lifting -power of
• Each cell was 1 foot square and 14 inch wide. Arranged as above, the battery was able to give an electric discharge between the poles as much as $ of an inch in length.

---

NEW ELECTRO -MAGNET.

45

2710 before the armature was detached . Theoretically, the maximum power would be 0-635 (sectional area) x 24 x 280
= 4267 lb.
Fig. 23. Scale

a a

Fig. 24 .

X
X

The weight of the pieces of grooved iron is 7 ·025 lb., that of the plain pieces constituting the armature 4:55 lb. The power

2710

per lb. of magnetized iron was therefore

= 234 lb.

11.575

46

ON A NEW CLASS

When the apparatus is in the position which is represented

by figs. 20 and 23 , it is evident that the zigzag ring of iron

is magnetized by the conducting wire in the same way as the

plain ring ( fig. 24) would be by the passage of electricity along the wire x x which is coiled upon it ; wherever such a

ring is cut, the display of maximum lifting -power is propor tional to the least transverse sectional area of the entire

magnetic circuit. In the above position it is impossible to
consider the instrument other than a single electro -magnet ;

but when the armature is turned until the plain pieces affixed

to it cover the grooves of the other pieces, a compound

electro -magnet is formed.

I remain, dear Sir,

Yours respectfully,

Broom Hill, near Manchester,
April 30, 1841.

J. P. JOULE .

On a new Class of Magnetic Forces. By J. P. JOULE *
[' Annals of Electricity,' vol. viii. p. 219.]
As it is my intention to bring forward in this paper an electro -magnetic principle in reference partly to the motion of machines, I hope that a few preliminary observations with respect to the ordinary electro-magnetic engines will not be
deemed out of place.
The great attractive powers of the electro -magnet, and the extreme rapidity with which its polarity is reversed by changing the direction of the current, very readily present to the reflecting mind an idea that its power may be made available for mechanical purposes. Accordingly, as soon as the general principles of electro -magnetism were understood, Professor Henry, Mr. Sturgeon , and, after them , a great number of ingenious individuals constructed various arrange ments of machinery to be set in motion by magnetic attrac tion and repulsion.
• Lecture at the Victoria Gallery, Manchester, February 16, 1841.

OF MAGNETIC FORCES.

47

At that period the expectations that electro -magnetism would ultimately supersede steam , as a motive force, were very sanguine. There seemed to be nothing to prevent an enormous velocity of rotation, and consequently an enormous power, except the resistance of the air, which it was easy to remove, the resistance of iron to the induction of magnetism , which I had succeeded in overcoming to a great extent by annealing the iron bars very well, and the inertia of the
electric fluid .
We are indebted to Professor Jacobi for the exposition of the principal obstacle to the perfection of the electro -magnetic engine. He has shown that the electric action produced by the motion of the bars operates against the battery current,
and in this way reduces the magnetism of the bars, until, at a certain velocity, the forces of attraction become equivalent to the load on the axle, and the motion in consequence ceases to be accelerated . Jacobi had not, however, given precise details concerning the duty of his apparatus ; nor had he then determined the laws of the engine. I was therefore induced to construct an engine adapted for experiment, and with it
found :
1st. That the counter electric action, or, in other words, the magneto -electric resistance to the battery -current, is pro
portional to the velocity of rotation and the magnetism of
the bars.
2nd. That the economical duty at a given velocity of rota tion is a constant quantity, whatever the number of similar pairs may be in series, provided the resistance of the battery is kept constant.
3rd . That at small velocities great advantage is obtained
by reducing as far as possible the resistance of the battery, and by arranging the coils so as to facilitate as far as possible
the transmission of the current.
4th . That the economical duty at a given velocity, and for a given resistance of the battery , is proportional to the mean of the intensities of the several pairs of the battery.
With my apparatus every pound of zinc consumed in a
Grove's battery produced a mechanical force ( friction in

48

ON A NEW CLASS

cluded) equal to raise a weight of 331,400 lb. to the height of one foot, when the revolving magnets were moving at the velocity of 8 feet per second.
Now the duty of the best Cornish steam - engine is about
1,500,000 lb. raised to the height of 1 foot by the combustion of a lb. of coal, which is nearly five times the extreme duty that I was able to obtain from my electro -magnetic engine
by the consumption of a lb. of zinc. This comparison is so very unfavourable that I confess I almost despair of the success of electro -magnetic attractions as an economical source of power ; for although my machine is by no means perfect, I do not see how the arrangement of its parts could be improved so far as to make the duty per lb. of zinc superior to the duty of the best steam - engines per lb. of coal. And even if this were attained , the expense of the zinc and exciting fluids of the battery is so great, when compared with the price of coal, as to prevent the ordinary electro -magnetic engine from being useful for any but very peculiar purposes.

New Class of Magnetic Forces.
A few weeks ago an ingenious gentleman of this town suggested to me a novel form of electro -magnetic engine. He was of opinion that a bar of iron was increased in length by receiving the magnetic influence, and that, although the increment was perhaps very small, it still might be found valuable as source of power on account of the great force with which it would operate. At that gentleman's request I undertook experiments to ascertain whether his opinion was correct, and if so, to ascertain whether the new source of power could be advantageously employed for the movement of machinery.
After some preliminary trials, I adopted the following
method of experiment. A length of 30 feet of copper wire go of an inch thick , covered with cotton thread , was formed into a coil 22 inches long and of an inch in interior diameter. This coil was secured in a perpendicular position, and the rod of iron , of which I wished to ascertain the incre
ment, was suspended in its axis so as to receive the magnetic

OF MAGNETIO FOROES.

49

influence whenever a current of electricity was passed through
the coil. Lastly, the upper extremity of the rod was fixed ; and the lower extremity was attached to a system of levers which multiplied its motion three thousand times.
A bar of rectangular iron wire, 2 feet long, 7 of an inch broad, and of an inch thick , caused the index of the multi
plying apparatus to spring from its position, and vibrate about a point to of an inch in advance, when the coil was made to
- complete the circuit of a battery capable of magnetizing the iron to saturation , or nearly so . After a short interval of time the index ceased to vibrate, and began to advance very
gradually in consequence of the expansion of the bar by the heat which was radiated from the coil. On breaking the circuit, the index immediately began to vibrate about a point exactly to of an inch below that to which it had attained .
By dividing the advance of the index by the power of the levers we obtain of an inch as the increment of the bar,
30000
which may be otherwise stated as 720000 of its whole length *. Similar results were obtained by the use of an iron wire
2 feet long and I of an inch in diameter. Five cells of a nitric - acid battery produced an increment of the thirty
thousandth part of an inch ; and when only one cell was employed I had an increment very slightly less, viz. the thirtythree -thousandth of an inch .
The increment did not appear to depend upon the thick ness of the bar ; for an electro -magnet constructed of an iron bar, 3 feet long and 1 inch square, was found to expand under the magnetic influence to nearly the same extent, compared with its length, as the wires did in the previous experiments.
I made some experiments in order to ascertain the law of the increment. Their results proved it to be very nearly
proportional to the length of the bar and the intensity of its
induced magnetism .

Trial was made whether any effect could be produced by using a copper wire, which was 2 feet long and about to of

The movement of the index was rendered visible to the whole of the audience.

50

ON A NEW CLASS

an inch thick ; but I need scarcely observe that the attempt was not attended with the slightest success .
A good method of observing the above phenomena, is to examine one end of an electro -magnet with a microscope
while the other end is fixed . The increment is then observed
to take place suddenly, as if it had been occasioned by a blow struck at the other extremity. The expansion, though very minute, is indeed so very rapid that it may be felt by the touch ; and if the electro-magnet be placed perpendicularly.on . a hard elastic body, such as glass, the ear can readily detect the fact that it makes a slight jump each time that contact
is made with the battery . When one end of the electro -magnet is applied to the ear,
a distinct musical note is heard every time that contact with
the battery is made or broken - another proof of the sudden
ness with which the particles of iron are disturbed .
With regard to the application of the new force to the movement of machinery, I have nothing favourable to ad vance . An easy calculation on the basis of the modulus of elasticity of iron will show that an electro -magnet, consisting of a bar of iron 1 inch square and 3 feet long, would exert a force of about ten pounds through the length of the twenty
thousandth part of an inch every time that contact with the voltaic battery is made or broken, provided the transmitted current is capable of saturating the iron. If, therefore, con tact be made and broken a hundred times per second, for an
hour together, we shall only have fifteen pounds raised to the height of a foot. The force would therefore be far too minute for the movement of machinery ; and the duty per lb. of zinc would be vastly less than that of the ordinary electro-magnetic engine,
In examining the bearing of the new property of the electro -magnet on magnetic theory, I would observe, that in the hypothesis of Ampère the phenomena of magnetism are referred to the attraction and repulsion of currents of elec tricity moving in the same or contrary directions. Fig. 25 represents the section of six particles of a magnetized iron bar, according to a modification of that philosopher's theory.

-
--
--
1
-

OF MAGNETIC FORCES.

51

The black circles represent atoms of iron ; and the shadowed
ovals around them represent atmospheres of electricity moving in planes at right angles to the axis of the magnet.

Fig. 25.

Fig. 26.

Fig. 27.

OO

This theory affords a good explanation of most cases of magnetic attraction . But the physical conditions which are demanded by it are impossible, and contrary to the analogy of nature ; for it supposes motion, or at least an active force,
to be continued against antagonist forces for an indefinite length of time without loss, in order to explain the pheno mena exhibited by a hard steel magnet.
The way by which it may be made to account for the fact that iron, after receiving a certain quantity of magnetism , is
incapacitated from receiving a further supply, or becomes
saturated, is to suppose that the electricity which revolves round each atom of iron has a centrifugal tendency. The velocity of the electric currents around the atoms of iron will tend to be proportional to the influence which urges
them ; and if the electricity be not endowed with centrifugal
force, it is difficult to say why it should refuse to travel beyond a certain velocity ; and thus the phenomenon of satu
ration would be unexplained. If, however, the momentum of electricity, and its consequent centrifugal tendency when rotated , be admitted, the currents will be prevented from
going beyond a certain velocity by their interference with
one another.
The hypothesis, however, is less successful in accounting
E2

52

A NEW CLASS OF MAGNETIC FORCES .

for the increase of length which I have noticed in a bar of iron when under the magnetic influence; for, as the electricity is supposed to revolve in a plane at right angles to the axis of the bar, the divergence of the fluid from each atom of iron
by centrifugal force would have the effect of shortening * the bar, which is opposed to experience.
Turning now to a modified theory of Æpinus, it will be found to explain tolerably well facts which are unaccounted
for by the former theory.
Let the black circles in fig. 26 represent in section six atoms of iron , the shadowed circles about them atmospheres of magnetism , and the rings beyond these still rarer atmo
spheres of electricity. Further, let the space between these compound atoms be supposed to be filled with calorific ether in a state of vibration, or, otherwise, to be occupied with the oscillations of the atoms themselves. Such a state of things may probably give an idea of part of a bar of upmaguetized steel or iron. Now , if an inductive influence be applied to the atoms, the magnetism may be supposed to accumulate
on one side of the atoms of iron, as represented by fig . 27, and the bar rendered magnetic.
Such a theory seems to me to afford a natural and complete expression of facts. It supposes nothing which we cannot readily comprehend, except the existence and elementary properties of matter, which are necessarily assumed by every theory, and which the Great Creator has placed utterly beyond the grasp of the human understanding.
When all the magnetism of each atom of iron is accumu lated on one side, the bar may be said to be saturated .
It is obvious that when the magnetism is thus being accu mulated on one side of the particles, the bar will increase in length in the direction of its polarity, and decrease in a direction at right angles to the length. The former fact I have proved by the foregoing experiments ; and there can be little doubt that a very delicate apparatus would exhibit the
diminution of the thickness of a bar of iron in consequence of the communication of magnetic virtue.
* Further on it will be seen that a bar under strong tension is shortened by magnetization . - Note by J. P. J. in 1882.

ON VOLTAIO APPARATUS .

53

The hypothesis will also account for the fact that at a certain degree of heat all the magnetic power of iron is
destroyed . I have already observed that the space left between the magnetic particles in fig . 27 represents the room taken up by their vibration . This vibration is called
heat, and will of course increase in violence and extent with the increase of the temperature of the bar. Now it is
natural to suppose that the atoms of iron have far greater weight and inertia than the atmospheres of magnetism and electricity which surround them — therefore that these atmo spheres will be in a state of vibration, while the atoms of iron remain in a state of comparative quiescence ; and when the vibration has reached a certain intensity, the inductive influ ence will not be able to arrange the magnetism in any definite
direction with regard to the atoms of iron .
The retentive power may be explained by supposing the magnetism to adhere to the atoms of iron to a certain extent. And if we make another supposition , viz. that an atom of iron , on combining with an atom of carbon, loses its attrac tion for magnetism in the side which is next the carbon, the superior retentive power of steel, in comparison with that of iron, will be explained.

On Voltaic Apparatus *.
[ Proceedings of the London Electrical Society.]
DEAR SIR, Having recently ascertained some circumstances rela
tive to voltaic apparatus, which have, I think, considerable
value in a practical point of view , I desire to communicate my experience to you , and, through you , to the Electrical Society.
If a plate of copper be placed in dilute sulphuric acid , it will be gradually dissolved, and, after a certain length of time, the liquid will be found to have acquired a blue tinge owing to the solution of oxide of copper. If now a plate of
• In a letter to C. V. Walker, Sec. Lond. Electr. Soc.

64

ON VOLTAIC APPARATUS.

amalgamated zinc be placed in voltaic association with the copper, a powerful current will pass along the connecting wire, equal in intensity * to that which would have been pro duced by a Daniell's cell. But in the meantime a part of the copper in solution will have precipitated itself on the amalgamated zinc, causing a local action, which will speedily destroy the plate.
But if, instead of allowing the copper to remain alone in the acid before the battery is completed , it be placed in the dilute sulphuric acid along with a piece of amalgamated zinc, connected with it by means of a copper wire, the pair thus formed will not, after the immediate effects of
immersion are passed away, be as intense as was the former
one ; but then it will work without local action.
The due consideration of these facts will enable us to understand why local action is so common an annoyance to those who work with acid batteries in which copper is employed as the negative element. It will also point out the following means for remedying that defect :
1st. Every part of the copper surface which is immersed in the liquid should be, su to speak, in sight of the amal. gamated zinc. Any part not so situated is not actively engaged in propagating the current, and is consequently liable to enter into solution, and then to be precipitated
on the zinc. If we use the Wollaston's arrangement, we
should not ( as is common ) bend the copper about the zinc,
but the zinc about the copper.
2nd. When the copper battery is immersed in its trough, it should be set to work immediately, and the current should be allowed to pass with as few intermissions as possible. If we wish to make such a change in our apparatus as will occupy any length of time, we should either take the
battery out of its trough, or else we should connect its poles by a wire, the conducting -power of which is sufficient to occasion a slight effervescence on the copper surfaces.
• It will be proper to observe that, throughout this paper, I mean by the word " intensity " electromotive force. It is always proportional to the quantity of current multiplied by the resistance of thewhole circuit.

-

ON VOLTAIC APPARATUS.

56

By adopting these precautions, I find that I am able to use a copper battery, charged with a strong solution of sulphuric acid, without being annoyed by local action .
I now endeavoured to ascertain the best method of giving the copper a surface adapted for the transmission of a large quantity of electricity. I tried the platinizing process first, and found that, after a plate of copper had been immersed for
an instant in an extremely dilute solution of platinum , its surface was closely assimilated in electric character to that of a plate of platinized silver - the quantity of platinum required for the purpose being far too small to be regarded in an economical point of view . Then I tried the plans which had been suggested by Professor Poggendorff and yourself*, particularly that of heating the copper to redness in air a method which has, I think, some advantage over the other
processes .
By this means, and by the precautions against local action
of which I have previously spoken, I was able to construct a copper -zinc acid battery, not much inferior in any respect to that of Mr. Smee, while at the same time the stabilityt of the copper plates gave me the advantage of a superior mechanical arrangement.

Fig. 28.

UDIO
The arrangement to which I am alluding is so convenient
in practice that, although it has, with the order of metals reversed, already received publication in the ‘ Annals of
• Proc. Electr. Soc. part ii. p. 92 ; also part i. p. 30. + I have tried the platinized plated copper suggested by Mr. Smee. Its stability renders it much better adapted for the mechanical arrangement of a battery than the platinized silver foil.

56

ON VOLTAIC APPARATUS.

Electricity '*, I hope to be permitted to describe it here, connected as it is with the general object of this com
munication .

Fig. 28 (p. 55) represents my battery. It is adapted for the common divided trough, and is constructed as follows :-1
take ten pieces of sheet zinc, each of which is 10 inches long,
34 inches broad , and about to of an inch thick . These I
bend over a gauge, into the shape which, for the Fig. 29.
sake of clear illustration, I have delineated in
fig. 29. I amalgamate those parts of them which will have to be immersed in the acid, polish the
top of each with glass -paper, and place in the bent part of each a piece of semicircular wood, furnished with a longitudinal saw - cut. Having
thus prepared the zincs, I place them in the cells of an empty trough . I then take ten plates of thick sheet copper, bend
them into the shape represented by fig. 30, Fig. 30.
prepare their surfaces by red heat, polish that part which is to come into contact with the
zinc, and fix them within the zincs by inserting
their bottom edges in the saw - cuts of which I have spoken. Lastly, the pieces of wood, a a a &c., are put in their places, a screw -bolt is passed through the whole series by means of a hole in the centre
of each , and the whole is secured by screwing on the nut.
When my object is to construct a battery of a large number
of small plates, I dispense with the pieces of wood and screw

Fig. 31.

tttttttttt
bolt, and simply insert the plates in the transverse saw - cuts of a piece of wood . This battery is represented by fig. 31 .
With a battery of ten pairs, similar to that represented by fig . 28, furnished with coppers which had been prepared by
• Vol. v. p. 197 .

--

ON VOLTAIC APPARATUS.

67

heat, and immersed in a dilute solution of sulphuric acid, I have raised 10 inches of platinum wire 3d of an inch thick , to a white heat. The great intensity, however, which produced
such striking calorific effects passed off in a minute or two ;
gas began to arise from the copper ; and the intensity became
constant, and about one half of what it was at first. The
inferior constant intensity was the principal remaining defect of the copper -zinc and acid battery ; I therefore endeavoured to devise means for improving its electromotive force. In this inquiry I ascertained the following facts :
1st. That the intensities of the arrangements of Grove and Daniell are very nearly in the ratio of five to three. My experiments on this point confirm those of Professor Jacobi,
who finds these intensities to be in the ratio of 22515 to
13552 *.
2nd. That Professor Daniell's battery has the same in tensity, whether the fluid in contact with the copper holds in solution the nitrate or the sulphate of copper .
3rd . That copper-zinc and platinum -zinc pairs have re spectively the same intensities as the arrangements of Daniell and Grove, if they are immersed in solutions of nitric acid sufficiently strong to prevent the evolution of hydrogen gas
from the negative elements. And
4th . That the intensity of the copper-zinc battery, at the instant of its immersion in dilute sulphuric acid, is equal to
that of a Daniell's cell .
With the view of making a practical use of the fourth fact, I caused a circular plate of copper, ten inches in diameter, and dipping to the depth of nearly three inches in dilute sul phuric acid, to be fixed on a horizontal axis, connexion being maintained, through the galvanometer, between it and a plate of amalgamated zinc. On revolving the disk, each part of its circumference was alternately exposed to the atmosphere and immersed in the acid . The electric current generated in these circumstances was equal in intensity to that produced by a Daniell’s cell, when it had to traverse 120 yards of copper wire of of an inch diameter. But when a wire of greater
• Vide Proc. Electr. Soc. part i. p. 40.

58

ON VOLTAIC APPARATUS.

conducting -power was used, the intensity, though considerably
greater than it would have been without the revolution of the
plate, was not equal to that standard , the quantity of oxygen absorbed from the atmosphere not being equivalent to the quantity ofcurrent produced in this latter case . By increasing the size of the disk, I have no doubt that the intensity would be raised as high as that of a Daniell’s cell, even when the elements are connected by a very short and thick wire ; but in that case the cumbrousness of the apparatus would, I apprehend, preclude it from being practically useful * .
I have also attempted to make a practical use of the third fact. I was aware that you had condemned the use of nitric
acid in the charge of the acid battery ; but, remembering that it had been used by Dr. Faraday in a very dilute state without producing much local actiont, I did not like to relinquish , without another trial, the great intensity which nitric acid gives to the acid battery.
I will not detain you by describing the extensive series of experiments which I have made on this subject. I will only
observe that I could work with an acid composed of 16 measures of water, 3 measures of strong oil of vitriol, and
1 measure of nitric acid sp. gr . 1.3, without being annoyed by much local action, the intensity, even when the circuit was completed by a short wire, being equal to that of a Daniell's
cell, and the quantity of electricity transmitted much greater
than in a comparative Daniell's cell. Still I was obliged ultimately to revert to your opinion ; for the zinc plates were generally acted upon violently after they had been used a few
times .
There is also an objection against the use of nitric acid, which has some weight in an economical point of view ; for when it is used, there is a portion of nitric acid decomposed by nascent hydrogen from the negative plate, besides the usual equivalent of nitric or sulphuric acid.
I have no further observations to make on batteries without
diaphragms, but am not willing to conclude without testifying
* See further on , where the effects of immersion are studied. + Experimental Researches (1139).

HEAT PRODUCED BY VOLTAIC ELECTRICITY . 59

my sense of the value of your method of constructing the acid

battery with diaphragms for long -continued and constant

action, such as is required for electrotype. I have myself used your process, and have witnessed with pleasure and

surprise the rapidity with which the copper spread itself over
the black lead which I had rubbed on the inner surface of &

jar more than two feet deep and three inches in diameter.

You state that dilute sulphuric acid alone produces a current of sufficient intensity for electrotype : for other purposes,

however, it is desirable to use the cupreous solution mingled

with a small quantity of free sulphuric acid, in contact with

the copper of a diaphragm battery,. I am, yours truly ,

Broom Hill, near Manchester, March 12th, 1842.

JAMES P. JOULE .

On the Production of Heat by Voltaic Electricity. By J. P. JOULE *
[ Proceedings of the Royal Society , December 17, 1840.]
Tae inquiries of the author are directed to the investigation of the cause of the different degrees of facility with which
various kinds of metal, of different sizes, are heated by the
passage of voltaic electricity . The apparatus he employed for this purpose consisted of a coil of the wire, which was to be subjected to trial, placed in a jar of water, of which the
change of temperature was measured by a very sensible ther
mometer immersed in it ; and of a galvanometer, to indicate the quantity of electricity sent through the wire, which was estimated by the quantity of water decomposed by that elec tricity. The conclusion he draws from the results of his experiments is, that the calorific effects of equal quantities of transmitted electricity are proportional to the resistance
opposed to its passage, whatever may be the length, thickness,
• The experiments were made at Broom Hill, near Manchester.

80

HEAT EVOLVED BY METALLIC
shape, or kind of metal which closes the circuit; and also that, cæteris paribus, these effects are in the duplicate ratio of the quantities of transmitted electricity, and , consequently, also in the duplicate ratio of the velocity of transmission. He also infers from his researches that the heat produced by the combustion of zinc in oxygen is likewise the consequence
of resistance to electric conduction .
On the Heat evolved by Metallic Conductors of Elec tricity, and in the Cells of a Battery during Elec trolysis. By JAMES PRESCOTT JOULE, Esq.*
[ ' Philosophical Magazine,' vol. xix . p. 260.]
1. THERE are few facts in science more interesting than those which establish a connexion between heat and electricity. Their value, indeed, cannot be estimated rightly, until we obtain a complete knowledge of the grand agents upon which they shed so much light. I hope, therefore, that the results of my careful investigation on the heat produced by voltaic action are of sufficient interest to justify me in laying them before the Royal Society.
CHAP. I. — Heat evolved by Metallic Conductors. 2. It is well known that the facility with which a metallic wire is heated by the voltaic current is in inverse proportion
to its conducting -power ; and it is generally believed that this proportion is exact ; nevertheless I wished to ascertain the fact for my own satisfaction, and especially as it was of the utmost importance to know whether resistance to conduction
is the sole cause of the heating effects. The detail, therefore,
of some experiments confirmatory of the law, in addition to
those already recorded in the pages ofscience, will not, I hope, be deemed superfluous.
• The experiments were made at Broom Hill, Pendlebury, near
Manchester.

CONDUCTORS OF ELECTRICITY .

61

3. It was absolutely essential to work with a galvanometer the indications of which could be depended upon as marking
definite quantities of electricity. I bent a rod of copper into the shape of a rectangle (AB, fig . 32), 12 inches long and

Fig. 32.

A

N

BВ

C

6 inches broad . This I secured in a vertical position by means of the block of wood C ; N is the magnetic needle,
34 inches long, pointed at its extremities, and suspended upon
a fine steel pivot over a graduated card placed a little before the centre of the instrument.
4. On account of the large relative size of the rectangular conductor of my galvanometer, the tangents of the deviations of the needle are very nearly proportional to the quantities of current electricity. The small correction which it is necessary to apply to the tangents, I obtained by means of the rigorous experimental process which I have some time ago described in the ' Annals of Electricity ' *.
5. I have expressed my quantities of electricity on the basis of Faraday's great discovery of definite electrolysis ; and I venture to suggest that that quantity of current electricity which is able to electrolyze a chemical equivalent expressed in grains in one hour of time, be called a degree. Now , by a number of experiments I found that the needle of my galva nometer deviated 330.5 of the graduated card when a current was passing in sufficient quantity to decompose nine grains of water per hour ; that deviation , therefore, indicates one degree of current electricity on the scale that I propose to
• Vol. iv. pp. 131, 132, and 476.