THE LI BRARY O F YORK UNIVERSITY 3 9007 0297 5619 4 Date Due FARADAY'S DIARY VOL. IV Faraday's great electro-magnet The powerful electro-magnet (pars. 7874, 8408, November 1845), constructed from the link of a great chain cable, and used by Faraday in his experiments on diamagnetism FARADAY'S DIARY Being the Various Philosophical Notes of Experimental Investigation made by MICHAEL FARADAY D.C.L., F.R.S. during the years 1820 -1862 and bequeathed by him to the ROYAL INSTITUTION OF GREAT BRITAIN Now, by order of the Managers, printed and published for the first time, under the editorial supervision of THOMAS MARTIN, M.Sc. with a Foreword by SIR WILLIAM H. BRAGG, O.M., K.B.E., F.R.S. Director of the Laboratory of the Royal Institution Vol. IV NOV. 12, 1839-JUNE 26, 1847 LONDON G. BELL AND SONS, LTD 1933 113 PRINTED IN GREAT BRITAIN BY W. LEWIS, M.A., AT THE UNIVERSITY PRESS, CAMBRIDGE ; CONTENTS The cross references are to the page numbers in the collected edition of the Experimental Researches in Electricity and the Experimental Researches in Chemistry and Physics. FOLIO VOLUME IV OF MANUSCRIPT (continued) 1839 November 12. 5764-5825. Voltaic experiments with pairs of wires in various electrolytes (continued) ..... possible thermo-electric effects. Fuzed electro- lytes. Conclusions. pages '^-i^ See Exptl. Res. Electy., vol. 11. Seventeenth Series. November 12. 5826. Henry's inductive coils . . page 14 November 14, 16. 5827-5852. Dilution of the electrolyte. Voltaic circuits without metallic contact . pages 14-20 See Exptl. Res. Electy., vol. n. Seventeenth Series. ...... November 16 to 22. 5853-5994. Voltaic relations of metals with plumbago and oxides ; plumbago as a neutral substance pages 21-40 See Exptl. Res. Electy., vol. 11. Seventeenth Series. 1840 January 10 to February 11. 5995-6087. Voltaic experiments concluded. Order of ten metals in seven solutions (6041) pages 41-^1 See Exptl. Res. Electy., vol. 11. Seventeenth Series. August 10, II. 6088-6149. Electro-magnetic induction: theory of induction; the electrotonic state . August 12 to September i. 6150-6192. Inductive action of an induced current; secondary and tertiary currents; effects of the "magneto-electric" cur- rent found identical with those of the voltaic current (6 1 87). Specification of coils used . September 14. 6 1 93-6 1 97. An induction experiment with Gassiot; spark in hot air . . . . pages 52-57 pages 58-64 pages 64-65 vi CONTENTS 1842 June I to 30. 6201-63 13. Electricity generated by the friction of steam issuing from an orifice. Electrical states of the issuing steam and insulated boiler. Nozzles of various shapes tried; a steam ..... A globe introduced. "steam electricity battery" contemplated (6248) pages 6(y-jS See Exptl. Res. Electy,^ vol. II, pp. 106-126. Eighteenth Series. On the electricity evolved by the friction of water and steam against other bodies. September 8 to 26. 6314-6367. Regelation; ice and water in contact; freezing together of pieces of ice, and of ice and various materials . . . pages 79-83 November 9 to December 23. 63 68-669 1, 6697-6768. Steam experiments resumed : effect of varying the form of the steam vent; various tubes tried; a cone apparatus constructed. Liquids injected into the issuing jet, and various substances inter- posed in its path. Action of oil in the jet . . pages 84-124 See Exptl. Res. Electy., vol. ii. Eighteenth Series. December xo^^^. 6692-6696. Electrification by friction: states of materials rubbed together . . . pages 116-117 December 23 to 29. 6769-6776, 6783-6875. Com- pressed air used in place of steam: electrification by liquids, dry powders, etc. introduced into the air jet . . . . . . . . pages 124-132 See Exptl. Res. Electy., vol. 11. Eighteenth Series. ........ December 24, 29. 6777-6782, 6876. Action of oil films on water surfaces ; a surface tension experi- ment pages 125, 132 See Exptl. Res. Electy., vol. 11. Eighteenth Series. December 30, 1842 to February 16, 1843. 6877-6966. Electrification by steam and air jets: further trials and modifications. Materials rubbed together; an order of electrification determined (6935); excitation with ice . . . . . . pages 132-143 See Exptl. Res. Electy., vol. 11. Eighteenth Series. CONTENTS vii 1844 ...... February 12. 6967-6971. Examination of a fractured Leyden jar . pages 144-145 ....... A March 9. 6972. proposed registering atmospheric electrometer pages 145-147 May 23 to June 15. 6973-7068. Liquefaction and solidification of gases by compression and cooling in closed tubes. Olefiant gas liquefied (6989); sulphurous and nitrous acids solidified (6998-9); a new method of drying the gases (7003); hydriodic acid liquefied and solidified (7022); cyanogen solidified (7039); hydrobromic acid liquefied (7058) pages 147-163 See Exptl. Res. Chem. Phys., pp. 96-124. On the Liquefaction and Solidification of Bodies generally existing as Gases. A July 10. 7069-7084. At Hampstead. lime-light experiment with the sun's rays . . . pages 163-164 August 2.2. to October 25. 7085-7148, 7152-7185. Liquefaction of gases. The cooling bath in vacuo: lower temperatures reached. Attempts to liquefy oxygen, hydrogen, nitrogen, etc. Specific gravities of liquid cyanogen (7 1 19) and ammonia (7 1 32) determined. Vapour pressures over ranges of temperature observed and tabulated. Fluosilicon liquefied (7154); ammonia (7168) and sul- phuretted hydrogen (7169) solidified . . pages 165-186 See Exptl. Res. Chem. Phys., pp. 96-124. ..... A September 13. 7149-7151. tree in Greenwich Park struck by lightning pages 178-179 November 4, 1844 to January 10, 1845. 7 1 86-7244, 7246-7299. Liquefaction of gases. Nitrous oxide (7188), euchlorine (7209) and hydrobromic acid (7214) solidified. Cagniard de la Tour's experiment. Vapour pressure determina- tions. Pressure tests on tubes at Mr Addams' ...... (7255 et seq.). Freezing points determined. Phosphuretted hydrogen (7276) and fluoboron (7287) liquefied pages 187-215 See Exptl. Res. Chem. Phys., pp. 96—124. ....... November 23. 7245. Matteucci unable to repeat an experiment pages 197-198 Vlll CONTENTS 1845 FOLIO VOLUME V OF MANUSCRIPT January 10 to February 18. 7300-7372, 7378-7380. Liquefaction of gases (continued). Vapour pressures tabulated; irregularities caused by impurities. Preparation of the gases. Thermo- meters (7306, 7) . . . . . . pages 219-246 See Exptl. Res. Chem. Phys.^ pp. 96—124. February 14, June 12. JS'J'^-JZll-) 74^9- Liquefied gases sealed up in tubes with acids. Examined (7419) P^g^^ 244, 253 ...... February 27, March i. 7381-7396. Solubility of gases in various liquids pages 246-249 May 8. 7397-7400. defiant gas from Prof. Graham page 250 May ij to 28. 7401-741 8. Metals and metallic com- ..... pounds at low and high temperatures: their magnetic properties, etc. pages 250-253 On See Exptl. Res. Electy., vol. Ill, pp. 444-446. the Magnetic Relations and Characters of the Metals. ...... August 18. 7420-7432. Solubility of nitrous oxide in various fluids pages 253-254 August 2.1. 7433. FlexibiHty of glass under pressure . page 7.^^ August 30 to September 5. 7434-7497. Polarized light in conducting electrolytes, and in transparent No dielectrics. effects of electrification on the light (7497) pages 256-263 September 13, 16. 7498-7537. Action of magnetism on light: a polarized ray passed through trans- An parent bodies in the magnetic field. effect on the ray found with heavy glass (7504) . . pages 26^^-267 On See Exptl. Res. Electy.^ vol. iii, pp. 1-26. Nineteenth Series. the magnetization of light and the illumination of magnetic lines of force, (i) Action of magnets on light, (ii) Action of electric currents on light, (iii) General considerations. A September 18 to 26. 7538-7654, 7657-7688. more powerful magnet borrowed; result with heavy glass confirmed, and effect on the polarized ray found to be rotatory. Best positions of the glass. CONTENTS ix Direction of the rotation. The effect found in a variety of solutions, oils and fuzed bodies, but not in gases, and only slightly in crystals . . pages 26j-^%6 See Exptl. Res. Electy., vol. iii. Nineteenth Series. September 22 to October 6. 7655-7656, 7689-7743. Miscellaneous: the heavy glass not magnetic; its action not influenced by electrostatic force or ........ heat. Electrolytes subjected to the simultaneous action of electric currents, magnetic forces and light pages 282-293 See Exptl. Res. Electy., vol. iii. Nineteenth Series. October 11 to November 5. 7744-7873, 7912-7922. Action of electricity on light: the polarized ray affected by a current-carrying coil as by magnets. Coils and tubes constructed. An effect of "extra light" traced to the heating action of the coils. ...... A reciprocal action of sunlight on a conducting circuit looked for pages 293-315 See Exptl. Res. Electy., vol. ill. Nineteenth Series. A November 3. 7874-7901. great horseshoe electro- magnet made. Magnetic rotation of polarized light measured in various substances . . pages 310-312 See Exptl. Res. Electy., vol. iii. Nineteenth Series. A November 4 to 10. 7902-79 1 1, 7923-8 1 07. bar of heavy glass suspended between the poles of the great magnet; a new property (diamagnetism) discovered (7902). The effect found in a great variety of substances, including liquids; peculiar behaviour of copper; bismuth found to exhibit the new property strongly . . . • Z*^^^*^ 3i3~333 See Exptl. Res. Electy., vol. iii, pp. 27-53. Twentieth Series. On new magnetic actions, and on the magnetic condition of all matter, (i) Apparatus required, (ii) Action of magnets on heavy glass, (iii) Action of magnets on other substances acting mag- netically on light, (iv) Action of magnets on the metals generally. November 10 to 15. 8108-8190. The field of the great magnet examined with a suspended bismuth bar. Metals, etc. suspended in liquids between the A poles. Conclusions. magnetic order of sub- stances (8180) pages 333-342 See Exptl. Res. Electy., vol. ill. Twentieth Series. X CONTENTS November A i<). 8191-8323. single magnetic pole set up: motions in its field examined with bis- A muth, heavy glass, copper, etc. conical termina- tion fitted to the pole. Small cubes and spheres, and powders, used as indicators . . . pages 342-357 See Exptl. Res, Electy., vol. iii. Twentieth Series. .... November 19. 8324-8329. EfTect of heat on the mag- netism of iron and nickel pages 357-358 See Exptl. Res. Electy., vol. Ill, pp. 54-82. Twenty-first Series. On new magnetic actions, and on the magnetic condition of all matter (continued), (v) Action of magnets on the magnetic metals and their compounds, (vi) Action of magnets on air and gases, (vii) General considerations. November 19 to 26. 8330-8454. Action of magnets on air and gases : tubes of air, etc. suspended in fluid media between the magnetic poles; bodies sus- pended in gases and in vacuo; various experi- ments on metals. Coils added to the great magnet (8362, 8408, 8409) pages 358-371 See Exptl. Res. Electy., vol. iii. Twenty-first Series. December i to 23. 8455-8640. Ferromagnetism and diamagnetism in metals: motions, in the magnetic field, of a series of metals and metallic com- pounds; a magnetic order of the metals. Liquids and gases in fluid media: further tests . . pages 372-390 See Exptl. Res. Electy., vol. iii. Twenty-first Series. 1846 January 10, 13. 8641-8665. Polarized light: a mag- netic effect sought on opposed rays in the same path pages 391-395 January 13. 8666-8672. Cobalt from Dr Miller, etc. ...... January 15. 8673-8674. An experiment on polarization for Herschell page 395 page 396 .... February 26. 8675-868 1. Brush and spark discharges in magnetic fields. References pages 396-397 March 9, 12, June 29. 8682-8694, 8695-8706. Mag- netism and light: a reciprocal inductive action of light sought in electro-magnetic circuits . . pages 397-401 1 CONTENTS xi July 24 to September ij. 8707-8747, 8755-8780. Magnetism and light: the effect in heavy glass magnified by repeated reflections; the method applied to air and crystals. Diamagnetism: the .... pointing of suspended tubes and bars of oxides, etc. before magnetic poles pages 401-412 On See Exptl. Res. Electy., vol. iii, pp. 453-466. the Magnetic Affection of Light, and on the Distinction between the Ferro- magnetic and Diamagnetic Conditions of Matter. ........ August 3. 8748-8754. Diamagnetic bodies introduced into wire coils: an electrical inductive action sought pages 407-408 October 17, 24. 8781-8839. Electro-magnetic in- duction: intensity of inductive force inside and outside a cylindrical coil ; action of a bismuth core; ...... induction in rotating metal cylinders. Miscel- laneous. . pages 412-^10 November 6. 8840-8842. Magnetism and light: heavy glass rotated between magnetic poles . . page 420 November 6 to 14. 8843-8913. Electro-magnetic induction: numerous experiments with magnets, coils, etc. with a view to detecting some con- tinuous effect in the circuit under induction . pages 420-43 November 14. 89 1 4-8922. Magnetism and polarized light: further experiments . . . . /^a^^i- 431-432 ...... November 23. 8923-893 1. Electrostatic induction: an experiment pages 432-434 December 31. 8932-8946. Freezing of aqueous solu- tions: purity of the ice . . . . . pages 435-437 1847 January i. 8947. References page 437 ....... January 2. 8948-896 1. Freezing of water with gases in solution pages 437—439 January 19. 8962-8973, 8976-8978. Magnetism and light: rotation of a cylinder of heavy glass .... between the poles of the great magnet; coloured light and coloured media, etc. pages 439-443 xii CONTENTS January 19. 8974-8975. Electro-magnetic induction: currents in a rotating metal disc . . . pages 441-442 March 25. 8979-8997. Hot wires in gases: action of hydrogen. Light from a hot wire in the magnetic field pages 443-445 .... April 22, May 24. 8998-9020. Iodide of nitrogen: preparation and experiments pages 445-448 ...... June 26. 9021. At Oxford: a discussion with Sir William Hamilton page 448 PLATES The great electromagnet constructed in 1845 . . Frontispiece .... The entry recording the discovery of an effect of magnetism on light fa-cing page 264 FOLIO VOLUME IV OF MANUSCRIPT (CONTINUED) 12 NOVR. 1839. 5764. Made some more experiments of the same kind. Thus, Nickel and Silver both being in weak Nitric acid, the nickel was Positive; both being in strong nitric acid, the silver was positive; but the state was preceeded by a moment in which the Nickel was Pos. So this is no case of inversion in strong and weak acid, except in consequence of the investing solution in the Nickel. 5765. Nickel and Copper in Strong Nitric acid. Copper was Positive from the first moment, strongly and constantly so, but in the dilute Nitric acid the copper was slightly but clearly Negative to the nickel. So this a good case of inversion with the same metals and the same acid. How then can we refer any of the effects to contact, which remains the same throughout? 5766. Again, Zinc and Cadmium being in strong nitric acid, the Cadmium is Pos. strongly and clearly to the zinc— but being in dilute nitric acid, the zinc is very pos. to the cadmium. So here a case as beautiful of contrast— against contact and for chemical action. 5767. Suppose a contact man were to say that it is only the very strongest acid that is able to render a metal negative to a piece of the same metal in dilute acid, and that the first portion of water added to the nitric acid makes it another thing as respects its contact power— then how far can this be carried; for Iron in a dilute + nitric acid consisting of i vol. acid 20 water is Positive to Iron in Strong Nitric acid, or in a dilute Nitric acid consisting of i vol. strong Acid and i vol. of water, or in a more dilute acid consisting of I vol. strong acid and 3 volumes of water, or in an acid con- sisting of I vol. strong acid and 5 vols, of water. Silver also, in the most dilute acid, is positive to silver in the four stronger states of Now the acid. how can this agree with the theory of contact ? 5768. Or if it is said that the force of contact of the acid, or rather its difference of force, becomes gradually greater as the acid is more and more dilute, the metal in the stronger acid always being Negative to the metal in the weaker; i.e. in the cases, as of nitric acid, where the weak acid is the one making the metal Positive (the reverse of course being supposed to be the case where the 1-2 4 12 NOVR. 1839. stronger acid determines the Pos. side (5712, 5713)); then how can the following cases be accounted for ? 5769. Copper Copper being in Strong Nitric acid, and in nitric acid diluted with three volume[s] of water, the latter was Positive to the former, and so far the results agree with the former results and with the notion just expressed. But Copper and Copper^ being + in the solution of i vol. Acid 3 vol. water, and another solution of I vol. acid and 20 vols, water, the latter was Negative to the former, instead of being Positive as in the cases of Iron and Silver. So that here an Acid of i vol. strong N. A. and 3 vol. water made the copper in it Positive to other copper, whether it were in a stronger or a weaker acid of the same nature. Surely contact could not change in this manner. And it was not the first addition of water which did this; for an acid consisting of i vol. Nitric acid and one vol. of water is with copper and copper as dilute acid to the strong acid, and as strong acid to the most dilute, i.e. + I A. 20 water. It is only when further dilute that it loses its relation of Neg. to the weakest acid and becomes Positive to both it and the strongest acid. 5770. Lead Lead. The case is the same and better with Lead, for the Acid of i vol. A., 3 vols, water makes lead in it well Pos. to lead, either in the weakest acid and the strongest acid. With it also dilution with i vol. of water is not enough to bring on this beautiful intermediate acid. The proportion of i and 3 is probably not the best— time would be required to work it out. 5771. Made some experiments with Sulphuric acid which I will enter, but must repeat them carefully: if correct, they afford still more curious contrasts. A was strong Sulc. acid. + B, I vol. strong S. A. i vol. water. + C, I vol. strong S. A. 20 vols, water. 5772. Tin Tin in these acids, two together in tube no. i. The A tin in was Pos. to tin in B or C, and tin in B was Pos. to tin in C. So that in the strongest acid was always Pos. to that in the weaker acid as before (5713). A 5773. Iron Iron and these acids. The iron in was Neg. to that in B or in C, and the iron in B was Neg. to that in C. So here the A iron in the weaker acid was always the Positive. striking 12 NOVR. 1839. 5 contrast to the Tin and I believe a true result (571 1). Though not in favour of contact. A 5774. Lead Lead in these acids. Lead in was Pos. to that in B C or in (5712). But lead in B was Neg. to that in C. So that B acid rendered the lead Negative either to lead in acid stronger or weaker (5828). A 5775. Copper Copper in these acids. Copper in was a little Neg. to copper in B, and it was in the former case Neg. to copper C in (5710). But copper in B is Pos. to copper in C. So copper in B is Positive to either stronger or weaker acid (5829). 5776. It is these two cases of contrast in Lead and copper that I must carefully repeat. 5777. Endeavoured to ascertain, if possible, whether there was really any thermo electric effect between Metals and fluids such as Potash, acid, etc. etc., or in fact any electrolytes, and especially those consisting of aqueous solutions. 5778. Began with a very strong solution of Caustic potassa. It was made of potassa fuza (good) and was cleared by decantation. 5779. Platina; Pladna; strong sol. SuL potassa. All was quiet and the needle at 0° when the whole was at common temperatures. By heating the side B I could deflect the needle about a degree, 1°, On B being Positive. heating A, it became Pos. full 15*^, and this went down on cooling. Reheated B side and it became Pos. io°— cooled it and it went back. Again heated B side— it was at the first moment the least degree Neg., but became Pos. as it boiled. Found that the boiling had thrown a little spurt of potash up to the junction of the Platina and copper at c, and this made that side much Positive, which, though it went down somewhat, still in part continued. One would have thought the metallic contact there would have been quite enough to have prevented such a circumstance having any power to produce a current, but it did not. Cleaned this away, and renewing the contact, left all quiet for half an hour to wear out irregularities. All was then cold and the needle at 0°. Heated A: it became Pos. about 5°; cooled it: it went down again. Heated B: it became Pos. about the same; cooled it: it went down again. As the heat rose the effect appeared apparently in proportion to it; but when all was hot and nearly boiling, on making it quite to boil it became much more Pos., as 6 12 NOVR. 1839. if boiling itself did something, for the difference in the heat of the fluid must have been small at that moment. 5780. Boiling by sending steam, etc. up to the higher part of the wire would heat it quickly, and by diminishing conduction from the immersed end would cause it to be hotter. Perhaps that was the cause of the increased effect. 5781. If the effect is thermo electric^ then Platina heated in sol. Potash is Pos, to Platina in the cold part of the same solution, but not more than 5° by diffnce. of about 60° and 212° F. 5782. Gold; Gold; and sol. strong Potash. At first the needle at 0°. Heated A: it became about 3° Pos., or when boiling, 5° or 6°. On A cooling the needle returned to 0°. Heated B : it became Pos. in turn about as much; on cooling, the needle returned to 0°. The fact of the boiling increasing the deflection is rather in favour of the idea that it is really thermo electric. 5783. The distance of the metal ends in these cases was about as much as in the various former experiments on chemical force, and therefore the chemical and the thermo effects are so far comparable. But the potash was very strong and therefore a very good con- ductor, far better than any of the diluted acids used. The thermo effect here is therefore by so much magnified. 5784. Iron; Iron; Strong sol. Potash. Was neutral and at 0° : then A heated and it became Pos. to 40° or more. On examining it, could see a cloud forming at it, when left quiet, apparently of prot oxide of iron. The effect gradually fell as temperature fell. Heated A again, which raised the Pos. state somewhat, but it looked as if there were an investing effect at A. Left all a minute, and then B of a sudden and spontaneously became Pos., though it had not been touched or heated— heated B to boiling, which raised it Pos. to 80°— left it and that state gradually went down— heated B again and the state rose again, but not so high as before. 5785. The wires looked bright when brought out, but were tarnished at the part in the air above the solution. 5786. So the Iron is evidently a good case of a current produced by chemical action and not by contact. It is in good contrast with those of contact, as Platina and Gold. 5787. Proceeded to use the strong solution of sulphuret ofPotash (5263, 5268) in place of Potash, to compare supposed thermo : 12 NOVR. 1839. 7 current of it on Gold and platina, as in Potash. It is a most excellent conductor of feeble currents. 5788. Platina; Pladna; Sulrt. Potash, Some motion, shewing some difference in wires, and that something besides heat can act with them at present. There was but a small and uncertain effect A and after a while the needle was at 0°. Warmed and boiled side —the Platina there became steadily Neg. 10° or about— increased to 20° gradually, i.e. after the spirit lamp was withdrawn. Cooled A and the needle went back to 0°. Warmed B : it became a very little Neg., perhaps i°— removed the lamp and left it and this increased a little— renewed the heat and the Neg. effect diminished —then left alone, it again increased gradually up to 7°— it fell slowly, but applying a little heat diminished it faster. Cooling at A first made it still more Neg. and then it fell to 0°. Heated again it did not become sensibly Neg. now during the time of heating, but afterwards on being left to stand rose to Neg. 12° or 15°. Heating lowd. it to 0°; then being left it became Neg. againheating again lowered it about 1°, and then being left, it became Neg. again. 5789. The effect is I believe chemical and not thermo electric. I think it is principally due to the action of the air (a current of which the heat tends to form in the tube) upon the hot sulphuret —the effect of the heat, when applied to lower the Neg. state, may be due to its forming currents, and so by mixing up the solution to destroy in part its polar state. Being cooled, all fell to 0°. 5790. At all event, the heat did not render the wire heated Pos., whether it be a thermo electric effect or not. 5791. Palladium i Palladium; Sulrt. Pot, strong solution. At first was 0°. Heated B : it became the least trace Neg.^ which went off in part during the heating— after a while it was at Neg. 2°— reheated B, which reduced its Neg. to o°— stillness brought it up to 3° again— is the air acting here— cooled B: its Neg. went down to 1° or 2°. Heated A: it became Pos. up to 3°, but on cooHng A returned to 0°; on reheating again, it became Pos. 2°. Here therefore there is no evidence of thermo effect or even of chemical effect. The two wires are not quite alike, but still the whole difference is very small; 2° or 3° only. 5792. It certainly does seem in many experiments as if Platina had 8 12 NOVR. 1839. a power— not mere indifference, but of a peculiar kind; here it has been much more effective in producing changes than Palladium. On 5793. taking out the Palladium wires, a little powder as dirt A wiped off from the Palladium and a very little also from B. These probably indicate some chemical action. Now 5794. Iron Iron in Sulphuret of Potash strong solution. A Wire B was Pos. to at the immersion— soon came to 0°. Heated A: the needle scarcely moved— cooled it: still quiet. Heated B: it was the smallest degree Pos. and then returned— then left quiet, On was Neg. 1°. cooling it retd. to 0°. The effect was very On small indeed and is as nothing. taking out the wires, the parts in the solution were bright and wiped clean on cloth, but a part of one which was in the air in the tube above the solution was tarnished and wiped black on the cloth. 5795. Diluted the solution i vol. to 12 of water and replaced the Irons, to see if action on them when dilute solution used (5647). At the first B was a little Pos. Heated B: there was very little change of the needle, but on cooling it, it became Pos. 8° or 9° and there was evident action, for the yellow colour of the solution was rendered rather green and sulphuret was there. Still, the lower part of the iron was bright and the upper part in the air was black; and the sulphuret may have formed at the upper part and washed down with the boiling. 5796*. Put some of the same solution into a tube bent and closed at one end. Put in a piece of wire so as to be entirely immersed in the solution and quite clean. Heated the end of the tube, where no air was, to boiling for several minutes and then left it still. At first there was no discoloration and the iron apprd. quite bright, but in a short time action appeared at one spot on the iron. Black sulphuret formed there in a little spot like rust, i.e. compact and limited, and from that place a little green cloud rose into the yellow solution. As the piece of wire had been touched by the Laboratory pincers, it is possible some irregularity there caused this action. At all events, the great difference between the action of the air in the tube before used, and the non action now air was excluded, was very evident. 5797. So Iron shews no effects in favour of a sensible thermo electric current. * [5796] 12 NOVR. 1839. 9 5798. Made experiments with red nitric acid in reference to possible thermo electric currents. But the results were here alto- gether new and of a different nature to those expected, and in reference to the thermo effects, I think their existence is very doubtful in any of the experiments made. 5799. Platina; Platina; Very Orange Nitric acid (5607, 541 1). A At the outset was Pos. 10° and continued so when at quite uniform temperature, as if there were some vigorous chemical On action going on here. heating either side, the Platina there became Negative 60° or 70° or more. The effect very striking but surely cannot be one of thermo electricity. 5800. The moment the spirit lamp was applied the needle moved, and I think almost before the Platina could have been sensibly warmed. 5801. Brought the whole to common temperatures and the needle was at 0°. Then dropped a little ether on the one side to cool it. That side quickly became Positive 15° or more: cooling the other side produced the corresponding effect. Must remember in rela- tion to these effects that the acid is an excellent conductor and so the effects are, as it were, magnified. 5802*. When with the same arrangement a joint of the copper and platina was heated, the deflection was not more than 5°, the heat being then far higher than before for the acid in the tube. The order of the current produced when any one of the spots a, c, d were heated is shewn by the arrows. 5803. When all was at a uniform temperature, I took out one wire, washed, dried and warmed it (not to 212°), and then put it into the acid; it immediately apprd. to be Neg., producing such a current. I then took it out, washed, dried and cooled it by a little ether, then quickly wiping and immersing it in the acid, it was Positive. 5804. Gold; Gold; Red Nitric acid. Repeated the same experiment, using gold wires. The effects were just the same and I think to as high a degree with this metal as the former. Either heating or cooling the liquid or heating or cooling the wires, the proper effects were produced. 5805. Palladium; Palladium; Red Nc. Acid. Palladium and the acid acts at common temperatures and a brown solution is formed lo 12 NOVR. 1839. A A at the wires. was Pos. Warmed side and whilst the chemical action was going on, it became Neg. This might be partly due to power like that of the Gold or Platina, and in part to the investing solution formed by chemical action. But it was striking. It at all events shews that the chemical power of Palladium (which was much increased by the heat) is not a strong current force. 5806. Would be well to make out what is that state of Platina, Gold and perhaps Palladium which makes their chemical actions differ so much as sources of voltaic power from the metals at the other end of the series, as Cadmium, tin, zinc, etc. etc. Remember the Gold on Nitro-muriatic acid (5595). 5807. Iron; Iron; Red Nitric acid. Violent momentary action when first put in and corresponding momentary effect on the needle. Then became inactive and quiescent to the usual degree. 5808. Heated B side; did not become Neg., but more Positive^ A cooled it and the effect went down. Heated and it in turn became very much Pos. So Iron is quite unlike Platina or Gold or the inactive metals in this acid. 5809. It would also appear that Iron and Platina, when in Nitric acid, are not quiescent because they are alike to each other, for they are evidently very different. Hence they should shew the effect of contact, if there is any such effect, where they touch each other. 5810. These effects of Platina, Gold, etc. seem almost to be effects produced by a power in these metals of altering the state of the acid around them (the Ns. acid, etc.), like that they have in mixed oxygen and hydrogen, and makes them sources of a current in the contrary direction to that produced by oxidizable metals —it is very wonderful. It is most probably related to the former effects, in which, though heating one of two platinas did nothing, yet heating the Platina of Silver and Platina ( ) made it Negative. Must examine these effects all together and bring them into proper relation. 581 1. Conceiving that the effect might be due to some peculiar action of the Nitrous acid in the Nitric acid, which though increased by heat, did not depend for its existence on it, but was in the fore- going cases balanced at the two extremities, I arranged the tube thus, making a colourless Nitric acid ; red nitric acid, and using 12 NOVR. 1839. II A Platina wires. Immediately B side became Pos. to side 70°. Changed the wires side for side, but still the B side was Pos. 70°. This therefore must involve the cause of the effect in the former cases, and now it appears to be independant of temperature, to be in the Nitrous acid in some way and to be an action making the Platina Pos. instead of Negative. So that in the former cases, the heat was perhaps only shewing an effect by diminishing the pre-existing action, not by causing a new one. But this, and the true locality of the action, I must search out carefully hereafter. 5812. Warmed A, or the colourless acid side, and it became More A Negative; cooled and its former degree of Negative state was reproduced. Warmed B or Red Nitric acid side and its state became less Positive, but the change was very little in degree compared to what it had been when both the Platinas were in Red acid (5799). In fact the difference of warm and cold in the same red acid is but little, compared to the difference of coloured and colourless acid to each other at common temperatures. 5813. Placed Gold Gold in the two acids of the tube (5804). A was Pos. at first for a moment, then B became spontaneously Pos. to about 25°. Took out, washed, wiped and changed the A wires : still that in was a trace Pos. at the first moment and then A B became Pos. about 25°. (Can at that first moment be taking up a state analogous to that taken by Iron at the first moment in Nitric acid.'^) So the permanent deflection, though in the same direction as with Platina, is nothing like so strong. Heating the A Gold in the or colourless acid side made it much more Negative, as before— cooling that side lowered it again. Heating B or the red acid side made it a Httle less Pos., but as with the Platina (5812), so here the quantity of effect was nothing like so much as that of the heat on A. 5814. The gold deepens in colour in the Nitric acid as if there had been some action upon it. 5815. Then put Gold Gold into Nitric acid, colourless on both sides, to compare the effect of heat on it now, when the metal being in the same acid was not Negative, At first the needle was quiet A and at 0°; then heat either at or B produced scarcely a sensible effect. So that in the above action the effect of heat on the metal in 12 12 NOVR. 1839. the colourless acid is still related to the metal in the red acid, i.e. to the Positive metal ( ), as in former cases. 5816. When all was at the same temperature and quiet, I dropped a little red Nitric acid in on B, and that instant the gold wire in that end became Pos. and strongly so for the moment. 5817. I examined the colourless and coloured Nitric acids for Muriatic acid, supposing a little of that might be present, and active as chlorine on the Gold or Platina. I found none in the colourless acid, but there was chlorine in the red acid. I therefore added some liquid Nitrous acid (from distilled nitrate of lead) to a portion of the pure colourless Nitric acid, so as to make a red acid which, when tested, was quite free from chlorine or muriatic acid. 5818. When this acid was used in place of the former coloured acid, either by itself with the platina wires (5799) or in conjunction with the colourless nitric acid (581 1), it gave exactly the same effects and as strongly as before. They do not depend therefore upon any chlorine or muriatic acid on the Platina or Gold. 5819. Tried an experiment or two with fuzed electrolytes to as- certain if correct results might be obtained with certainty from them. 5820. Platina; Platina; fu\ed Nitrate of Silver. This experiment was made in a bent tube. The actions at first rather strong and A irregular. After a while, was rather steadily more Pos. than B when the temperature was as nearly as might be alike. Then which- ever side was made distinctly hottest became a little Negative, but the effect went down much in degree whilst the heat was continued and sometimes even to the cooler side becoming the Negative. If A had been made Neg. by continued heat and had gone back in degree, then heating B made it, B, more Neg. by far than it would otherwise have been or than it could continue. This appears to be that common effect ( ) due to the forces accumulated by a certain current and coming into play in the contrary direction when that current is stopped. On 5821. the whole, the apparent effect of heat is to make the wire heated Negative; but that this is a pure thermo-electric effect is very doubtful- for it falls much, though the heat be continued, and on changing the heat to the other side the current produced is 12 NOVR. 1839. 13 over strong at the first. These effects are more accordant with chemical than thermo-actions. 5822. Is it possible that Platina can determine the separation of silver at the heated side and so become an active negative surface ( ) ? On taking out the platinas, they were well washed and wiped, and as they looked a little different from their first ap- pearance, bng. rather more rough, I examined them by a drop of nitric acid to find metallic silver upon them. There certainly was a trace of silver on them, especially on B side wire. But it would require polished wires and great care to decide the notion of reduction by the Platina. It is however worth deciding. 5823. Platina; Silver; Chloride of lead. Silver is Pos. to Platina On in fuzed chloride of lead. taking out the metals, there was no evident action, but quite enough to produce the effects would have been inappreciable. Such experiments with fuzed electrolytes are not easy to make without good glass, free from lead, and an especial heating apparatus. 5824. It appears to me that the experiments on Nitro muriatic acid (5593), Nitric acid (5410, 5606), nitrous acid, etc. etc., as well as the others on strong and dilute acids, hot and cold acids, etc., are well fitted to illustrate many points: as for instance, the locality of the generating force as at this or that surface; the manner in which heat affects chemical affinity in local actions; the constancy of chemical force (in certain of the cases of dilution) at a given temperature; the nature of compound and complicated electro- lytes, as Nitric acid, hydrated sulphuric acid, etc. etc. 5825. All those effects, so numerous, which relate to the state the metals take in consequence of a previous current; or of investment; and perhaps some of those where Platina, being Neg., becomes more Neg. by heat, should be collated and compared with Schon- bein's Polarity of liquids; Schonbein's indifference of Iron in N. Acid, etc.; De la Rive's peculiar state of Platina electrodes by current; Marianini's single pair producing same state; and putting Platina plates into oxygen and hydrogen; as well my perhaps as results with clean and dirty Platina in the sixth series. 14 12 NOVR. 1839. 5826. When going to work on Henry's late dynamic induction experiments, remember that the following parties have Henry's inductive coils— Gassiot; Leeson; Edwd. Solly; The Adelaide Gallery; Daniell. 14 NOVR. 1839. 5827. Repeated the expts. with Lead and copper (5774) but made an acid B weaker, thus A was Oil of vitriol. + B I vol. Oil Vit. 2 vols. Water CI ,, „ +20 „ „ A 5828. Lead in was well Pos. to that in B, and it was before Pos. C to that in (5712). But lead in B was well Neg. to that in C, and also to that in A. So that this B acid rendered the lead in it Nega- tive and well so to lead in acid of the same kind, either stronger or weaker (5774). A 5829. Copper in was a trace Neg. to copper in B, and it was in the former case (5710) a little Neg. to copper in C. Also copper in B is a little Pos. to copper in C, and a trace Pos. to copper in A. So copper in B is Positive to copper in stronger or weaker acid of the same kind (5775). A 5830. Whilst comparing the coppers in and B, I thought the small effect produce[d] might be an effect of dilution (5675). I therefore used Platina wires instead of coppers in the same acids, but did not obtain the same effects. So that I believe the result is not due to the mere dilution, but to the chemical relation. The result with lead is far too powerful to admit any thought of this kind. 5831. Made a comparison of different metals in three different nitric acids, to see how their relations ran and changed about. The table on the next page^ gives the tabulated results. The acids were A— strong Nitric acid. B— + I vol. strong acid i vol. water. C— I ?j 5? + 20 vols, water. ^ i.e. par. 5832. ^ u u uu uu uu u u uu ^ 8 •55 o 8 -^5^ U •o 2 oo 2o o. • bD C -i^i •t-j (V o ^ ^ -c^ O ^1 = OS ^ Oh ;-H 8 -So 8 0) TG 1—1 C/D 00 a; m 00 Dh a. £S Uo U U ^ o o ^ ^ K-l ^— CL> 72 «H-( (u ^ U ^ ^ P5^h Ph dilute M.A. Red Nc. acid Do. ~^ 5> >j » ~^ >> Sol. oxalic A. Sulrt. Potash Sol. Potash Do. Sulrt. Potash Red Nc. acid Platina dil. N.A.<- Silver-*- strong N.A. Copper-^ Sulrt. Potash Iron dil. N.Acid » >5 ?> » ~^ ?5 Strong N.A. sol. iodide Pm. sol. Acetate Lead sol. Potash dilute N.A. iodide potash J? acetate lead sol. ~^ ?5 >7 sol. potash full current full current feeble current full current good current good current good current feeble current feeble current, very small or weak current a current feeble current powerful current current feeble current doubtful very powerful fair current feeble current doubtful ^ In the manuscript, all the arrow heads are red, except that printed in heavy type in the twelfth line of the table. 14 NOVR. 1839. 17 Silver ?> Copper Silver Lead Zinc Sulrt. Potash Strong N.A. Sulrt. Potash Iron Platina dil. N.A. Iodide Pm. Potash dil. N.Acid Iodide Pm. Potash Tin -> Cadmium Lead -> -Copper diluted N.A. Platina Strong Pale N.A.Lead -> Iron -> Tin -Cadmium Zinc Strong current fair current Do. strong current good current not sensible not sensible not sensible feeble, but de- composes very feeble, not decompose not sensible very feeble, not decompose feeble current good from first moment good current moderate good current moderate In these six experiments, the ends in the dilute acid were first immersed, and then those in the strong acid; hence probably the reverse currents with copper and cadmium. Tin Strong S.A. Iron diluted S.A. strong current Lead copper Iron moderate current strong current copper feeble current 5834. It appears to me probable that any one of the very numerous combinations which can be made out of the following table would produce a current, and that some of them would be very powerful. Platina Gold Palladium Rhodium Silver r Nickel Copper Lead Si .0 Tin Zinc Cadmium F D IV S3 ^dilute Nitric acid „ sulphuric acid Muriatic acid Solutions of vegetable acids Iodide of Potassium Iodide of Zinc solution of salt many metallic solutions 2 i8 14 NOVR. 1839, Besides these cases there are also 5835. All the cases of currents with one metal and one acid heated at one of the junctions ( ). 5836. All those of one metal with the same acid, but strong at one end and diluted at the other. 5837. Perforce all those where, different metals being used, the effect of dilution or heat is to cause a current in the reverse direction to that which it is supposed the metallic contact ought to produce (5695). 5838. Arrange two plates of iron and Platina with two drops of strong nitric acid between them. There is no current. Then heat How one end until action is caused, and there is a current. can this depend upon contact } 5839. Or all being quiescent, dilute one drop a little and there is How action. can this be contact ? 5840. Batteries ofno metallic contact Iron or Iron dilute acid dilute acid Copper Silver Sulrt. Potash Sulrt. Potash Iron, etc. Iron, etc. or 5841 *. a is strong nitric or sulphuric acid, and h is dilute acid of the same kind; c are wires or plates of one metal only, being copper, iron, silver, tin, lead or any of those metals which become Pos. and Negative by difference of dilution of the acids. 5842. or Strong and weak solution of potash with zinc, lead, cadmium, etc. will do the same thing. 5843. If the fluid above be sulphuric acid and Iron the metal used, the arrangement will give a current in one direction— but if tin be the metal used the current will be the other way. 5844 1. or the following arrangement with two metals not touching. I, 3, 5 contains strong sulc. acid; 2, 4, 6 dilute sulc. acid; the metals a, a are tin, and ^, b iron. The current will be in the direction of the arrow. * [5841I t [5844] i6 NOVR. 1839. 19 5845. Repeated the experiments as to copper and Silver of (5832). Silver in N.Acid C was compared with copper in the acids C, B and A. When the copper was in C, there was scarcely a sensible current. When it was in B, the metals were as nearly alike as well could be: the silver was in the least degree positive and the copper in B acid was acted upon. When the copper was in acid A, or the strong, there was first of all a hesitation in the needle action and then the silver was positive. So that the results are but feeble and in the last point are not like the former ones : the feebleness of the differential action involves these variations. 5846. In reference to the same table of results (5832), ascertained the relative state of silver, copper, iron and lead to each other in + the N. A. B, i.e. i vol. acid i vol. water. The lead was Positive to Iron, copper and silver. The iron was Pos. to copper and silver, and the copper was pos. to Silver. 5847. Made this arrangement, in which two metals, iron and platina, are connected at two different places by the same strong nitric acid, but with the power of heating or diluting one place to observe the effect. The following is the practical arrangement. The two tubes contain the same strong pale nitric acid ; the wires />, p are platina; the wire i is iron, and a galvanometer is in the circuit. All was neutral at first. Heated the iron end, etc. at h. It is surprising what a little difference this made; the deflection was perhaps not more than 20° and yet the heat caused action to go on on the warmed iron and the metal dissolved there, though with difficulty. This is a very striking result in connection with the peculiar condition of iron in nitric acid. 5848. In this case it was not so much that the cold iron could not transmit the current produced (though it is known from the former expts. with thermo currents that iron is far worse than platina in this respect), as that the hot iron could not produce a strong current, its action being almost entirely local; and the following result shews that. 5849. The heated part b was cooled down— then a drop of water was put in there on to the nitric acid— at first there was no chemical action and no deflection of the needle. Moved the water and acid about with the end of the wire. In a few moments proper chemical action came on, the iron not evolving Nitrous acid, but nitrous gas apparently, or a lower grade of nitrogen compound with 20 i6 NOVR. 1839. oxygen; and now the acting iron was powerfully Pos., and the current ran round the circuit properly and in a manner very different to that of the former case by heat. 5850. As the heat rose from the action there were redder fumes evolved at the place of action in the tube, and I thought the deflection of the needle was then not so strong. The action was apparently more local. How 5851. are all these effects referred to the theory of contact.'^ 5852. Some more special results on voltaic currents without metallic contact- see (5833). Iron dil. N.A. Mur. acid sol. common salt common water Platina hydr. Nitrous acid most powerful current Do. Do. Powerful current In all these the current was excellent. The nitrous acid was that green hydrated acid before described ( ), and it seemed to make no difference to the Galvanometer needle whether the iron and platina terminations in it were in contact with each other, or connected only by the fluid. The results were most striking in relation to metallic contact. Iron ?> ?? Copper Tin Lead Cadmium .mc J? dil. N.A. Mur. acid dil. S.A. dil. N.A. Mur. A. dil. S.A. dil. N.A. Mur. A. dil. S.A. dil. N.A. Mur. A. dil. S.A. dil. N.A. Mur. A. dil. S.A. dil. N.A. Mur. A. dil. S.A. Platina sol. iodide Pm. small current' small current Do. Do. Do. not sensiblev/ small current Do. Do. Good current Good current small current good current good current not sensibley/ good current good current very feeble \/ All caused decomposition of the iodide evolving iodine against the Platina except Copper, Cadmium and Zinc in dil. S. A. i6 NOVR. 1839. 21 5853. Plumbago is often Neg. to Platina in solutions which would seem to have no power of acting on the platina to induce a current, and yet the plumbago would not seem capable by any action of its own, like that of deoxidizement (as with per oxides), to be an active source of what may be called a Negative current. Made some experiments to elucidate this point. 5854. Plumbago; Platina; Mur. Acid. The plumbago was one of On Morden's slices and was fresh: the Muriatic acid was pure. first immersion the Plumbago was strongly Neg.^ but quickly settling to plumbago a trace Negative and that constantly. 5855. Took the Plumbago which has been soaking in Muriatic acid (strong) for 7 weeks (5241)^; it had communicated iron to the acid and coloured it yellow; it had also become rotten and broke easily. This plumbago was also clearly Neg. to the platina, and when steady, to the amount of about 10°. 5856. Took another fresh piece of Plumbago, and it in Muriatic acid was clearly Neg. to platina. The acid quickly acted upon it and in a few minutes was yellow from iron dissolved. I boiled this plumbago in the acid for 5 minutes, then cooled it a little, but trying it whilst still warm, found it was powerfully Pos. to Platina —its state having been thus reversed. Cooled it quite and it was still Pos. a little to Platina. So there is no permanent effect to produce a neg. current here. 5857. Put the same piece of plumbago back into the acid and boiled : removed the acid and put fresh so as to exhaust all chemical action. Did this again and at last left the Plumbago in the acid on the sand bath for a day or two, to exhaust it of chemical power (p. 1326'). 5858. Plumbago; Platina; Nitric acid., i.e. i vol. strong N.A. + 4 vols, water. Fresh plumbago was Neg, to the platina— warmed it in the acid and it quickly became very Pos. to Platina— cooled it and it then was either 0° or the smallest possible trace Neg. to the Platina. Warmed it again and it was well Pos. again. 5859. The Plumbago and dilute N. Acid of (5241)^ the 23rd Septr. was put into a glass. The acid was colourless and the Plumbago was Neg. to Platina— warmed it a little and it was still Neg.— warmed it pretty well it was still a trace Neg., but not as before in degree— left it on the sand bath for i hour— now much iron ^ Par. 5893. ^ This should be 5214. 22 i6 NOVR. 1839. dissolved out and the acid, which was colourless, became yellow from iron. The plumbago was now well Pos. to the platina— being cooled, the plumbago was again well Neg. to the Platina. (This + acid was i vol. strg. N.A. 5 vols, water). 5860. Put this and the other plumbagos which had been used in the N. A. into a flask with fresh N. A., to heat on the sand bath and become exhausted of any chemical reagents (p. 1325^). 5861. Plumbago; Platina; Potassa. Plumbago has been soaking in solution of caustic potassa (5241^) since 23 Septr., or 54 days. It is clearly a little Neg. to Platina— about 3°. The plumbago does not seem altered and the solution is not coloured. I boiled the plumbago in the alkaline solution for 5' and now it was clearly Pos.; more than it was before Neg. 5862. Fresh plumbago in Potassa is clearly Neg. to Platina and then falls to about 5° Neg. Warmed it a little and the Neg. state of the plumbago zWr^czW— heated it more and now it became clearly Pos. to Platina, and when cold was still a slight degree Pos. This increased effect of Neg. on first heating is very interesting, but the whole result shews that Plumbago has no permanent power of producing a current by contact. 5863. Plumbago; Iron; Potassa. Fresh Plumbago is negative to Iron in Potassa, as it is to Platina, and perhaps rather more so than to Platina. This is natural enough, but the explication of its relation to Platina will probably shew its relation to iron. A 5864. Plumbago; Platina; Sulphuret ofPotassa, fresh piece of plumbago in strong solution of sulphuret of potassa (5263, 5268) was very Neg. to Platina— being heated for a minute or two in the solution, it was more Neg. to Platina than before- heated still longer and then the Plumbago became fairly Pos, to the platina. 5865. The piece of plumbago left in sol. sulrt. potassa (5241 since 23 Septr., or 54 days, was found in the least degree Pos, to platina, at once and at common temperatures. 5866. It is clear therefore the power that fresh plumbago has is not inherent but exhaustible, as a chemical power would be. 5867. Left the fresh piece, etc. in sulrt. potassa on the sand bath to take up a final state (p. 1333^). 5868. Per oxide of Manganese is Neg. to Platina in certain solu- ^ Par. 5888. - This should be 5214. 3 Par. 5931. i6 NOVR. 1839. 23 tions. Is this an effect of contact or of an exhaustible chemical action ? Per oxide of lead has the same power, as Schonbein and others have shewn. Made a few experiments with these oxides to illustrate this point, and first with per oxide of lead, being that prepared by well boiling in distilled water and found to be a con- ductor of a thermo-current ( ). 5869. Per ox. lead; Pladna; Mur. acid. Per oxide of lead very Neg. indeed; much chlorine evolved and chloride of lead formed. Is quite in accordance with theory of chemical action. 5870. Per oxide lead; Sol. Salt; Pladna. Per oxide when cold is a little Neg. to Platina— by heat is more and well so. Boiled them together for several minutes— was still so. Continued the boil- ing for some time and found that plenty of chloride of lead was formed. Hence sufficient reason for the action and state of the per oxide. 5871. Per oxide lead; Pladna; strong pure N.A. When cold the per oxide is Neg. to the Platina. Being heated in the acid, there seemed to be an action and the continued evolution of a little gas —then it was the least trace Neg.— heated it again and after some time still it was Negative, as if the generating action were still going on. A 5872. little per oxide of lead, being boiled in nitric acid, soon changed in colour and disappeared, leaving a grey insoluble sub- stance in its place: this substance dissolves in water, crystallizes from its Nc. acid solution, and is precipitated by Sul. Soda. It is doubtless Nitrate of Lead, and from its formation from Per oxide and Nitric acid arises the Neg. state of the Per oxide and the current produced. 5873. Per oxide lead; Platina; dilute S.A. The per oxide is Neg. fairly to the Platina— heated the acid and the per oxide was more Neg. than before. 5874. Imagng. that the addition of a little organic or reducing matter added to the electrolyte ought to increase the action on the per oxide of Lead, I added a little alcohol to the acid and thought that the per oxide was more Neg. than before. 5875. Per oxide lead; Platina; Sol. oxalic acid. Per oxide is most powerfully Neg. Strong action goes on, carbonic acid is formed and evolved and oxalate of prot oxide of lead is produced. If enough per oxide be triturated with the acid the latter becomes 24 i6 NOVR. 1839. entirely neutral. This and muriatic acid furnish I think good illustration of all the actions of per oxide of lead in producing currents and becoming Neg. to Platina. 5876. Per oxide oflead; Platina; Sol, Tartaric acid. The per oxide was Negative. There was action and the evolution of a little gas (carbonic acid) at the per oxide, but the effect was not to be com- A pared in strength with [that] of oxalic acid. little nitric acid was added : the action at the per oxide was increased, but the effect was local, for the needle was not more deflected. 5877. Per oxide oflead; Platina; sol, Sulrt, Potash, The per oxide was very Neg. to the Platina. The needle stood at 70° steady. Per oxide of lead in sol. sulrt. Potassa becomes Sulphuret of lead directly. So this deflection depends upon chemical action. 5878. Per oxide lead; Platina; sol, Potassa. The per oxide is Negative to the Platina— when made hot, very Neg., i.e. to 20° nearly. After boiling some time it was still Negative. A 5879. little of the Alkali, being neutralized after this boiling, gave full odour of chlorine, so that some chlorine compound is here present— but it should not alter the relation of the per oxide and potash. Per oxide of lead boiled in a solution of potash gives a solution of the protoxide in a short space of time. It is easily shewn by Hydro sulrt. ammonia. 5880. Now worked with per oxide of Manganese.^ using such as is employed for yielding oxygen. Employed fragments of it. 5881. Per ox. Mang.; Platina; Mur. Acid. Chlorine produced and the oxide fairly Negative, as was to be expected from theory of chemical action. 5882. Per ox. Mang.; Platina; Strong pure N.A. Per oxide Negative moderately. Added some water and alcohol (5874): violent action came on between the Alcohol and acid, but on adding more water, that went off except at the piece of oxide, which appeared to effervesce and was more Neg. than before. Per oxide boiled in Strong N. A. gives a solution of protoxide and of Iron. Dilute N. A. with a little alcohol and per oxide of Man- ganese, being put on the sand bath, in 5' all the per oxide was dissolved and proto nitrate formed. 5883. Per ox. Mang.; Platina; dil. S.A. The per oxide is in the least degree Neg. to Platina. In much stronger acid it is the same. i6 NOVR. 1839. 25 A little heat very much increases the effect. The reduction of Per oxide to prot oxide in Sulc. acid is one of the processes for ob- taining oxygen. 5884. Per oxide Mang.; Pladna; sol, oxalic acid. The per oxide is Negative but not strongly— only in small degree— heat increases the effect a little. The oxide is very compact and hard compared to the per oxide of lead (5875). 5885. Per oxide Mang.; Platina; Sulrt, Potash. The per oxide is slightly Negative to the Platina. 5886. Per oxide Mang.; Platina; Potash, Clean crystd. native per oxide Manganese was put away in pure sol. of caustic potassa (5214) on the 23 Septr. The solution was now coloured a bright yellow, as if it contained chromic acid or some such thing. The oxide was still and well Neg. to Platina— being heated in the Potash a few minutes it became more Neg.— boiled it well, still more Neg. Put in on the sand bath to boil some time. The solu- tion became deeper and deeper in colour and the per oxide was still Neg. 20° perhaps. It looks as if it were really an active chemical agent continuing to give up something (oxygen) under the in- fluence of the potash and not easily exhausted of its power. On 5887. Put it to boil again. the 21 Novr. was still Neg., but in a very small degree, to the Platina (5984). 18 NOVR. 1839. 5888. Plumbago; Platina and N, Acid, The plumbago in N. A. of the 1 6th (5860) is still a little Neg. to Platina- most at first and soon falls to very little. Warming the Plumbago in the N. A. made it more A^^^.— hotter was more Neg.— hotter still, still was more Neg. 5889. The N.A. had taken a little iron from the plumbago; but now put the plumbago into a flask and heated it in N. M. Acid for 15'. Turned the two pieces into water; took one out, washed, wiped and put it into strong M.A. It was powerfully Neg. to Platina. The other piece in the water was a very little Neg. to Platina— put a little acid in and it shewed more Negative— diluted the acid in which the first piece was and that became less Neg. than before. Put both pieces into the same acid and water. Each was 26 i8 NOVR. 1839. then Neg. to the Platina wire, but one of them was powerfully Pos. to the other, the large being positive to the smaller. 5890. Broke a small piece off from each. The piece from the Positive was pos. to the piece from the Neg. Broke two fragments off from one piece and found one well Pos. to the other. So soon is this difference of Pos. and Neg. established and so easily is it shewn by plumbago in such an acid. Can the difference depend upon the solution in the pores of the plumbago ? 5891. The two pieces of plumbago in the dilute Mur. acid, when lying one on the other and communicated through the galvano- meter by the platina wires, gave a current, though one was pressed against the other with considerable force. So that such contact does not communicate the acting parts with each other so well as the wire of the Galvanometer. 5892. I put both pieces into pure strong M. A. in a flask on the sand bath, that the fluid within might be displaced by the simple Mur. acid, and so left them (5906). 5893. Plumbago; Platina; Mur. acid. The plumbago in Mur. acid of 16 Novr. (5857) was found at first contact (all being cold) a little Pos. to Platina, but the effect quickly sank to 0°, and then Pos. or Neg. state was not sensible either way. Washed, wiped and put the plumbago into fresh Mur. acid, but there was no action to platina. The needle was at 0°. 5894. This neutral plumbago, which has been 7 weeks or more in M.A., was put into distilled water and boiled. The water was removed and fresh put on, and this was repeated two or three times at intervals. The next day, the 19th Nov., this plumbago was the slightest degree Pos. in distilled water to Platina— in cold common water it was scarcely sensibly Pos. to the Platina plate —but in cold strong solution of pure Mur. acid it was freely and permanently pos. to the platina plate about 20°. In this case the Pos. Plumbago would have a weaker solution within it than that against the Neg. Platina. 5895. Put the plumbago and Mur. acid into a flask on the sand bath to see if it would become neutral in Muriatic acid. After an hour it was cooled and was still Pos. to the platina plate— much at the first introduction of the plate and less afterwards— scarcely On at all after a while. taking the platina out, washing, wiping i8 NOVR. 1839. 27 and reintroducing it, it was again more Neg., and fell by degrees almost directly as before; but this effect on the platina was by no means so great in this case, where the plumbago is Pos., as it was found to be in another case where the plumbago was Neg., to be described hereafter (5909). 5896. Took some fresh slices of plumbago, placed them in a crucible and gradually raised their temperature to full redness. Kept them so for a time and then allowed them to cool. A 5897. piece of the heated plumbago (5896) in strong pure Mur, acid was neutral to Platina. Another piece of the same heated batch was also neutral in the same acid to the Platina. But when the two pieces were compared together, one was very strongly Positive to the other. 5898. I put the two pieces into a dilute solution of Mur. acidthere was no sensible difference between either and platina, but the one which was Pos. before was also well Pos. again when compared to the other. The effect is very striking. 5899. Took a piece of the heated plumbago (5896) and broke it in two. These were put at once into the same Muriatic acid in the same dish. Both were a little Negative to Platina wires, but fearing wire had hardly surface enough, I compared them to a platina plate of surface equal in the immersed part to their own. One was trace Pos. to the Platina plate. One was very Pos, to the other very Neg. Removed the Neg. Piece— then the Pos. piece was scarcely sensibly Pos. to the Platina plate. Put in a little piece of the Neg. one not nearly equal in surface to the platina plate and then the Pos. plumbago was well Pos. to it. 5900. So that with equal surfaces, or still more unfavourable proportions, the Plumbago could discharge the current of the Pos. Piece when the platina plate could not. 5901. Then on a sudden the little piece (just now neg.) became Pos. to the larger and formerly Pos. piece. 5902. Put another Httle piece of the first Neg. one in and it was neg. to the former little piece. So that even small pieces of the same portion of plumbago differ from each other more than plumbago does from Platina, and also change in their state in a very complicated way. 5903. All these effects were in pretty strong pure sol. of Muriatic 28 i8 NOVR. 1839. acid, which both conducted well and acted quickly on the iron in the plumbago. 5904. After an hour, looked at these pieces in the M.A. They were a little and variously related to the Platina wire— but were strongly contrasted one with another, and more than before. They seem by rest to store up power, as if the platina wire by pressure broke through external films, and related internal films to each other, which could not otherwise be in communication— but then the current, which is strong at first, always goes down in strength, as if platina took up an opposing state. 5905. Still the plumbago present no constant permanent effect of contact, 20 NOVR. 1839. 5906. Worked a while this morng. with the pieces of plumbago which were left to digest in Mur. acid (5892) on the i8th instant. The acid now shewed scarcely a trace of iron, but it contained fine acicular crystals of chloride of lead derived from the glass of the flask. 5907. They were transferred into fresh strong M. A., and being cold, were very nearly indifferent to each other; but both were well Neg. to a clean Platina plate at the first moment, and then the effect went off until the plate was removed, rewashed and restored. 5908. When Plumbago and platina were connected in the acid and 0°, if the platina plate were more depressed^ then plumbago again was Neg. to it, which effect went off as before. 5909. If the neutral platina plate were taken out of the acid, washed, wiped on clean cloth and restored again, the plumbago was neg. though it remained unaltered in every circumstance in the mean time. 5910. If the neutral platina plate were washed and cleaned by sand paper^ still it became Pos. when first restored to the acid in connection with the plumbago. 591 1. It seems as if the platina had a power of producing a current by some change it suffered in the M. A., but that change (on its surface effected)^, its power ceased, for the change could go no further. I The sense seems to require the closing bracket after "surface." 20 NOVR. 1839. 29 5912. If the clean platina were put into the mur. acid, not touching or connected with the plumbago, it did not lose this power, for at the moment of touching the plumbago the momentary current was produced. 5913. If the clean platina in the Mur. acid were touched by other platina (as the wire of the galvanometer by which connection with the plumbago was made), still no current was produced, nor could its state of tension in the M. A. be discharged, for the moment the plumbago was brought into the circuit the current was produced and the Platina was Pos. 5914. So that the exciting force appears certainly to be in the platina, and the current is produced as the platina acquires a certain state, analogous perhaps to that of Iron in Nitric acid. The plumbago appears to be first a body having no chemical tendency for the chlorine of the M. A., and next a most excellent discharge; and so shews the production of this state and the current produced during its development. 5915. That Platina cannot discharge Platina of its first tense state, though Plumbago can, is quite consistent; their opposite tensions are equal. 5916. The current only can bring on the inactive state of the Platina. A 5917. platina wire which had been discharged by the Plumbago in the Mur. acid, and so made to assume its inactive state, was very slightly Neg. to fresh undischarged Platina, as if the latter had now its state of tension in excess; but the effect was not to be compared to that of plumbago in facilitating the current. 5918. Plumbago seems to exert no action and to have no opposed final state. It appears to offer an open door to the electrolyte for the transfer of any current, determined elsewhere, which tends to pass through it. 5920 \ If the cause of the current is in the mutual action of the Mur. A. and Platina, which is enough to bring on a certain state in the Platina (the first step as it were of chemical action), then Gold in M. A. ought to be more powerful in relation to Plumbago than Platina has been; and yet should shew but little Pos. Power to Platina, because of its tension when undischarged and its ^ 5919 is omitted in the MS. 30 20 NOVR. 1839. imperfect power of communicating the current force to the electrolyte. 5921. So Gold and Platina in the M.A. in which the plumbago was. The Gold was just sensibly Pos. to the platina. But the Gold and Plumbago : now there was a strong current, and the Gold was far more Pos. to the Plumbago than the Platina had been; also its excess of power over the Platina was far greater than its effect when compared directly with Platina itself. 5922. If Plumbago is quite indifferent as a motor and only power- ful as a discharger, then it ought to surpass also in discharging current in the reverse direction, i.e. where it would be in the relation of Positive. Put a little piece of per oxide of lead in dil. S. A. and made platina and plumbago in turn the opposite termination in the acid. There was good current in both cases, the platina or plumbago being Pos.— but I am persuaded the plumbago was best. It requires a more delicate current than the per oxide of lead gave. 5923. Plumbago will probably be an excellent developer of the state of Electro motive bodies. 5924. Plumbago and Water. Two pieces (different) of the heated plumbago (5896) were put into a basin with common water. The small piece was Neg. to Platina plate and the large piece was still more Neg. to the Platina plate. But when the pieces were compared together, there was far more effect than between the platina and the most Negative piece^ the small piece being now Pos. and the large piece Neg. to each other. So common water shews the effect of plumbago. 5925. Two pieces of plumbago heated (5896), broken from the same piece, were put into distilled water, and being compared, there was very little difference or current sensible. Added a few Now drops of nitrate of soda and stirred all up. when they were compared, one was strongly Neg. and the other Pos., the needle standing at 30° or more. Both were the least trace Neg. to the Platina Plate. The effect was very striking and shewed clearly that the effect was not one of mere extent of external surface, for the platina plate was more than equal to them in its immersed part. 5926. Warmed the water and pieces and then they were in the least degree Pos. to the Platina plate, and differed very little from each other. 20 NOVR. 1839. 31 5927. When warmth is applied, must be careful about the difference of temperature in different parts ; for Plumbago may be able to shew thermo electric effects though the metals hardly can. 5928. Into a basin containing warm water and platina, put a cold platina, and the latter appeared to be slightly Neg. to the former (5927); but must work this out carefully with Plumbago. 5929. Plumbago appears to act as a most ready discharger of currents to an electrolyte. But in relation to the generation of the current, Carbon might take oxygen and hydrogen easily, and if it can, then plumbago may help to generate either a Pos. or a Neg. state according to circumstances. 5930. Make a Ritter's secondary battery with Plumbago and see if it has any reaction. 5931. Plumbago and Sulrt. Potash, The Plumbago of 16 Novr. (5867) was examined in its solution and found to be still a little Pos. to the Platina on the morng. of the i8th. It was again examd. this morng., having remained on the sand bath all the while since in its solution: still well Pos. I put a platina wire round it and returned it in its solution to the sand bath for a few days (5951). 5932. It occurs to me that perhaps the plumbago, etc. were not all at one temperature, and that the difference may be due to a difference of this kind in the parts concerned. If plumbago can shew the thermo electric effect of electrolyte, then it is very possible some effect of this kind may be here. Especially as the Plumbago is Pos, A 5933. Piece of the heated plumbago (5896) was broken into two pieces and put at the same moment into a solution of sulrt. potassa. They were neutral to Platina— then in a few minutes one was Pos. to Platina, and whilst contact continued was the next moment Neg. to Platina. The other was Neg. a little to the Platina —but both gave an unsteady needle as if changes were going on within. When compared with each other, one was Pos. and the other Neg., with a much greater deflection of the needle than either had shewn to the Platina. 5934. It appears that either this sulphuretted potash solution or an acid solution (both being electrolytes which act on the 32 20 NOVR. 1839. plumbago) develope these differences of piece with piece far more than Potash does. See below (5935). A 5935. Heated Plumbago; Platina; Potash, piece of the heated plumbago ( ) was broken in two and put into a solution of potash (caustic); both were a little Neg. to Platina, either the wire or plate— only a little. When compared together, one was a little Pos., the other Neg.; but the difference was no ways com- parable to the effect in acids or sulrt. potash, for now it was not greater than between either plumbago and the platina. 5936. Hence the effects in the acids, etc. may be effects due to the acid strata in the plumbago, to which the plumbago serves as a conductor, and not to the direct activity of the Plumbago as an electromotor. 5937. The theory of contact is not sustained by Plumbago in Potash, for one is Pos. and the other Neg. Contact, if any thing at all, should give consistent results. 5938. In illustration that reason why Plumbago will seem Neg. to a metal when Platina will not, is because it can discharge a current better, compared both with Silver in dilute Sulc. Acid; and found that silver appears very little Pos. to Platina in such acid, but very Pos. to Plumbago in the same acid, though Plat, and Plumbago differ very little in that acid. A 5939. Plumbago; Platina; N. A. piece of the heated plumbago (5896), was broken in two and put at once into strong Nitric acid —both were Pos, directly to Platina wire and plate. Still one was much more Pos. to the other than it had been to the platina plate of equal surface. So here the Plumbago had the Pos.^ not the Neg. state, and the two pieces differed at once from each other. Against contact theory. 5940. Plumbago in water. Have boiled some slices of plumbago (Morden's) in common water in a flask (containing also a platina wire to assist discharge and assumption of a final state) for several hours. Then left all to become cold together. 5941. Being turned out into a Wedgwood's basin with the water, all were found a little Negative to a fresh platina plate (i.e., one just cleaned (5909)) ; but eflect with all went down to 0° or nearly so. 5942. There were slight variations when the pieces were com- pared together, but not much. 20 NOVR. 1839. 33 5943. Took one of the pieces, broke it into two, put the two into strong Muriatic acid. There was very Httle difference between them, but both were Neg. to the Platina plate, and though more On at first than afterwards, still a little Neg. permanently. washing and wiping the platina, it became each time more Pos. to the plumbago. A 5944. third piece of plumbago from the water, on its first introduction into the Mur. acid, was well Pos. to the other two: it then shortly changed and became Neg, to them, and was then a little Neg., though scarcely sensibly so, to the platina plate. Then pieces i and 2 were powerfully Pos. to piece 3, though they On and 3 also were nearly insensible to the Platina plate. every repetition, i and 2 were very Pos. to 3, holding the Galvanometer needle nearly steady at 60° or 70°. i and 2 were also one Pos. to the other now a little, and all were acted upon by the acid, the iron appearing in the solution. 5945. The most Pos. of them was not sensibly less Neg. to the Platina plate than the most Negative one was. 5946. There was a storing up and an exhaustion of power even among the plumbagoes themselves, i.e. the first contact of i and 3 was more powerful than the continued contact. Still, a good current of 70° or so could be kept up for a minute or more and the effect probably depended upon the partial exhaustion of the film of acting fluid. 5947. Plumbago which is most Neg. to other plumbago is probably that which is best fitted to shew the effect of discharge with Platina, etc. etc., and is probably the most neutral and inactive, i.e., if the solution within its pores is the same as that without. Otherwise, that solution may be actively Neg. to the solution around. 5948. Had some plumbagoes in two different muriatic acids which, though strong, differed in strength, and one had more iron A in it than the other. piece in its acid was 0° to Platina plate. Another piece in the other acid was also 0° to the same platina plate. Put the first piece out of its acid into the other acid, and it was then well Pos. to the Platina plate and very Pos. to the Plumbago there. 5949. So that differences in the solutions are shewn to be very important. FDIV 3 34 2o NOVR. 1839. 5950. In a little while the piece which had been so Pos. at first became Neg. to the other piece, to which it had been Pos. and well so— and this second piece was rendered Negative to a piece to which it had before been neutral. 21 NOVR. 1839. 5951. The Plumbago, Platina and Sulrt. Potash of 16 Novr. (5867, 5931) being examined, the Plumbago was still Pos. to the Platina, as if conversion of the iron in it to sulphuret was still going on. 5952. Decomposed a pretty strong solution of sulphuret of potassa by a small battery of 30 Pr. one inch plates, feebly charged, using Platina electrodes. Sulphur was evolved at the anode, and a little gas, probably hydrogen, at the cathode. The sulphur was very pure and, in appearance, abundant. 5953' Will Platina and sulphuret of Potassa form a secondary A battery.^ Yes, for two platina wires, and B, were put into a A solution of the sulphuret; was connected with the Positive and B with the Negative end of the small battery (5952) for about 10 seconds; they were then separated from the battery and con- nected with the wires of the galvanometer. Immediately a con- siderable current was produced, and that wire which had been made by the battery Positive to the other was, when connected by the galvanometer, Neg. to the other through the solution, and that constantly so, whichever way the wires were used. The expt. was very Good and striking. 5954. Iron and Nickel are Negative to Platina in solution of Sulphuret of Potassa (5269). What is the cause of this: contact or chemical action.^ (5980). 5955. Iron and Platina come ultimately to a state of rest ( ). 5956. Iron; Platina; and sol. Sulrt. Potash (5269). Used plates of the metals. The iron being ever so well cleaned by sand paper and clean cloths, etc., was always Neg. to the Platina on the first immersion, but when well cleaned soon went down and became 0°. 5957. An iron plate which was dim from rust formed by vapours of the laboratory in the course of the last day or two, was far more powerfully Neg. to the Platina, and the base of rust became a haze of black sulphuret of Iron. 21 NOVR. 1839. 35 5958. I believe that the effect of clean iron is due to a film of oxide, and in this respect consider that iron is an exceedingly oxidizable metal. When in state of sponge, it fires spontaneously; with water and platina it gives an instant voltaic current— when out of sul- phuret, if washed in water and wiped, it gives instant smell of hydrogen— if cleaned with sand paper and dipped in water, it gives instant smell of hydrogen; so that when exposed clean to the air, it is not wonderful if a film of oxide instantly forms on it; and it is probably the reduction of this oxide at the moment of contact with the sulphuret which renders the iron Negative. 5959. With respect to the power of Plumbago to conduct Neg. or reverse currents ( ), a clean iron wire in sulrt. potash was not sensibly more Neg. to Plumbago than to Platina— but the piece of plumbago was found to be Pos. to the Platina in the solution. Using the Ironplate^ it was very sensible to the plumbago, but not in any very particular degree, as contrasted with Platina. 5960. Nickel; Platina; Sulrt, Potassa (5983). The Nickel was at first well Neg, to the Platina— it then fell very quickly to 2°. Then rewashed, wiped, etc., and was now Neg. only a little. Again Now washed, wiped, etc., and it was now less Neg. than last time. cleaned with sand paper and wiped dry (i.e. not having been wetted) : it was more Neg. than last time, but not so much as the first. 5960I. This piece ofNickel is a small bar. It is not perfectly solid, but has little rough depressions and holes which I cannot file out, and these contain oxide, etc. etc., black matter from former experiments, etc. These probably are quite enough to produce all the effect by the oxide, etc. which they contain. 5961. Must remember that a plate in sulphuret, if taken out and exposed to air, is Neg. when put in again (5127, 5177, 5325). So these holes, when the matter in them is exposed to air, would by that exposure be Neg. to the platina when reimmersed. 5962. Some results with Plumbago, Platina, etc., using the Green hydrated Ns. acid as the Electrolyte (5353). 5963. Platina; Platina; Green Ns. Acid, It is impossible to clean two platina wires so well and so alike, though both were from the same piece of wire, as that they shall be 0° when first immersed; though all care be taken to immerse them together. The effect first 3-2 36 21 NOVR. 1839. produced is quickly at 0°, a balance of the conditions of each wire having taken place. 5964. If, when the two wires give 0° effect, one be raised not more than the | of an inch and instantly lowered again, so that the part which had been for a moment in the air should be again reimmersed, that wire became strongly Neg. in consequence of the change of the acid on its surface ( ); so delicate is this electrolyte, in consequence of its good conducting power, etc. etc., in allowing currents to form and pass. 5965. Plumbago; Pladna; Green Ns, Acid (5353). Made the circuit complete, dipping only one end of the piece of plumbago used into the acid. The plumbago was strongly Negative to the Platina— the force gradually diminished, but after some time 20° or 30° of deflection remained. 5966. Put the piece of plumbago quite into the acid, so as to cover it over and exclude the effect of air at the part where it passed the surface of the acid; and now touched it with one or other of the platina wires employed, so as to make it the electrode at one or other side of the surface at pleasure, the untouching platina being of course the other electrode. Now the plumbago was strongly Pos. to the Platina, having changed, as has happened before with the Nitric acid and the Muriatic acid. It continued Positive for all the succeeding experiments. 5967. That platina wire which has for a moment touched the Pos. Plumbago, when removed from it and placed by the side of the other wire (still in the acid), is strongly Neg. to the other, no A A matter which it was. Suppose the wires to be and B: is touching the Pos. Plumbago and B is consequently Neg., for it is A the other electrode. Then removing from the plumbago and putting it by B, B becomes for the moment powerfully Pos. and A is as powerfully Neg. All this depended simply on the circum- stance that, to put a wire on the plumbago, it was raised a little, and when removed from it, it was depressed and so became Neg. (5964). 5968. Per oxide Manganese; Platina; Green Ns. Acid (5353). The per oxide is Neg. to the Platina. The manganese was disintegrated, and being left in the acid for a while, it converted it into Nitric acid, being itself reduced in its degree of oxidation; and then Iron 21 NOVR. 1839. 37 and protoxide of manganese were found dissolved in the acid. All very consistent with chemical theory. 5969. Per oxide Manganese ; Plumbago; Potash. The old preparation of Per oxide and potash was taken (5887) at common temperatures, and a piece of plumbago put into the solution. The Plumbago was well Neg. to Platina. The per oxide was a little Neg. to Platina. The Plumbago was very Neg. to the Per oxide, which was Positive. 5970. After a little while, all three became neutral to each other. 5971. After a while longer, the Plumbago was 0° to the Platina plate and was then well Positive to the per oxide (as if it did conduct a reverse current well); and this relative state continued. 5972. These changes are very remarkable, and it seems as if the previous currents had some strong effect in causing alterations in the relative states of the bodies. Must try this. 5973. Plumbago; Platina; Sol. Sulc. Acid and chromate potash. The plumbago was Neg. to the platina, and more if the solution was warm, but there was no remarkable or special action. Not more than in sol. of the acid or salt alone. 5974. Can such solid bodies as per oxide lead, per ox. manganese or sulphurets of some of the metals, be used in place of the electro- lytes in constructing an active circuit? 5975*. Zinc; Platina; Per oxide of lead ( ). The per oxide was that which had been well washed, etc. When all were pressed together there was a current. Zinc being Pos. to the Platina. When left quiet, the current fell. When the zinc was moved so as to produce a grinding action, pressure being continued on a, then the current was renewed, but it could not be continued. On removing the plates, etc., the per oxide was found adhering con- siderably to the zinc, and the latter seemed a little tarnished. 5976. Lead wdiS a little pos., not much, and was tarnished at place of action. Cadmium— not sensible effect. Iron— not sensible. Zmc- produced the effect again. Zinc with Powdered oxide of Manganese : no effect. 5977. As a piece of wet paper between zinc and platina produced * [5975] 38 21 NOVR. 1839. a powerful current, proceeded to dry the oxide in a basin. Found some moisture came off. Then put it hot between Zinc and platina. Now very little deflection and that as if Zinc were Neg. 5978. Put zinc and Plat, together and warmed place of contact: is as if Zinc Neg. 5979. So no exciting action of the per oxide of lead and no evi- dence it can act as an exciting electrolyte. Its neutrality when in such contact is surely against contact alone doing any thing. 22 NOVEMBER 1839. 5980. Procured a clean piece of iron (5958), rubbed it over with sol. sulphuret of Potassa and put it into a quantity of solution so as entirely to cover it. It was now Pos. to a platina plate feebly. Whilst the contact was complete, rubbed it with a piece of wood under the surface, so as to expose a fresh surface to the solution. This did not make it neg., but the smallest degree more Pos. So that iron is not Neg. to sulphuret potash and Platina, except because of previous action of air, etc. A 5981. bar just clean from the Instrument Makers was actually Pos. in the first instance, but by soiling its surface and tending to oxidize it, could make it give a first effect of Neg. 5982. Different Pieces of iron differ, some inclining more than others to be Neg. at the first immersion; but none continue neg. if entirely under the solution, though rubbed. 5983. Nickel and Platina in Sulrt. Potash (5960). Put the nickel under the solution; rubbed it there. It was Pos. to the Platina— On whilst in contact rubbed it; it was still Pos. a little. taking it out it was found a little tarnished, as if sulphuret had formed on it. Then dipped it in water, washed, wiped and restored it, but now it was, as at first, strongly Neg. to the Platina. So that is evidently the consequence of a state taken by washing and exposure to air, and not a state due to contact with the sulphuret. So that difficulty is cleared away. 5984. Potash; Platina; and Per ox. Manganese. Reverted to the experiments and preparation of (5887). The potash was yellow. Examined it for Manganese: could not find any; only a little silica or alumina derived from the ore or the vessel. 5985. The per oxide has changed colour on the exterior, and 22 NOVEMBER 1839. 39 when a portion of it was washed and dried, it had a greenish surface, and is evidently tarnished as ifit had suffered from chemical action; but what is that action? Can it be partial reduction by some organic matter in the potash ? 5986. Added a little alcohol to the solution of potassa and now the Per oxide was well Negative to the Platina plate, continuing even at 30° or 40°. So this is the explanation of the whole, and what the alcohol does here, the dust and other reducing matter which falls into the potash does there. 5987. The well crystallized Manganese shews a much better effect than the permeable amorphous kind. The dynamic action is probably better developed and determined. 5988. Plumbago; Platina; Rhodium; and Mur. Acid. Expected that Rhodium would have less action than Platina on Mur. acid, and so might prove neutral to Plumbago, which might be Neg. to A platina. piece of plumbago which had been digesting several days in Muriatic acid was taken and cooled with a little of its acid, the last being used as the electrolyte. Plumbago was Pos. well to Platina at the first contact, and the effect then went down nearly to 0°. Upon breaking and remaking contact, there was very little difference, but the Plumbago the least Pos. 5989. The Plumbago was still more Pos. to the Rhodium at first contact, but that went down nearly to 0°, and then on breaking and remaking contact the Plumbago was only a very little Pos. to the Rhodium, but more than to the Platina under the same circumstance. 5990. The Platina was a trace Pos. to the Rhodium. The effect of Fresh Platina to Rhodium is as nothing compared to the effect of either fresh Platina or Rhodium to the Plumbago. 5991. Hence it appears that Rhodium is more Neg. than Platina; but that it probably produces a first pos, effect, since it takes up the second or inert state. Must try this with Platina, which is Neg. to it instead of positive to it. 5992. The Piece of plumbago was turned into common water, and in it, it was a little Pos. to Platina in the water. Here the Plumbago was wet with strong acid and the platina with water, but the water is a bad conductor. The Plumbago was washed and wiped dry by a cloth; then put into the acid again, and now it was 40 22 NOVEMBER 1839. well Neg. to the Platina, so that the current was reversed with the states of the Platina and plumbago. The effect gradually went down, and in 5 ^ the Plumbago was at 0°, or rather a trace Positive to the Platina. 5993. Conduction without decomposition of the electrolyte. Try to decompose by Gassiot's dry piles; if can not, then perhaps obtain proof of conduction without decomposition. 5994. Voltaic pile : reference to reservations, etc. Exp. Researches. Paragraphs: see note to 856; 875,6, 7; note to 921; to 928; end of 960. Opinion reserved on source of V. Electy. at 857, 872 par. ^ Presumably 5 minutes. 10 JANUARY 1840. 5995. The current of a single pr. antimony-bismuth element passed across five interruptions of strong solution of sulphuret of potassium ( ), the interruptions being by platina plates. The immersed part of the plates was about i| inches by | wide on an average. 5996. The same thermo current passed 5 similar interruptions made by platina across the dilute green nitric acid ( ) from Grove's experiments, and so freely that I think it would have passed 20 almost as well, and probably a hundred sensibly. 5997. Platina and Iron associated in dilute sol. sulphuret potas- sium, I vol. Strong sol. ( ) and 10 vols, water. Iron was at first Negative but in ten minutes it was quite neutral. 5998. Platina and Iron were associated in the hydrated green nitrous acid ( ) and were soon neutral. The thermo current excited at x travelled round the circuit moderately, but nothing like as if, for iron, oxide of iron or platina had been used. The chemical action of a little dilute Sulc. acid at x caused a current which passed powerfully. 5999. Oxidized iron plate and platina in red nitric acid strong. The acid contained some muriatic acid and was in so far bad. The iron was a clean wire, bent thus, and then blued in the spirit lamp flame, as Schonbein directs. Whilst this with Platina was in the acid, the iron ox. was Pos. at first 60°, but soon fell to 25°, and in 15' was 20°, but part of this is due to the opposing state the platina takes up. 6000. When at 20°, the antimony-bismuth thermo current could pass well, raising the needle to 50°. This would not have happened had it been metallic iron instead of oxide. The chemical action of dil. S. A. at x gave an excellent current, passing in the opposite direction to the one tending to form in the nitric acid. 6001. Made the same arrangement, but used a dilute acid consisting of I vol. of above and i vol. water. The oxidized Iron was Pos. and rather strongly so, also irregular, the needle vibrating much. On examg. the iron, found that the oxide was dissolved away at the part cut by the surface of the acid at cz, and metallic 42 lo JANUARY 1840. iron was acting there. Cut ofF this part down to oxidized the termination by heat, attached the scaled piece to a platina wire Now and replaced it, putting the whole under the surface. there was nothing like so much current, but the scale iron was still Positive and the needle at 20° steady in less than a minute. 6002. The thermo current at x passed very well indeed. 6003. The chemical current of iron and dil. S.A. at x passed powerfully. 6004. In 40' the deflection was only 5° and steady and the thermo current was well conducted. 6005. I put a platina plate down into the acid before and against (i.e. in contact with) the platina plate to discharge its state; this caused the iron to seem negative for a moment, but I expect it also discharged the first platina plate of its state. Then taking out the second plate, the scale iron was pos. about 10° only. 6006. Boiled dilute solution of potash caustic, then cooled it. Iron and platina in it. At first immersion, iron was Positive to the platina, but soon fell much and nearly to 0°. 6007. Chemical action at x passed feebly— the action between the Iron and platina being by dilute Nitric acid. 6008. The following pairs of metals were arranged with dilute solution of potash at one end and dilute Nitric acid at the other, to ascertain whether the current it formed would pass the potash. 6009. Iron and silver: current did not pass sensibly. Iron and Palladium 1 Iron and Gold TIron andJ rP>li at• ma 1 ^ 1 r 11 the current p^assed feebly. Iron and scale oxide Iron 6010. Platina and rusty iron in dilute sol. of sulphuret of potas- sium. The Rusty iron strongly Negative. 601 1. Oxide of Iron made into a paste with water and applied on a platina plate— dried well— then this made into a pair with clean platina in dil. sulphuret of potassm. : it was well negative. 6012. An Iron wire dipped into a solution of dilute nitric acid- shaken— left a few seconds— put into vapour of ammonia to throw down oxide— warmed to dry it— then washed well in water— dried again. This was well neg. to Platina in sol. sulrt. potassium. 6013. Fuzed pure protoxide oflead ground fine, made into a paste 10 JANUARY 1840. 43 with water and applied uniformly on a platina plate, then dried by warmth and cooled. This with a platina plate into dil. sulrt. of potassium was well Negative, and the deflection increased as the coat was penetrated and the action went on nearer the substratum of platina. If left long enough the action came to an end. 6014. The cleaned platina plate was quite neutral to the other platina. 6015. The investing coat must be put on uniformly so as to cover the whole of the immersed part. If it be dabbed on so as [to] make little spots, one part of the platina being clear and the neighbouring part covered, then there is local action and no great current. 6016. When a platina plate similarly covered was connected with a clean platina in dilute Nitric acid, the first was strongly positive to the second, not negative, and no oxide was left. 6017. When a platina plate covered with red lead was in dil. N. A. in connection with a clean platina plate, the red lead was strongly Negative, as in Sulphuret of potassm. because it was now giving off oxygen. When in Sulrt. potassm. the red lead was Neg. and became sulphuret of lead. 6018. When a platina plate with per ox. lead was in dil. N. A. as before, the Per oxide was very Negative. In dilute solution of sulphuret of potassium it was also Negative. 6019. Carbonate of lead on platina plate to clean plat, plate in dil. sol. sulrt. potassm. Carbonate was Negative and deflection increased as above for a while. 6020. Carbonate of lead on platina plate in dil. N. Acid was Posi- tive to clean platina plate. 6021. Per oxide of lead in solution of caustic potash was very negative at first to clean platina— the force of the current quickly fell, and then adding a little alcohol or a little sugar did not sensibly increase the current. The effect is probably quickly produced at the surface of contact between the per oxide and the alkali. 13 JANUARY 1840. 6022. Iron and platina plates arranged in strong sol. sulrt. potassium. Iron strongly Negative 30° or more. In 10' it was at 4°. 44 13 JANUARY 1840. Then separated the platina plate, put another down into the solution touching it so as to discharge the state of the first platina plate, removed the second plate and at the 20' from the com- mencement of the experiment rejoined the platina and iron plates. The iron was Neg. only 3°. The thermo current of Antmy. and bismuth passed so fairly as to make the 3° equal to 18°, so that to 15° of thermo current passed. 6023. Iron and platina plates in a strong solution of caustic potassa (potassa fusa). The iron was Pos. at first considerably, perhaps 40°. In 20' it was Pos. 10°, in 100' it was 2°. Separated the platina and discharged its state as above. Then reconnected it with the iron: the latter was Pos. about 8°, falling quickly to 4°. So that Platina, being Neg., had acquired a reacting state in the sol. of potash. The antimony bismuth thermo current did pass, but very poorly, both the iron and the solution of potash against it— scarcely sensible. 6024. Made a fresh solution of proto sulrt. potassm. by passing sul. hy. through sol. potash. By dil. S.Acid the solution gave much sul: hydrogen and very little sulphur. It tasted somewhat and I think must have been pretty good. It has been going on from the eveng. of the nth until this morning, or 40 hours. 6025. Lead and Platina in this proto sulrt. potassm. Lead well Positive about 40°; quickly fell to 10° and then seemed to remain steadily there. It does not go down so beautifully as in the yellow sulphuret. 6026. Bismuth and platina in the proto sulrt. potassm. The Bis- muth was at first well Positive, but after about 20', nearly at 0°, so fell more completely than the lead. 6027. The Negative platina assumes rather a strong reacting state in this solution. Two 6028. platinas in, not alike: difficult to make them alike. Scraped and cleaned one well— then came nearly together— all the difference was due to a little insensible matter lodged in a crevice near the end. On 6029. taking out one and reintroducing it from the air, it is very slightly Negative to the one which remains in, but the difference is not very sensible. Taking one out, washing, drying and reintroducing it, made it ^ little Negative to the other. 13 JANUARY 1840. 45 6030. Action by production of iodides. 6031. Lead and Platina in sol. iodide potassm. Lead is well Positive and an insoluble iodide of lead forms on it, but not as an investing crust, for it wipes off. 6032. Cadmium and platina in sol. iodide pm. Cadmium very positive. 6033. Cadmium and platina in sol. iodide of line, 6034. Lead and platina in iodide zinc— lead very Pos.— continues so. The iodide of lead formed is rough; part wipes off, part adheres. 6035. Silver and platina in iodide line. Silver very pos. indeed and sharply so— the power falls slowly— silver becomes coated with iodide. 6036. Silver and Lead in iodide zinc. Lead Pos. to the silver. Action by production ofchlorides. 6037. Silver and platina in Mur. acid, i vol. sol. strong, i vol. water. The silver was strongly positive. The current diminishes and in 10' was about 8°. Put in a second plate of platina to the first in the acid. The Positive state of the silver rose instantly, shewing the reacting power acquired by the first platina plate. 6038. Native Carb. Iron— argillaceous Iron ore applied as paste to a platina plate, dried, etc. and then associated with platina in dilute sulrt. potassm. The carbonate was Neg., but not strongly. 6039. Per oxide of lead in strong solution of caustic potassa: part soon dissolves and lead is found in solution. 6040. Dilute Nitric acid, i vol. strong and 5 vols, water, had some per oxide of lead put into it; in i| hours a very sensible portion had dissolved, and sul. hydrogen shewed it in the acid, though it could shew none in a portion of the acid preserved for comparison and in which per oxide had not been put. OS Sulrt. sol. water ^C J3 3 vii Potass. strg. yellow vols, 5 vol. I Iron Nickel Bismuth Antimony Lead Silver Tin Cadmium Copper Zinc 6 5 4 38 1 7 9 2 10 Oo Ph X cu vi 'IS Sulrt. dilute colourless Hydro Iron Nickel Bismuth Lead Silver Antimony Tin Copper Zinc Cadmium 6 5 48 1 3 7 2 9 10 Ow ;3 5 CT3 sol. 1—1 V caustic potassa Silver Nickel Copper Iron Bismuth Lead Antimony Cadmium Tin Zinc ^6 strong .a 1 5 2 6 4 8 3 9 7 10 iv Nitric pure acid strong Nickel Silver Antimony Copper Bismuth Iron Tin Lead Zinc Cadmium 5 1 3 2 46 78 9 10 G a. c G n3 sol. water iii M.A. strong Antimony Silver Nickel Bismuth Copper Iron Lead Tin Cadmium Zinc vol. 3 5 9 I vol. 1 4268 7 10 I CD Qh O^ M Qh >^ u O V. water S2 Oil ii S.A. vols, Silver Antimony Copper Bismuth Nickel Iron Lead Tin Cadmium Zinc .5 dil. vol. 1 2 3 4 56 8 7 9 o I 13 10 Acid water -S o i N.A. strong dil. vols, 7 vol. I Silver Copper Antimony Bismuth Nickel Iron Tin Cadmium J_5 13 JANUARY 1840. 47 6042. Nature of Electrolyte, Becquerel's acid and alkali battery is probably connected with Daniell's results on electrolysis of sulphates. Water playing a part. 6043. So also prot. oxide of lead on platina plate is Pos. to the Nitric acid dilute in which it is dissolving. 6044. How would Daniell's results work out if pursued with such a salt as the per sulphate of copper, when the oxide is supposed to be a per oxide. 6045. Iron and platina in strong solution of pure potassa. Iron is Pos., but quickly falls, quicker than in dilute solution of the same potassa. Platina assumes reacting state rather strongly, so iron seems to fall faster. Iron continues Pos. to perhaps 8° or 5° for an hour or more when freed from back action of platina. The arrangement conducts chemical action at x. 6046. Nickel and platina very like iron and platina in all respects in this strong sol. of potash. Chem. action passes, etc. 6047. Silver andplatina in this strong solution of potassa : scarcely any motion— at 0° very nearly from the first. 6048. Used dilute solution consisting of i vol. above sol. potash and 6 vols, of water. This dil. sol. pure potash. Iron and platina \ The iron was pos. and more permanently palladium I think than in the strong solution. Perhaps „ Gold the air dissolved in the diluting water may „ Silver help this action. Nickel Pos.— it falls soon— washing and Nickel and Platina wiping does not restore it, but sand paper Silver Pr»a^\l'il^a\-dmm I cleaning; always makes it Pos. a2;ain— so a hr\im r or -j oxide • is rr ormedj, etc. I Copper and silver— Copper Pos. and tarnishes. Silver— Platina \ Silver very nearly neutral, not quite. Is a little Silver— Palladium Pos. at first. J 6049. Iron wire scaled by spirit lamp is Pos. to platina in dil. sol. Potash and steadily so at first. I think the solution penetrates the crust. 14 JANUARY 1840. 6050. Iron wire oxidized in spirit lamp flame and platina in strong pure pale nitric acid. The scale iron quite under the surface as 48 14 JANUARY 1840. before ( ). Oxide film is Pos. at first; the action falls and after an hour is very small, but still there. Thermo current scarcely passes. Chemical action at x passes current well. 6051. Same experiment with dilute Nitric acid, i vol. of the strong pure acid and 4 vols, of water. Used the same iron-scaled wire. There was the least possible deflection, almost at 0°, but the Iron a trace Pos. The thermo current passed very feeble, but the chemical current at x very well. 6052. When two platina wires are in the strong pure Nitric acid and at 0°, if one be taken out, washed and wiped, it is hardly possible to wipe it so dry (the wires had been sand paper cleaned) that, when immersed, it is not for the moment positive strongly. Yet if, being wiped, it is warmed in the lamp to dry it perfectly, then cooled and immersed, it will be quite passive. So easily does the needle shew the current due to state of surface of the metals. 6053. Iron and Platina in the solution of proto sulphuret of potassium. The iron at first well Negative; then quickly neutral and then went on to be Pos., rising to about 10°. This change occurred in about 5'. The first state is produced by deoxidation of the iron, but the sulphuret is I think quickly dissolved and then sulphuration of the iron and solution of sulphuret goes on. The solution partakes of potash characters. 6054. Chemical action at x made current pass well. So also did thermo current pass, because though one metal iron, yet large plates were used. 6055. Nickel and platina in the proto sulphuret of potassium. The nickel was Positive from the first, and though the current fell, yet it continued slightly positive all the time. 6056. Thermo action did not pass sensibly, but chemical action did well. The metal plate was small compared to the former iron plate. 6057. The tube contained strong sol. of yellow sulphuret of A potassium in and dilute solution (i vol. strong, 5 vols, water) in B. A 6058. Platina and Platina in and B. After first irregular action, A settled quickly to the platina in being about 2° Pos. to the Platina in B. A 6059. Iron and Iron in and B. After first action, settled to iron 14 JANUARY 1840. 49 A in being about 2° Pos. to iron in B. Changed the irons about, but was same thing. A 6060. Platina and Iron. After first action of iron, the Platina in was as nearly as might be 2° Pos. to the Iron in B. A 6061. Iron and Platina. Still the same result, i.e. the iron in was about 2° Pos. to the Platina in B. 6062. So quite clear that contact of Platina and Iron went for nothing; and strong and weak went for nothing, except the small effect of 2°, and this I think due to the gradual combination of the strong and the weak solution. Is in fact constant in direction only for that circumstance, and the nature of the metals go for nothing. 30 JANY. 1840. 6063. Iron in green hydrated Ns. Acid. Clean iron and platina wires. Iron first very Pos., then neutral nearly- though platina is discharged. 6064. Iron wire scaled by heat with Platina in the same acid: first a little Pos., then neutral nearly. 6065. Clean iron in strong solution of pure potash— was considerably and constantly Positive. Perhaps due to air. Used a weaker solution of potash; the iron was still positive— added a little drop of sulphuret of potash to remove the oxygen— still Iron was positive— added a little more, yet iron pos.— more still and still iron was positive. 6066. Scaled iron wire in strong sol. of potash— iron pos. and continued so. In dilute sol. of potash it was still somewhat positive. Added a single drop of sulphuret to remove air and now the scale iron was powerfully negative. Is in striking contrast with the iron, where the sulphur does not reduce an oxide but combines with a metal. 6067. Clean iron plate in strong sol. of sulphuret of potassm.: after 14' put in a platina plate and connected: the iron was about 2° Neg. In 7' more put in a large platina plate not connected; in 5' more connected them and there was the least possible deflection, the Iron being a trace Neg. Hence no reaction back of platina here and iron neutral. A 6068. weaker solution consisting of i vol. strong and 6 vols. FDIV 4 50 30 JANY. 1840. of water was made; the same iron put into it and left a few minutes. Then connected with platina and it, the iron, was a little pos.— going down if contact continued, partly from accumulation On of force at the platina. breaking contact a while and then remaking, it was again Pos. Took the iron out, washed, wiped and dried it and replaced it: it was then well Neg.— but in 4 or 5 seconds was again pos. The Pos. is no doubt a sulphuration of the iron. 6069. Made a very weak solution of sulphuret of potash. The iron was Pos., and if washed and wiped, it was positive from the first. But here the iron quickly blackened in places, from the formation of sulphuret upon it. 6070. Rusty Iron was Negative in this weak solution. 6071. Nickel in strong solution of sulphuret of potash was Pos. to a platina plate. In 5' it was the least degree negative to a large platina plate. So the first action was from matter in the interstices. When At 8' after, it was quite neutral. taken out, it has a yellow tint on it, as if an investing coat had been formed of sulphuret. 6072. Prot. oxide of lead on a platina plate in moderate muriatic acid is a little Pos. to Platina. 6073. Prot. ox. lead on platina plate gave no motion of needle by association with clean platina in dil. S.A. 6074. Plat. Plate dipped in potash is Pos. when put into dil. N. A. in which another plate is present. 6075. Tin and dilute S.A, Acid i vol. O. V., 13 vols, water. Put the two clean tins in in succession. The last in is always Neg, Taking one out, cleaning and restoring it, it is Neg. If both in and action equal, moving one makes it neg. So that the coat which forms at the tin increases the final current. This may be from the first salt of tin produced tending to acquire some other state, as an acid salt, and so adding to the force of the current. 6076. With Mur. acid, the first in was also Positive, not the last. Moving one about made a difference, but not as with Sulphuric acid. 6077. With dilute Nitric acid, weak and very weak, the last in was Positive— also the one moved in the solution was Positive. So this acid not as the Sulphuric or the muriatic. ; II FEBY. 1840. 51 6078. Examd. the Iron and Platina filings of 26 Octr. (5359). The iron was rusted, the surface being covered with a net work of rust and the intermediate parts being bright. This net work had no reference to the platina filings; for the latter were some- times on a bright part and sometimes on rust, but the form of the rust no ways disturbed by them. 6079. In reference to the Negative deflection by heat of several cases, as 5394, 6, 428, 40, i, 70, ascertained, if two pieces of the same metal were taken and one piece plunged in first, which, if there was any difference, was positive. 6080. Used first dilute N.A. of i vol. N. A. and 9 vols, water. 6081. Lead; Lead; Dil. N,A, (5394, 470). The one in first by a few seconds is positive to the one in last, but there is so much difference between the two wires that a particular one always became pos. in the end. 6082. Zinc; Zinc; dil. N.A, (5396). Whichever is in first is Positive to the other— distinctly well indeed. Can this effect depend in part on the circumstance that the one which is in first is well cleaned— or on that of a storing up of force accumulated during a moment by a state of tension assumed by the first im- mersed wire ? 6083. Copper; Copper; dil. N.A. The one first in was the least trace Positive to the other— but nothing like in the degree of zinc or lead. 6084. Iron; Iron; dil. N. A. The one in last is Positive -regularly but only for a moment. No 6085. Copper; Copper; strong sol. potash (5428). difference. No 6086. Lead; Lead; strong sol. Sulrt. pot. (5440). difference if one was in quickly after the other; but if one in first by a minute or more, then the last in was Pos. 6087. Zinc; Zinc; strong sol. sulrt. pot. (5441). The last in was a very little Positive and only for a moment. 4-2 . ; lOTH AUG. 1840. 6088. In relation to dynamic induction (see Researches 1660, etc.). 6089. For the present let P > or < P represent a primary, i.e. a common voltaic or Leyden current; let S > or < S represent T a secondary effect, and let > or < T represent a tertiary Q —©— effect; and let S or S represent the supposed state of the secondary wire after S > or < S has occurred. Then are not the following propositions true ? 6090. It is clear that whilst a primary current is risings thus P a short current or rather a demi current < S is produced— that whilst P Q > continues, S is - S or apparently nothing, and that when P Q > ceases, S passes into S >, 6091. The cycle of conditions in the primary wire is: the natural state ; the rising of the current P > ; the continuance of P > ; the fall of P > ; the natural state, etc. The corresponding cycle of conditions in the secondary wire is : the natural state ; < S —0— then S ; then S > ; then the natural state. Now —P— 6092. is it possible that P wire natural or , and < P and also P —S— can be exactly alike in their action on pro- , ducing not the least difference of effect on or in it ? 6093. P > and < — P P are magnetic; — is not. — — 6094. As S approaches P > or < P, it is acted upon by them and becomes more and more < S or S and as it re- ceeds and is relieved from this influence, it returns more and more — — nearly to its natural state S Q Q 6095. So it would seem that S or S- is in a peculiar state of tension and not in the natural state; and that these changes with distance are the rising and falling of the state by such altera- tion of distance. 6096. But then how is this related to the effects with magnetic curves, for when a wire is taken towards a horseshoe magnet so as to pass between its poles, S > is produced; but when it has passed through and is receeding to a distance, still S > con- Now tinues? the state of that wire at the first and last must be the same surely, and the neutral state; it can hardly be supposed — — Q to be S at one time and then afterwards S . lOTH AUG. 1840. 53 6097. Especially as, by making it go round the pole, one could make the S > state continue for any length of time, and this would imply an enormous accumulation of S Q tension; and yet that wire might be removed without any return through the corresponding < S state and would in no respect differ from a common wire. 6098 . Surely this would imply that S > or < S are discharges of a certain state which the metal wire, l)emg a conductor^ cannot retain; but which, if it were a non conductor, it probably would retain. This would, as on former occasions, direct one's views to the surrounding insulating dielectric. But at present go on to view the action as one at a distance and exerted solely between the conducting bodies. Q 6099. That the wire S has a state seems to be shewn by this, that the moment the < P which sustains it ceases, an < S is produced. 6 100. The < < P and P S and S > currents must not be confounded with A > currents. They are in no way similar. < P current, however short it is, produces S > as it rises and < S On as it ceases, and nothing during its continuation. the contrary, an < S produces T > during its continuation; but it cannot be said to produce either < T or T > as it rises and it certainly produces no current of any kind as it ceases. (See 6187, etc.) 6101. S > and < T S, as also > and < T, etc., are only the ends or elements of currents, and have not the full properties of complete or continuous currents, and must be carefully dis- tinguished from them at present. 6102. But then magneto electric currents ought to have very different properties from ordinary voltaic currents, and so very probably they have. This an important matter for research; for a magneto electric > should continue to produce an < S, whereas a voltaic > does not, but only an effect of that kind at its commencement or increase. 6103. Work this out (see 6187, etc.). 6104. If a difference, then follow that into other forms of current: Thermo, etc. etc. etc.; Friction, etc.; Electrolytic, etc. A 6105. complete current is equivalent to S > and < S and . 54 lOTH AUG. 1840. Q a certain duration of the supposed S state. But an S > is only equivalent to an < Q S and no tonic state. 6106. Approaching a wire to a P > or a < P is equal to an < S or an S >, i.e. is equal to the rising of the P > or < P current. So receeding from P > or < P is equal to the falling of these currents in effect. 6107. Still the question is, has the approached wire an electro- tonic state, or is it in a discharged and indifferent state ? If it has Q an electrotonic, that state may be represented as before by S or s-e-. O 6108. S seems to be in an electrotonic state because, when the P wire is either removed or its current stopped, it produces an < Q S; so that whilst as S it neither carried a current nor yet was indifferent, it was not as P —P— P > or as natural, i.e. Q 6109. But then supposed S lengthened out as in Henry's experiments and that one half of it (that away from P >) were removed and carried to a distance. It could produce no effects and nothing particular would happen either with it or the left How half. then can the S wire be in a tonic state .'^ Surely it is the medium around which is tonic; the conducting matter is relieved, the wires being only places where, because of their nature, discharge occurs. Then both may be in dependant con- dition on P ^, but the one discharged whilst the other is sustained. 61 10. If so, what is the state of the S insulating dielectric and what are its qualities ? 6111. When a wire S has been brought alongside a P > and had an < S produced in it, another wire brought up either before or beside or even behind the first S wire, or placed any where else near P >, will have its < S produced with little or no interference. (But see Henry's screening effects connected my with this point and also non screening results). 61 12. Will masses of metal placed in any way as circuits or other- wise affect the inducing power of P > or < S? 61 13. And must remember here that these lines of inductive force as given by a magnet do not seem to differ in the least through conducting or non conducting matter provided both be still. lOTH AUG. 1840. 55 61 14. But if the electrotonic state were discharged in the metal and sustained in the insulating matter, and if the action were carried on by contiguous particles, how could it be a matter of indifference whether the space was filled up by a metal or an insulator ? 61 15. That S > or < S are not currents is shewn by their being active during their whole continuance and to their full amount, whereas currents are not active at all for their continuance. By their producing < T or T >, whereas currents produce nothing. By their producing only < T or T ^, whereas a current began and finished produces both < S and S >» 61 16. Any increase of a current is equivalent to either a beginning or exaltation of state; and decrease is equivalent to the contrary. This shews that the continuous current is a very different thing to the change of degree or state of that current. 61 17. Any approximation of the S wire to a current? > is equivalent to an increase in the force of that current, or the addition of the beginning of another current; any removal is equivalent to the contrary change of the P >. This is a very simple and natural relation, and shews that it is not the current as current which produces any of the secondary effects— except the electro- tonic or quiescent state if it exist. ^p 61 18. ^ When p currents are placed side by side, as thus ^ p^ ^^, or thus ^ p, then we have the effect of two interfering electrotonic states (or their equivalent states), either exerted in the same or in contrary directions; and we learn by the facts that Magnetic attraction and repulsion is the consequence. 61 19. So we ought to trace, in attracting and repelling magnetic poles, the full and consistent effects of the supposed electrotonic states. 6120. Then remember these take place just as well through metals as through air or shell lac. 6121. In Magnets the Electrotonic states are sustained. What does this lead to in simple helices ? in soft iron cores } in mere ordinary magnets } 56 lOTH AUG. 1840. 6122. When the S currents, or rather halves, act like complete currents, as in deflecting the needle, electrolyzation, etc. etc., what is the amount of the effect? It ought not to be the same both in kind and in quantity. Perhaps its amount is only a half. 6123. Will an S ^, whilst producing the same continuous effect as a P also continue to induce a corresponding < T. The P > used will not induce a corresponding < S whilst it con- tinues. 6124. Must make out the true relation of static induction and the electrotonic state— and the direction of the forces in that relation. 6125. Usually there is no corresponding effect at the rising of a current Hke that at the conclusion. But probably produce it by beginning the discharge by an electric spark. 6126. The cause of the non-effect at rising is perhaps because the action is expended in retarding the primary current itself— but why? 6127. Follow out this in the action of a long wire carrying a current, and also into effect of beginning spark in cases of Gassiot, Crosse, etc. 6128. The inductive result is the experimental analysis of a voltaic current or an electric current, and these two, S > and < S, appear to be the elements of the current; then a current will be proved to be a compound action; and then the two elements will be important in the consideration of all inductive and electrical phenomena, and especially in all those of discharge. A 6129. current appears most clearly to be a compound effect. II AUG. 1840. 6130. If P > differs so much from S >^ does S > and < S differ from each other except in direction? i.e. would < S turned round make S > or no ? 613 1. Perhaps not, if < S and S > have any relation to the places of separation of oxygen, hydrogen, etc. —^ — 6132. Call the rising S > or < S, S and ^—^ S; and the —^ ^— falling S > or < S, S and S. Then a P > on — —^; rising would produce S, and on falling an S —^ ^— < P on rising would produce an S and on falling an so a S. II AUG. 1840. 57 6133. Then if there be these currents, what curious association of these with each other and with P > or < P may be made, even in tlie same carrying wire ? and must be examined. S—^—> 6134. What will g 2 ^ the same wire do? 6135. What ^T^P —^? 6136. Will g I make only a stronger S ^ 6137. What will ^ 2 g do? 6138. What ^~7^? 6139. What ^ ? Not the same as the last. 6140. Is S ' > and S ^ > alike? 6141. What will g I do? 6142. What g^T^? with a great many other changes. 6143. Surely it is impossible the distinctions can exist. The great- ness of the consequences almost forbid it. Still, if Henry be right my and if first papers be right, there must be some difference between a principal and an induced current. See (6187). 6144. Will ^ 2 g make an ordinary current? They ought. If so, what will be its direction, P > or < P ? and what decides the direction ? 6145. But P > and < P make nothing. ^g S_i 6146. What particularly will ^ make ? ^ 6147. 6148. Also ^_2_g -"^ S— And also g 2 6149. Turn a current round in the wire, so as to traverse in its course thus, using a secondary current, not a primary. See 6187, etc. 58 12 AUG. 1840. H 6150*. Arranged thus : is a double helix, of Palmer's apparatus, G either with or without an iron core. is a pretty fair galvano- N meter. When the ends of the wires P and were put in contact G with a voltaic pair, was deflected on making contact and also on breaking contact— but not on continuing contact. So induction N could act and insulation was right. But when P and were made to touch the Edge and axis of my revolving copper plate moving between the poles of a horseshoe magnet, no effect took place at G, though an induced current was running round where the voltaic current was running before. H 6151. Whether the iron was in or out of made no difference. 6152. Perhaps the exciting apparatus was not strong enough or G the test part not delicate enough. Yet when the wires of were put at once in connection with my revolving wheel, then the Galvanometer was deflected, perhaps 30° even. 6153. So no evidence here of any peculiarity in the induced cur- rent, yet it ought to have it if voltaic induced currents have it. 6154. Use own powerful ring Electro magnet for the first magnet. Henry's helices and larger helices in place of H, and a more delicate Galvanometer. 31 AUG. 1840. 6155. Have arranged to examine the inductive action of an induced current. I arranged a powerful Electromagnet. Then I set my revolving copper wheel at work and carried its current round a large coil of Henry's; and then induced by that upon another coil, whose current circulated round a delicate galvanometer. The arrangement as seen nearly in plan was as below t. D, a Daniell's battery of 3 cells, 18 inches high. N M, own large ring Electromagnet : as connected, and S now the north and south ends. This includes the Primary current. W, my revolving copper wheel, with collectors at the axis and periphery, passing away to Gassiot's large flat coil A. [6150] t [6155] 31 AUG. 1840. 59 A, Gassiot's coil of 300 feet of copper plate inches wide, W A covered with cotton and forming a coil 20 inches in external diameter, having the direction given by the curve. and my include the whole circuit of the first induced current; current or Henry's indary current. B, Daniell's coil of 200 feet of copper ribbon, forming a helix or coil of III inches in diameter, covered with cotton. This had the direction shewn and was placed concentric with and over the first. The forms of these really were as below\ G, a delicate Galvanometer indicating thus: when zinc was at A and Platina at B and the tongue between, the deflection was as shewn and the current as marked. When the deflection was in the reverse direction, the current was the other way. Now 6156*. direct revolution of the wheel affected the galvano- meter, and so did the reverse revolution— in the opposite direction. So the secondary current thus obtained could induce a tertiary. Direct revolution is that marked by the arrow on the wheel. 6157. Direct revolution gave this current and reverse revolution this. So with direct revolution the currents throughout would be as marked above^. 6158. But now the constant secondary current did not induce a constant tertiary current, for the Galvanometer came to rest, and though the secondary current continued in the direction of the arrow in the helix, yet the contrary tertiary current ceased instantly. So these currents are not peculiar in any such quality. 6159. Then it was to be expected that the secondary current would induce at its beginning and its ending, like other currents, and in contrary directions; and this was found to be true; for keeping the wheel in continual direct revolution and making and breaking contact at the periphery of the wheel, it was found that on making contact the galvanometer deflection was thus and on breaking contact thus. 6160. So the secondary current is like any other current in this, that at its beginning it induces a current in the neighbouring wire in the reverse direction^ and in ending it induces a current in the ^ i.e. in margin. * [6156] ^ i.e. in diagram below [6156]. 6o 31 AUG. 1840. same direction. But it is unlike the primary current in this, that it induces most strongly at its beginning, whereas the primary induces most strongly at its ending, and hence the reason why, as Henry experiments, the difference between the action of the primary and the induced currents. 6161. When all the connexions were complete, causing direct motion of the wheel made the tertiary current thus. Reversing the motion of the wheel caused a reverse current thus, and these being alternated made the Galvanometer needle swing very well. 6162. But even stopping the wheel causes a tertiary current, the reverse of that of beginning revolution; and reversing the motion is in fact doubling the effect which stopping the revolution of the wheel would alone do. 6163. So causing direct and reverse motion of the wheel is the most effective way of shewing the tertiary currents. But still every current can induce two currents, one at its beginning and another at its ending, and none merely during its continuance. 6164. The superiority of the current induced at the beginning of the secondary current rather than the end may depend upon the circumstance that both the circuits concerned are entirely metallic, for this allows the secondary current, when it is about to stop, to induce in its own wire, which it would rather do than in the neighbouring wire, and so on. 6165. Probably when, with the machine now constructing, the secondary current is made to begin and stop by making and breaking contact at the edge of the wheel more effectually than now I can do, the breaking contact may be as strong. SEPTR. I, 1840. 6166*. The arrangement was as before and as above^ The direction of the permanent induced current when the motion of the copper wheel was continued is as marked by the arrows above^ * [6166] ^ i.e. in diagram below. SEPTR. I, 1840. 61 6167. When the wheel revolved direct, the Galvanometer was thus. When wheel revolved reverse, thus. This agrees with the former observation ( ). 6168. I caused the secondary current to begin and cease, not by beginning and ceasing to rotate the wheel, but by continuing the direct rotation of the wheel and making and breaking contact at the edge of it, as at E. In this way the current had more sudden terminations. 6169. Making contact caused deflection by the tertiary current, thus; and breaking contact cause deflection and current thus. So that making contact is equivalent to beginning to rotate the wheel, and breaking contact to ceasing or reversing rotation of the wheel. All this agrees well together. 6170. I now kept the wheel still, but made and broke contact at the battery, and there was certainly motion at the needle of the tertiary current. Making contact sent galvanometer this way and breaking contact this way. But when the contact at E on the wheel was broken, the same effect took place as if it were the action of the large electromagnet at a distance and not any induced electrical current. 6171. Indeed, such making and breaking of contact ought to do nothing, for then the radius of the wheel and the magnetic curves do not move across each other, but the curves are projected end ways, so to speak, through the radius. It is as if a wire were brought right up to a magnetic pole and not carried hy it and across the curves. In the former case, no current is produced; in the latter case, one is formed. 6172. But try this again unexceptionably. 6173. To prove the great point that no current induces during its continuance, I connected the Daniell's battery with the first coil (Gassiot's), and though the secondary current at making con- tact was very powerful, yet the continuance of the primary contact gave no effect at all. 6174*. I turned the two helices round so that Daniell's was in the wheel circuit, the battery and the magnet being reconnected as before ( ), and Gassiot's helix the one induced upon. The wheel therefore sent a secondary current which, in a coil of 200 feet, induced a tertiary current in a coil of 300 feet connected *[6i74] wx:->^ 1' ^ \> ^ //^ <^ / ^ f/i. ^ ^ ^ ^ ^//\ ; -^ 62 SEPTR. I, 1840. with the Galvanometer G. The arrangement was as in the next page^ The effect was about the same as in the former case. The direct motion of the wheel caused deflection thus, and the reverse motion thus. So that here, on making direct revolution, the secondary and tertiary currents were, as before, in opposite directions. 6175. Placed Solly's helix on Daniell's and connected them into Now one consistent helix of 263 feet in length. the effect stronger. 6176. Repeated the making and breaking of battery contact ( ) and obtained an effect in the same direction as before. But when the secondary circuit was interrupted permanently, still the same effects. So must be Magnetic induction at a distance. 6177. I wished to obtain the inductive effect of the breaking of the secondary current by itself and quite clear from any other action of making or reversing that current, and proceeded thus. The wheel was turned direct continually, the other contacts not yet being made ; then they were made and broken in the following order: contact made in wheel circuit— made in galvanometer cir- cuit—broke in wheel circuit— broke in galvanometer circuit— made in wheel circuit, etc. etc. etc. In this way the galvanometer or tertiary current depended entirely upon the breaking of the wheel On or secondary circuit. repeating the action several times, so as to accumulate the effect of the breakings, upon the galvano- meter the effect was thus for the current induced by the breaking, and this is just the reverse of that due to making. It was not nearly so strong as the latter, according to appearances. 6178. As to revolution, stopping and reversing the motion of the wheel, the effect with the present arrangement was thus: Revolving the wheel direct caused tertiary current thus Stopping the above revolution Revolving the wheel reverse Stopping the reverse revolution and there was the pull at the needle each time of stopping as at the time of beginning to revolve. 6179. So the secondary current being continued and during its continuation does nothings and these currents are just like other currents. ^ i.e. in the diagram above [6174]. SEPTR. I, 1840. 63 6180*. Now threw the Electro magnet and wheel out of use and arranged the Daniell's battery with the compound Daniell-Solly coil ( ), and allowed that to induce upon the Gassiot coil ( )• On making battery contact the galvanometer was thus On breaking battery contact „ „ thus So the induced or secondary current was on making contact the reverse^ and on breaking contact the same in direction as the primary current. 6181 1. Put in our great cylindrical helix of thick wire so that one of the helices in it should, with the Gassiot coil, form the secondary current circuit, and the other helix in it with the Galvanometer form the tertiary current circuit. But whether the iron core was in it or not, I could not perceive clear indications at the galvano- meter on making and breaking the contact at the battery. 6182. The helix consisted of two thick wire coils, each perhaps about feet long. Perhaps this was not length enough for a good induction. 6183. There is I think great power lost at each inductive step. 6184 ^. Arranged things thus for the Tertiary current. Daniell's battery with Gassiot's large coil ( ) formed the primary circuit. Daniell's coil over Gassiot's, with Solly's coil and a galvanometer, formed the secondary circuit; and Gassiot's long fine wire helix in the middle of Solly's coil, with a Galvanometer, A formed the tertiary circuit. Both galvanometers, and B, indi- cated the same way, i.e. thus. [6180] t [6181] t [6184] 64 SEPTR. I, 1840. Now 6185. had a result; on making battery contact, B Galvano- meter moved thus, and on breaking contact, thus. 6186. Again with the same arrangement. A Making battery contact, was thus and B was thus. So each induced current was the reverse of the inductric current, as in former cases of making contact. Only, as the making and ceasing of the secondary current must have been at the same moment that it induced the reverse current in the tertiary circuit, so the making influence seems by far the most powerful, and there are natural reasons, plenty, why it should be so. 6187. So all my visions of new kinds of current are gone. See (6143). See also 6100-4, 15? 22, 3, 8 up to 6187. 6188. Gassiot's coi/ consists of 300 feet of copper ribbon, inch wide, covered with cotton and made into a circle having an external diameter of 20 inches and an internal diameter of 8 inches. 6189. Daniel!'s coil: 200 feet of copper ribbon i| inches wide, in a coil i2| inches external diameter and 8 inches internal diameter; the copper ribbon covered with cotton. 6190. Solly's coil: copper ribbon i| inches wide, covered with cotton, made into a coil 9^ inches external diameter and 7^ inches internal diameter; there are 27 circumvolutions at a median length of 28 inches, making 63 feet of ribbon. 6191. Our double helix of thick wire. One wire 37 feet and the other 42 feet long and of inch thick. 6192. Gassiott's Wire helix. 6 coils of 40 feet each; also one coil of very fine wire about 4000 feet long. 14 SEPTR. 1840. 6193*. Gassiot made an experiment here to-day to shew me that the induced current always went in a peculiar direction and against expectation across air under certain circumstances. H is a double iron helix, the one described in the preceeding page I, and the iron wire core was in it. Z, C were three Daniell's U cells. The terminations of the second wire at T, were of fine On copper, parallel to each other and about of an inch apart. I i.e. at par. 6191. * [6193] H 14 SEPTR. 1840. 65 making or breaking contact at no discharge of the secondary- U current happened at T, at common temperatures; but when a lamp flame was applied there, then the current induced on breaking contact at x almost always passed as a spark or brush. how 6194. It is striking to see the effect of the hot air: great it is. Holding the spirit lamp under immediately makes the spark pass at the hot place. 6195. It is not the carbonaceous flame only that does this, for the hot stream from the end of a blow pipe does it also. A 6196. very striking circumstance is that, as the discharge always T takes place at uniform temperatures, either from the end at or at U, so it is always at that end which is negative. Thus, if the secondary current were moving as marked, then the discharge would be at T. If the current were reversed, then it invariably took place at U. Ascertained this very carefully by several examinations of the coil, etc. We 6197. This constancy must be important. should have ex- pected that the discharge would have been from the Positive point to the Negative large surface. Perhaps it has relation to Belli's results (Exp. Researches, 1520). He found the negative electricity my more easily dissipated than the positive. See also similar con- clusion at 1 501 of the Exp. Researches. FDIV 5 1 JUNE 1842. 6201^. Have been making a few expts. on electricity of steam, to see whether it might not be from friction against metal, as the metal cock or pipe. Have Boiler of the London Institution, which has going from it a pipe with 2 stop cocks. One of them I directed over the hot chamber of a furnace, to procure dry steam in the jet, and the other went into the air. 6202. The steam, whether at the hot cock or the air cock, and whether collected against an insulat[ed] conductor or a plate or by a wire or a flame in the jet, was always either not electrified at all or else Positive, Often not electrified apparently when it might have been expected to be electrified, and then appearing to be so. The electrometer moved up and down as if the state taken on by the successive portions of steam were capricious; but still the steam was always Positive if electrified. 6203. Had an electrometer at a distance of several feet from the cock and sent puffs or a stream of steam and air near it. It hap- pened that 2 or 3 times at first, this electrometer, with a flame on it, was charged Negative; but afterwards it over and over again under the same circumstances became Positive^ and I could not Do again make it negative. not as yet see the reason of this. 6204. At the beginning, there were 11 inches depth of water in the boiler; worked about i\ hours on and off; then there was about 9 inches depth of water left, so that only 2 inches of water On had been used. filling the boiler up again, we found that a gallon of water occupied just an inch in depth. 2 JUNE 1842. 6205. Worked again with the boiler, but now supported it on three pieces of shell lac, so as to insulate it. 6206. Put the Gold leaf electrometer at a distance from the end cock and threw jets of steam towards it, as yesterday (6203). A flame was on the electrometer and it was obtained in a Negative state, as yesterday, over and over again. Indeed, it was obtained only negative at this distance. I Numbers 6198 to 6200 are omitted in the MS. 2 JUNE 1842. 67 6207. Had a new collector— a short wide tube and 3 or 4 wire gauze diaphragms in it. This, with the insulated conductor, acted well at the side cock over the hot air. The sparks from the conductor were good, and continued to come as long as the blast of steam was running against it. 6208. Insulated the boiler, and instantly found it negative as soon as the jet of steam was allowed to issue at the side cock (over the fire). Obtained a constant stream of sparks from it as long as the jet continued. Now 6209. opened the valve hole full size; made good fire under the boiler and let steam escape freely. Not the least signs of electricity in the boiler by the gold leaf electrometer— or from the issuing steam by the same electricity^ and flame. So at low pressures no electricity evolved here— is as if the mere evaporation does not cause it. 6210. Put in the valve and with the lowest pressure began to try the cock issue of steam, connecting the electrometer with the boiler. Found this an admirable method, for the Electrometer gradually became Negative as the steam issued out and the whole was under beautiful management. 621 1. Best way by far is not to examine the issuing steam, but the state the boiler is left in. 6212*. The present steam pipe has two cocks, one at the end, the Now other at the side. found that, if the end cock were opened at this pressure, the boiler gave no electricity; but if the side cock were opened, the boiler gave beautiful charge. So the electricity is not caused apparently by evaporation, but at the cocks. This is still for the friction view. 6213. The steam has to turn a sharp angle inside the side cock, but not at the straight cock or end cock. 6214. Here then two steam issues from the same boiler, and one generates electricity, the other not, at this pressure. At a higher pressure both do so, but the side cock is by far the best; the end cock but little. 6215. The steam valve is a circle of 0-5 of inch diameter; the valve and weight together i lb. 2 oz. This is not more than 6 lb. upon the square inch for our highest pressure. Now 6216. put in an old side connecting piece to the end cock * [6212] ^ ? electrometer. 68 2 JUNE 1842. and then let steam out there. Found the boiler now constantly Positive, not Negative— but to a slight degree only. So by the exits may make either the steam or the boiler Positive or Ne- gative. 6217. Put a conductor and gatherer to the issuing steam of side piece and an electrometer attached. Put another electrometer to Now the boiler. the side piece termination gave steam which made its conductor Neg. and left .the boiler Positive. This was done many times. But when the steam was at the highest force, occasionally had the steam Positive and the boiler Negative. 6218. At last the steam lifted the valve, and this issue now made the boiler negative, though when quite open it could not. The steam I found by Electrometer and lamp was Pos. 6219. But on letting out steam from the end side piece, its steam was Negative. For first it neutralized the effect of the boiler electricity produced by the valve issue, making the boiler neutral, and next when sent against a collector it made it Negative. 6220. So then at this moment Pos. steam was issuing from the valve and Negative steam from the side hole at the end, and the boiler was neutral. 6221. Hence see the power of the issue and that the electricity is generated there. 6222. Again and again the Valve issue made boiler Negative, The End cock made boiler Positive. The side cock over the fire by itself always made the boiler Neg., as the valve did, but when the side connection was put on ( ) it often made the boiler Positive. 6223. The side cock of the apparatus (London Institution and a different shape to ours) was put on at the end of the pipe instead of the other cock. The first time it was used there it made the boiler Positive, but afterwards always Neg. 6224. Held wires and other things about half an inch from the end-cock in the jet, to see what friction would do; the wires, etc. being connected with a Gold leaf electrometer. In this way, bright copper wire, oxid. copper wire, plumbago, platina and wet string always were Neg., and made Electrometer so strongly. But with wet string put the boiler out of insulation, and then no effect, as if the string had gathered its state from the boiler. ; 2 JUNE 1842. 69 6225. When boiler was uninsulated, still metal of rod gave strong Neg. when held in the stream. 6226. Worked from 10 until 2 ocl'k., about 4 hours: in this time On level of water had sunk from 11 inches to 5 finches. filling it up to 1 1 inches again, 4I gallons of water were required. The boiler is smaller below than in the middle. 6227. Held glass and shell lac in issuing jet of steam, but per- ceived no effects. 6228. May perhaps equal a voltaic battery by this mode of obtaining steam. 6229. The ordinary charcoal smoke of the furnace shews no effect, nor does it leave the furnace sensibly charged. 4 JUNE 1842. 6230. Boiler up and insulated. The side and end cocks the same as before ( ) ; the side cock over hot air stove. 6231. The side cock instantly and always renders the boiler Neg. 6232. The End cock at first with low pressure rendered the boiler Pos.^ with little exit of steam, but Neg. with more exit. After- wards, with higher pressure, always rendered the boiler Negative when the cock only was used. 6233. But when an end piece was on the cock, then over and over again gave the boiler Positive, 6234. At last, when the pressure of steam was up, the end connector did nothing, i.e. the boiler was not charged. But if the side cock was opened for an instant only, it made the boiler well Negative. 6235. Have a two way end piece thus; when put thus on to the pipe*, it gave no signs of electricity to the boiler. But when put on thustj it made the boiler fairly Positive. 6236. When piece on thus, the passage of steam was so quick through the termination that all went out at a, and air actually entered at h. So finger drawn to h. Expect that steam should not touch sides of channel again when it has passed the place of quickest motion, or it becomes discharged again. On 6237. repeating the last expt., now had no effect on the boiler all null. Taking off the cross piece and using the end cock alone, it now always made the boiler Neg.., as the side cock does, but to a very much less degree than it. [6235] t 70 4 JUNE 1842. 6238. Used another end piece which had a very open way, for the former piece had a contracted way in the middle. This open piece on the end cock gave the boiler Positive, It was in a pretty clean state. A 6239. jet, a gas pipe, etc., were put on the end cock, but nothing particular occurred. Suspect all the effect takes place at the first contraction of passage and that after attachments only tend to discharge steam and modify effects. 6240. It is clear that a certain cock will sometimes make the boiler Pos. and sometimes Neg., and from yesterday's expts., the state of the steam is reversed at the same time, being always con- trary to the boiler. 6241. It is very remarkable to see how little steam from the side cock will make the boiler Neg.^ whilst it is comparatively difficult to obtain effects from the end cock and piece. I rather think this is because the side cock has a clear unbroken way through, which allows the steam that has brushed the edge of the inner aperture and become charged to travel unmixed down the channell and escape, whilst the others mix it up. 6242. I think much depends upon the free escape of the steam after it has past the place of friction, and also on the free space before that place. 6243. Used the side cock and held stretched string across the issue of steam close to jet. Boiler uninsulated and an electrometer attached to it. The boiler electrometer perfectly quiescent, but the string electrometer became charged Neg.^ for the string was Neg. This not derived either from the boiler or the finally charged steam, for first is neutral and the second pos. ; was due to friction of string and steam. Now 6244. a good expt. Boiler uninsulated and could not charge its electrometer. The string across distant steam, i.e. 3 or 4 inches from the cock jet, was a collector and made its electrometer Positive^ that being the steam electricity. But the same string close to the issue, so as to be rubbed by the steam, then became Negative— 3x16. these effects could be produced at pleasure. So the same stream of Steam, issuing from a metal cock, gave either Pos. or Negative electricity; whilst the boiler, being uninsulated, could do nothing. Hence the generation must be by the 4 JUNE 1842. 71 friction of the steam, at least when the string was rendered Negative. Now 6245. used Platina wire instead of string. Excellent. For boiler did nothing all the while to its electrometer, whilst the wire near the cock was Neg.^ but in the steam more distant was Positive. The first kind was excited on the wire by friction. The last only collected from air rendered Pos. by friction in cock. When 6246. a string in jet, beautiful vibration or rather revolu- tion either on one side or the other of the centre of jet. Must, to have friction, adjust so that string is in the centre and not moving, or else it does not charge. Hence again effect of the fric- tion. .... 6247. Used Plumbago as collector', then was Pos. „ „ as exciter close to cock : then . Neg. „ Copper wire as collector Pos. „ ,, „ exciter Neg. „ Chip of wood as collector Pos. „ ,, „ excitor Neg. .... „ wire gauze diaphragm as collector „ ,, „ „ as exciter . . . Pos. Neg. All these bodies near the jet were Negative^ but in the more distant steam became Positive, 6248. Easy theoretically to make a famous Steam electricity We battery that may perhaps replace even the Voltaic battery. shall see. I see no positive objection. The indications are quite as good as those relating to the voltaic battery were in the first instance. 6249. The effects with a wire gauze diaphragm are very beautiful. Held near the jet, the gauze and all connected with it is Neg. Held a few inches off, it is Positive, and by moving it to and fro I could make the Positive electrometer collapse and then become Neg. or Vice Versa ; or I could find out and hold the gauze in a place where the effects were equal and it was not charged at all. A quarter of an inch from this on either side would make the diaphragm Pos. or Neg. at pleasure. 6250. If a diaphragm of fine wire gauze were in a glass tube, and a stream rushing by it, it might be exceedingly powerful as a charger. 72 4 JUNE 1842. 6251 *. a, circle of very fine and close wire gauze in a glass plate b ; c, steam chamber; escape chamber, open to air except that wire gauze collectors at e; collecting wire[s] go from a and e to /, which are the electrodes. 6252. Silk thread, stretched across retort ring, was used wet and dry. On the whole, when placed near to the jet and when the electricity conveyed away, could not be either of issuing steam or of boiler, but that produced by friction; it was Neg.^ as the other rubbed bodies were. 6253. With this silk thread on retort ring observed a certain peculiar first effect at this side cock, as follows. After resting a while, the least puff of steam sent against silk and ring as a collector would instantly open the leaves of its electrometer; but on repeating the trials the puffs lost their effect, and at last the con- tinuous stream did, not so much as a puff before. The steam was now evidently not so electric as before, i.e. directly after the rest. On shutting the cock, waiting a few minutes and beginning again, the same effects recurred. The electricity was always Positive, but the point was the going down of the power, the force of the steam not going down. This was with a low pressure. 6254. When a metal collector was used, the same was the case. 6255. This was not the case at the end cock, but that very feeble as to any action compared with the side cock. 6256. The cause is doubtful— is not dependant on the degree of pressure— is apparently dependant on some state assumed by the stop cock and not on the steam. 6257. Insulated the boiler and tried it for this first effect at side cock at low pressure, and effect was certainly far stronger on beginning of the emission of steam than afterwards. 6258. Must find out the cause of this. 6259. The end stop cock has no power at low pressures as com- * [6251] 4 JUNE 1842. 73 pared with the side cock— and the above phenomena of changes are brought out at the side cock better at low pressures than at higher ones. 6260. Have worked from 10 to | p. i ocl'k. to-day, or 3I hours; have consumed 3^ gallons of water. 6261. The pipe for carrying off the issuing vapour is excellent. It prevents all interfering inductive effects of the steam on the electrometer and keeps the place dry. 6262. The electricity of steam and steam boilers is not due to evaporation, but friction. SATURDAY, 11 JUNE 1842. 6263. Have had a new tube made, inches long and of inch in diameter within; at end of it a stop cock with steam way equal to the tube, and after it a brass globe inches in diameter, with a wide mouth at furthest side to which various apertures, stop cocks, etc. could be attached. The object was to make this globe a steam chamber and if necessary in it to de- electrify the steam (supposing it electric in the tube); and so experiment on effect of friction of steam alone. This globe had aperture of inch in diameter. 6264. All this in its place and the boiler tested as to its insulation and keeping of charge, which it did well. Now, when the principal cock open and the steam issuing by the wide aperture of globe alone, there was no signs of electricity either in the boiler or in the steam. Yet if boiler charged by shell lac whilst the steam was going off, it kept that charge unchanged. So evaporation of water in the boiler and its issue by this open aperture neither charged nor discharged the boiler. It did nothing. 6265. Shutting the steam cock, the issue of steam by valve of the boiler did slowly make the boiler Neg.—friction. 6266. I had plates with small holes in them made, to close the outer aperture of the globe, so as to give fair issue of steam out of a small hole. Steam issuing out by these did not charge the boiler. A 6267. fine wire held in this issue of steam produced no elec- tricity, either in the wire or the boiler. 74 SATURDAY, ii JUNE 1842. 6268. Put a straight stop cock in thus*, and now the boiler was well Positive. 6269. Put on a right angled connecting piece, which had of course a female screw inside at both apertures. This made the boiler well Positive more than before, and when the conductor and connector were arranged to catch and try the steam, its electrometer was Negative, So boiler Pos. and Steam Neg. 6270. After a few minutes, repeated the last expt., both boiler and steam conductor having their electrometer. The steam was now something Neg. and the boiler Nothing— OYtr and over again. Then on a sudden both conductor and boiler electrometers diverged well, the Conductor Neg. and the boiler Pos., as before. The mode of issue of the steam apparently made this difference. 6271. Then the last result happened again and again, i.e. the steam conductor was Neg. and the boiler Pos. ; and also with this side issue I obtained by a wire the place of Neg. on the wire near the issue. 6272. It appears to me that, when the issue of steam is rattling and noisy, that then the exciting condition is very influential. This noise depends upon the way in which the air and the steam conflict, principally. 6273. There was more water present in the steam of these last experiments. The furnace beneath the issue was not over hot. 6274. Took of[f] the right-angled connecting piece and now the straight cock alone did nothing, i.e. it produced no divergence of the boiler electrometer— very remarkable. 6275. There is a draining cock on the steam pipe just before the large issue cock; it of course points downward. Put the side piece on to this, and the issue of steam here produced no effect on the boiler electrometer, but did a little affect the steam conductor electrometer, making it a little Neg. 6276. Took off the right-angled piece and tried the cock alone, and now it affected neither the boiler nor the conductor electro- meters. 6277. Worked from 10^ to i2|, or two hours; used 2 gallons of water. * [6268] 30 JUNE, 1842. 6278. Worked again. Boiler filled up to the mark— well insulated. Worked with my tube and globular termination. Rather in- different about the fire beneath the termination or exit. 6279*. Had a new stop cock of ours at the end. No electricity in the boiler produced by it. 6280. Put on an old cross piece to this new stop cock. This old cross piece has its three female screws connected by a narrow air or steam way, as in the figure. The issue of steam was open right and left. The boiler electrometer opened fairly and Neg. 6281. After a few minutes repeated the experiment, and now boiler electrometer quite nul. 6282. After a few minutes more, repeated expt. again, and now the boiler somewhat Neg,., as at first, and more so as the pressure was higher. 6283 1. Took off cross piece, and had the straight stop as at first, and now boiler neutral or nul. 6284. Now put on New cross piece, which has the female screws carried throughout, so as to be of large bore and rough through- out—and now the boiler electrometer opened out well and was Pos. Charged a Leyden phial by \l.—Pos. 6285. When the steam was left issuing, the power (electric) went down much, though all the time the pressure in the boiler was such as to lift continually the regulating valve. So diminution of power was not due to diminution of force of steam. 6286. Whether the steam was allowed to escape by opening the great or the small cock, i.e. the one on the one side or the one on the other side of the Globular steam vessel, still the boiler became Pos. 6287. When new cross piece was put on thus, there was no issue of steam at the side hole, but an indraught of air there, and the boiler was only a little Pos. Nothing like so good as if cross piece was in the former Position. 6288. Stopped up the side hole by a stop cock closed and now the boiler was fairly Pos. 6289. When the steam issued only by the regulating valve of the t [6283] 76 30 JUNE, 1842. boiler, it shewed its power of producing Neg. electricity by causing the boiler electrometer to diverge gradually Neg. This valve is always Neg,^ i.e. it always charges the boiler Neg., giving Pos, steam— at least, I expect so, but have not tried it. 6290. I now stopped up the end hole of the same new cross piece, leaving the side hole open, and now boiler was very Pos.^ more so than when we had two lateral issues. Had also the cracking, crushing, sharp sound in the issue of steam which always seems to accompany good evolution of electricity. 6291. Tried the steam of this side issue by insulated conductor, its tin collector and the electrometer. It was very variable, the leaves opening and closing continually and making the conductor electrometer sometimes Pos. and sometimes Neg. The boiler electrometer was always Pos. 6292. After a few minutes, repeated the expt. and same variable results. Now 6293. put Grove's Stop cock on at the side aperture and had a fine electrical effect at once. But the boiler was now regularly Neg. and the steam Pos. 6294. Still, this full effect did not continue, for though the issue of steam went on, the Electrometer fell and often became quite iVw/, rising a little now and then. 6295. The going down is an effect common to the two issues, but this stop cock makes boiler Neg., and the cross piece alone made it Pos. Remember the cross piece is roughed by screw thread and that the stop cock of Grove is smooth through- out. Now 6296. put Grove's Stop cock on to the globular vessel; it made the boiler freely and well Negative. 6297. Turned the cock end for end— same effect as before. 6298. Put on in the same place a new stop cock of ours, and it always made the boiler Pos. So these two stop cocks in direct opposition to each other as to their electrifying power. Now 6299. the only difference I can see as yet is that Grove's Stop cock is smooth throughout, and ours had a female screw thread turned in it at each end, to the depth of a quarter of an inch. I believe that is the cause of the difference. The smooth tube makes boiler Neg., the rough tube makes boiler Pos. This 30 JUNE, 1842. 77 falls in well with all the preceeding effects (Is the effect of Oil, see further on). 6300. Again experimented with our stop cock as before, but now four times in succession it made boiler Neg, Turned end for end- still boiler Neg, 6301. More steam so as to produce rattle probably would have made it Pos. 6302. Put some fine wire up Grove's Stop cock so as to offer resistance and roughness, but now issue of steam quiet and small and no electricity. Steam must get away. Removed the wire and now the boiler made Neg. 6303*. Tied fine wire gauze over the end of Grove's Stop cock and then put it on to Globe. The furnace and boiler was Neg,^ and better by far than if no gauze were there. So this appears to improve the effect. 6304. I observed, whilst continuing the issue of steam, that it was not so noisy as before— but more quiet and tranquil. But whenever it broke for a few moments into a crackling sound, as often happened, then instantly there was an increase of electricity shewn by the boiler electrometer; and when by increase of pressure this rough issue was continued, then more electricity was produced than if quieter. 6305. Still, the first effect was always the best (6294, 6295). 6306. I then put in this Stop cock (of Grove's, so covered at the mouth) sideways. It made the boiler well Neg. 6307 1. Took off the cock and left the issue at the side by aperture of cross piece, and now the boiler was Positive, So the effect of the exit very evident. 6308. Perceived by the shaking of the ball and issue of water that it was half full of water, i.e. up to the level of the issue aper- ture. Drew it off by a syphon and then lo ! little or no electricity, either by the good cross piece or by Grove's stop cock, both so effectual before. 6309 ^. Put common water into the globe till it was as full as it could be in its usual position, our New stop cock being at the t [6307] 78 30 JUNE, 1842. end, and now there was very good excitement, the boiler being Negative. Put on the rough cross piece and still good excitement, the Boiler being now Positive. 6310. Now put Grove's Stop cock into side and there was no On effect, the steam issue being quiet in sound. steam getting higher, then effect produced and boiler Negative. 631 1. Took out the water from Globe and now all bad again. 6312. Worked from 9^ hours till \x\ hours, or 2| hours; had used 3^ gallons, bringing down the depth from 11 inches to 7 inches. Made up to 11 inches in depth again. 6313. Probably steam alone does nothing, but water and metal are the bodies which by friction produce electricity. SEPTR. 8, 1842. 6314. Change of water and ice into each other. 6315. In Lavoisier's Calorimeter the water thawed at one part- froze at another. 6316. In the Glaciers, according to Agassiz, water freezes within the crevices. 6317. When wet snow is squeezed together, it freezes into a lump (with water between), and does not fall asunder as so much wetted sand or other kind of matter would do. 6318. In a warm day, if two pieces of ice be laid one on the other and wrapped up in flannel, they will freeze into one piece. 6319. All this seems to indicate that water at 32° will not continue as water, if it be between two surfaces of ice touching or very near to each other. 6320. Also that in such cases, an accumulation of such pieces of ice in one heap or portion can have freezing going on within, whilst no part is below the freezing point and whilst thawing is going on at the outside. Now 6321. the freezing must evolve heat, and this heat must either be conducted to the outside through the mass of ice as a solid, or else be carried off by the water that is flowing off from the part where the ice is two } too] far separated for the effect [ to take place. 6322. The effect appears due to this, that a particle of water is cold enough to freeze when solid ice is on both sides of it, though it is not cold enough to freeze if ice be only on one side of it, equally cold water being on the other. 6323. In that case, the interior of a piece of ice is warmer than the exterior, where it is melting, for it will require a higher temperature to melt a particle of ice having ice on both sides of it, than a particle having ice on one side of it. 6324. In that case, ice is warmer than ice and water. 6325. The ice probably acts as a nucleus, but it appears that the effect of one surface of ice on water is not equal to the joint effect of two. 6326. The phenomena relate to the stability of ice and water, or of these state[s] of the Ox, Hy. combination^ and therefore are 8o SEPTR. 8, 1842. connected with the beautiful results of Macaire or Marcet on boiling water. I forget which it was made the expts. lately. 6327. If ice can serve as a nucleus, so may other bodies, and there may be some which assist the conversion of water into ice, and perhaps other favouring the conversion of ice into water, of course excluding combination forces. 6328. In the case of water and steam, water against water changes with more difficulty into steam than water against other bodies. So in this case, water against water changes more difficultly into ice than water against ice or perhaps other bodies. Water there- fore seems to be the more stable state— the normal state. Now as to experiments. 6329. Put one piece of ice upon another in a deep dish, covered over to keep in cool air. After half an hour, one was frozen to the other. Could even find two pieces sticking together after 5 minutes, or even 2 minutes, but only at a few points. A 6330. piece of ice wrapped in flannel and left. In half an hour the flannel is frozen to it, as if it were glued on by the surface filaments. 6331. If four or five thicknesses of dry flannel be pressed by the hand on to the melting surface of the piece of ice, in less than half a minute it is found sticking to it, as if frozen. 6332. If a single thickness be pressed on by the hand for 5 or 10 seconds, notwithstanding the warmth of the hand, signs of sticking will be perceived. A 6333. piece of flannel put on a piece of ice and wrapped up in flannel, soon well frozen to it, remaining dry as it was at first. Again, put it under the ice, i.e. at the bottom. After 3I hours found it all swimming in water, but flannel much frozen to the ice. 6334. Linen cloth: a duster. Ice wrapped in it -some sticking after half an hour, but cloth and ice very wet. 6335. Linen cloth folded up and pressed on to the ice for a few seconds, stuck as if frozen to it. Wiped the ice dry and pressed a dry place of the linen on to it for a minute. It was much frozen, i.e. it stuck very much to the ice. 6336. In this way linen stuck well to the ice. 6337. Ice wrapped up in a linen cloth did not stick at the top surface, but did at the bottom surface, where the weight of the ice preserved a contact— but not much adhesion— was half an hour. SEPTR. 8, 1842. 81 6338. Piece of linen canvas put on top of ice and wrapped up in flannel, after half an hour stuck a little. Again, put on under the ice, after 3I hours was very wet and flowing, but frozen to ice a little. 6339. Tin foil, copper foil and paper in pieces were laid on to ice wrapped up in flannel and left. In half an hour did not stick, but they are very wet and had not been kept close— water between them and the ice. The wrapping flannel stuck every where. A 6340. better way for trial experiment is to put the things to be tried between two flat pieces of ice and wrap them tight up together in flannel; or put them in a deep basin for a while and so that the water shall drain from the experimental part. 6341. Dry flannel between two ice: frozen strongly to ice in 40'. 6342. Do. „ „ : in two hours took some force to separate the pieces of ice, probably 4 or 5 lbs. weight. 6343. Flannel wetted by boiHng it in distilled water for an hour and put wet between two ice: well frozen to ices in 40'. 6344. Canvass linen between 2 ices: froze and stuck in 40'. 6345. Wet Canvas Do. : in 40' froze the two pieces together. 6^-'4^6. Black silk ribbon dry 11-m „ „ soaked 1 waterjr put ^ , between • ice, ' h1 ad1 frozen a little to the ice in 40'. The pieces, arranged as before, but better, and left for i| hours, both well frozen to ice. Could lift the lower piece of ice by the upper. 6347. White silk ribbon dry— and soaked— h^V^^tv^ ice— had frozen a little in half an hour. 6348. The day was warm and I beUeve the effect is due to an actual congelation of the water through the influence of the solid bodies present. I do not think the ice could have been of a temperature within lower than that at its surface, so as to have caused freezing by conduction of cold. 10 SEPTR. 1842. A 6349. piece of ice sawn into two pieces ; the flat surfaces put together; in less than 10'' the two are frozen into one piece by the touching points; that it is not water between which holds them is shewn by this, that not only do they resist direct 82 lo SEPTR. 1842. separation, but also any lateral or twist motion. The feeling soon shews this to the judgment. 6350. The two pieces put into water at the laboratory tempera- ture and then brought into contact. In 10 seconds and less were frozen together. 6351. When the pieces are pressed together, the adhesion by freezing is almost instantaneous. 6352. Tried the following woollen substances by putting them between 2 pieces of ice ( wrappers for 10 minutes: ) and binding them up in flannel Dry white flannel Thoroughly wet flannel White Kerseymere Green drugget Brown drugget Gray drugget Brown Marino Black Bombazine Worsted cruels in threads : white, red, green, blue, black, gray. All these adhered by freezing in the course of 10 minutes. A 6353. piece of ice was put under common laboratory water; a piece of white flannel thoroughly saturated with water was put under the water on to the ice, and then a second piece of ice was put on to the flannel— all being under the water and pressed moderately together. In less than a minute the flannel had frozen to both ices, though under the water the whole time. 6354. Another good and quick test method for flexible things, as cloth, is to place a piece of ice with a flat side upward; fold a linen cloth into a pad with a smooth under surface; dry the ice with the pad, at the same time cooling the surface of the pad; then lay on the piece of flannel or other body to be tried and hold it down with the linen pad, or with a pad of flannel, for half a minute or a few seconds. The body pressed against the ice will be found frozen to it if of woollen. 6355. Thus flannel (dry) pressed against dried ice freezes very quickly to it. 6356. Wet and saturated flannel Do. frozen to the ice in 20 or 30 seconds.